Self-Consistency Improves Chain of Thought Reasoning in Language Models

2022/03/21 by Xuezhi Wang, Wang, Xuezhi, Wei, Jason +9 · 965 citations
Computer Science · #Topic Modeling #Natural Language Processing Techniques #Advanced Graph Neural Networks

paper · pdf · doi:10.48550/arxiv.2203.11171

Abstract

Chain-of-thought prompting combined with pre-trained large language models has achieved encouraging results on complex reasoning tasks. In this paper, we propose a new decoding strategy, self-consistency, to replace the naive greedy decoding used in chain-of-thought prompting. It first samples a diverse set of reasoning paths instead of only taking the greedy one, and then selects the most consistent answer by marginalizing out the sampled reasoning paths. Self-consistency leverages the intuition that a complex reasoning problem typically admits multiple different ways of thinking leading to its unique correct answer. Our extensive empirical evaluation shows that self-consistency boosts the performance of chain-of-thought prompting with a striking margin on a range of popular arithmetic and commonsense reasoning benchmarks, including GSM8K (+17.9%), SVAMP (+11.0%), AQuA (+12.2%), StrategyQA (+6.4%) and ARC-challenge (+3.9%).

Cited by

• Single LLM Debate, MoLaCE: Mixture of Latent Concept Experts Against Confirmation Bias
• Beyond Correctness: Exposing LLM-generated Logical Flaws in Reasoning via Multi-step Automated Theorem Proving
• CME-CAD: Heterogeneous Collaborative Multi-Expert Reinforcement Learning for CAD Code Generation
• Anka: A Domain-Specific Language for Reliable LLM Code Generation
• Scoring, Reasoning, and Selecting the Best! Ensembling Large Language Models via a Peer-Review Process
• SPIRAL: Symbolic LLM Planning via Grounded and Reflective Search
• Scaling Unverifiable Rewards: A Case Study on Visual Insights
• Evaluating GRPO and DPO for Faithful Chain-of-Thought Reasoning in LLMs
• Agent2World: Learning to Generate Symbolic World Models via Adaptive Multi-Agent Feedback
• Agentic Software Issue Resolution with Large Language Models: A Survey
• Calibrating LLM Judges: Linear Probes for Fast and Reliable Uncertainty Estimation
• State-dependent error correlations shape voting thresholds in committees of AI agents
• Training Language Models to Cooperate with Inference-Time Controllers
• A Frozen 12B Beats Frontier Models on Verified Work: 100% Accuracy, 0 Tokens, Bit-Exact, Forever
• Confidently Wrong: Exception Chain Collapse in Frontier LLM Rule Evaluation
• Verbalized Particle Posterior: Bayesian Inference over Natural Language Hypotheses
• Accuracy Hides How Language Models Fail: Measuring Failure States Under Matched Output Budgets
• Hierarchical Group-Conditional Conformal Risk Control for Selective Prediction in Language Models
• Statistically Supported LLM Ingredient and Recipe Data Collection in Computational Nutrition
• ConsistencyGate: Preventing Memory Contamination in LLM Agents via Self-Consistency Admission Control
• Visual prompt engineering for video models
• Beyond Prefill-Decode Disaggregation: Dissecting LLM Inference for Heterogeneous Platforms via Dynamic Operator Scheduling
• Conformal Cascade: Distribution-Free Accuracy Guarantees for Multi-Tier LLM Inference
• When Does Few-Shot Prompting Help? A Systematic Empirical Study of Shot-Count Effects Across Model Scale, Architecture, and Output Parsing Robustness
• HALLELUAI: A Hallucination-Aware AI System for Ultra-Realistic Image-to-Video Generation at Scale
• MEMENTO: Memory-Guided Memetic Code-as-Policy Evolution
• How Well Can AI Generate Backlogs from App Mockups?
• Agent Team Work Zone: An Automated, Persistent Workspace for Long-Lived Claude Code Agent Teams
• Do Language Models Converge to Themselves? Recursive Self-Refinement as Textual Relaxation
• Measuring and Improving Behavioral Consistency in Large Language Models through Fact-Heuristic-Emotion State Enforcement
• Evolving from Lessons: Skill-Augmented Table Graph Reasoning for Operation-wise Table Question Answering
• DeepLook: Deeper Thinking with Lookahead
• DeepLens Diagnosis Agent: Agentic Workflow Design Lets a Small Reasoning Model Compete with Frontier LLMs
• Evaluating large language models for diagnostic reasoning from unstructured clinical narratives in epilepsy
• Procedural Knowledge at Scale Improves Reasoning
• StAR: Segment Anything Reasoner
• Method Decoration (DeMe): A Framework for LLM-Driven Adaptive Method Generation in Dynamic IoT Environments
• What Makes a GitHub Issue Ready for Copilot?
• AegisAgent: An Autonomous Defense Agent Against Prompt Injection Attacks in LLM-HARs
• Pioneering Multimodal Emotion Recognition in the Era of Large Models: From Closed Sets to Open Vocabularies
• PanoGrounder: Bridging 2D and 3D with Panoramic Scene Representations for VLM-based 3D Visual Grounding
• Multi-LLM Thematic Analysis with Dual Reliability Metrics: Combining Cohen's Kappa and Semantic Similarity for Qualitative Research Validation
• Reason2Decide: Rationale-Driven Multi-Task Learning
• Reaching Agreement Among Reasoning LLM Agents
• Demystifying LLM-as-a-Judge: Analytically Tractable Model for Inference-Time Scaling
• Understanding Chain-of-Thought in Large Language Models via Topological Data Analysis
• A Large Language Model Based Method for Complex Logical Reasoning over Knowledge Graphs
• Population-Evolve: a Parallel Sampling and Evolutionary Method for LLM Math Reasoning
• HARMON-E: Hierarchical Agentic Reasoning for Multimodal Oncology Notes to Extract Structured Data
• Thinking Beyond Labels: Vocabulary-Free Fine-Grained Recognition using Reasoning-Augmented LMMs
• MDToC: Metacognitive Dynamic Tree of Concepts for Boosting Mathematical Problem-Solving of Large Language Models
• Multi-Agent LLM Committees for Autonomous Software Beta Testing
• Reflective Confidence: Correcting Reasoning Flaws via Online Self-Correction
• Training LLMs with LogicReward for Faithful and Rigorous Reasoning
• External Hippocampus: Topological Cognitive Maps for Guiding Large Language Model Reasoning
• Eidoku: A Neuro-Symbolic Verification Gate for LLM Reasoning via Structural Constraint Satisfaction
• Reasoning Palette: Modulating Reasoning via Latent Contextualization for Controllable Exploration for (V)LMs
• Trust-Region Adaptive Policy Optimization
• Constructive Circuit Amplification: Improving Math Reasoning in LLMs via Targeted Sub-Network Updates
• Multimodal RewardBench 2: Evaluating Omni Reward Models for Interleaved Text and Image
• FlowDet: Unifying Object Detection and Generative Transport Flows
• Emergent Bias and Fairness in Multi-Agent Decision Systems
• A Network Arena for Benchmarking AI Agents on Network Troubleshooting
• Agent Tools Orchestration Leaks More: Dataset, Benchmark, and Mitigation
• QuadSentinel: Sequent Safety for Machine-Checkable Control in Multi-agent Systems
• ToolForge: A Data Synthesis Pipeline for Multi-Hop Search without Real-World APIs
• BRAID: Bounded Reasoning for Autonomous Inference and Decisions
• City Navigation in the Wild: Exploring Emergent Navigation from Web-Scale Knowledge in MLLMs
• Characterizing Mamba's Selective Memory using Auto-Encoders
• Step-GUI Technical Report
• Dual-Density Inference for Efficient Language Model Reasoning
• Evaluating LLMs for Zeolite Synthesis Event Extraction (ZSEE): A Systematic Analysis of Prompting Strategies
• RFKG-CoT: Relation-Driven Adaptive Hop-count Selection and Few-Shot Path Guidance for Knowledge-Aware QA
• Do Large Language Models Know What They Don't Know? Kalshibench: A New Benchmark for Evaluating Epistemic Calibration via Prediction Markets
• Beyond Fast and Slow: Cognitive-Inspired Elastic Reasoning for Large Language Models
• An Exploratory Study of Bayesian Prompt Optimization for Test-Driven Code Generation with Large Language Models
• Multi-Agent Medical Decision Consensus Matrix System: An Intelligent Collaborative Framework for Oncology MDT Consultations
• Intention Chain-of-Thought Prompting with Dynamic Routing for Code Generation
• OmniDrive-R1: Reinforcement-driven Interleaved Multi-modal Chain-of-Thought for Trustworthy Vision-Language Autonomous Driving
• Reasoning Relay: Evaluating Stability and Interchangeability of Large Language Models in Mathematical Reasoning
• Let's (not) just put things in Context: Test-Time Training for Long-Context LLMs
• Error-Driven Prompt Optimization for Arithmetic Reasoning
• ORIBA: Exploring LLM-Driven Role-Play Chatbot as a Creativity Support Tool for Original Character Artists
• Socratic Students: Teaching Language Models to Learn by Asking Questions
• How Prompts Move Language Model Behavior: Frames, Salience, and Construal as Semantic Control
• Hold Onto That Thought: Assessing KV Cache Compression On Reasoning
• When Actions Teach You to Think: Reasoning-Action Synergy via Reinforcement Learning in Conversational Agents
• FutureWeaver: Planning Test-Time Compute for Multi-Agent Systems with Modularized Collaboration
• Limits and Gains of Test-Time Scaling in Vision-Language Reasoning
• PIAST: Rapid Prompting with In-context Augmentation for Scarce Training data
• Visual Species Recognition with Large Multimodal Models as Post-Hoc Correctors
• BAMBO: Construct Ability and Efficiency LLM Pareto Set via Bayesian Adaptive Multi-objective Block-wise Optimization
• Encoder-Free Knowledge-Graph Reasoning with LLMs via Hyperdimensional Path Retrieval
• Decoupling Template Bias in CLIP: Harnessing Empty Prompts for Enhanced Few-Shot Learning
• Enhancing Clinical Note Generation with ICD-10, Clinical Ontology Knowledge Graphs, and Chain-of-Thought Prompting Using GPT-4
• Chain-of-Image Generation: Toward Monitorable and Controllable Image Generation
• ReasonBENCH: Benchmarking the (In)Stability of LLM Reasoning
• Metric-Fair Prompting: Treating Similar Samples Similarly
• Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement Learning
• Attention as Binding: A Vector-Symbolic Perspective on Transformer Reasoning
• CoT4Det: A Chain-of-Thought Framework for Perception-Oriented Vision-Language Tasks
• Nanbeige4-3B Technical Report: Exploring the Frontier of Small Language Models
• TRACE: A Framework for Analyzing and Enhancing Stepwise Reasoning in Vision-Language Models
• Prompting Science Report 4: Playing Pretend: Expert Personas Don't Improve Factual Accuracy
• The Missing Layer of AGI: From Pattern Alchemy to Coordination Physics
• MARINE: Theoretical Optimization and Design for Multi-Agent Recursive IN-context Enhancement
• CureAgent: A Training-Free Executor-Analyst Framework for Clinical Reasoning
• RoBoN: Routed Online Best-of-n for Test-Time Scaling with Multiple LLMs
• The Seeds of Scheming: Weakness of Will in the Building Blocks of Agentic Systems
• RevoNAD: Reflective Evolutionary Exploration for Neural Architecture Design
• Algorithmic Thinking Theory
• Dual-Use AI Face Swap Apps Are Mostly Unsafe: A Systematic Safety Audit
• Beyond Factual Accuracy: Evaluating Global Reasoning Integrity in RAG Systems with LogicScore
• On the Limits of Test-Time Compute: Sequential Reward Filtering for Better Inference
• Distance Is All You Need: Radial Dispersion for Uncertainty Estimation in Large Language Models
• From Hypothesis to Premises: LLM-based Backward Logical Reasoning with Selective Symbolic Translation
• CrowdLLM: Building LLM-Based Digital Populations Augmented with Generative Models
• SPARK: Stepwise Process-Aware Rewards for Reference-Free Reinforcement Learning
• Network Self-Configuration based on Fine-Tuned Small Language Models
• From Imitation to Discrimination: Toward A Generalized Curriculum Advantage Mechanism Enhancing Cross-Domain Reasoning Tasks
• In-Context Distillation with Self-Consistency Cascades: A Simple, Training-Free Way to Reduce LLM Agent Costs
• WISE: Weighted Iterative Society-of-Experts for Robust Multimodal Multi-Agent Debate
• Distribution-Calibrated Inference Time Compute for Thinking LLM-as-a-Judge
• When Does Verification Pay Off? A Closer Look at LLMs as Solution Verifiers
• Think in Parallel, Answer as One: Logit Averaging for Open-Ended Reasoning
• ReVSeg: Incentivizing the Reasoning Chain for Video Segmentation with Reinforcement Learning
• DETAIL Matters: Measuring the Impact of Prompt Specificity on Reasoning in Large Language Models
• ALARM: Automated MLLM-Based Anomaly Detection in Complex-EnviRonment Monitoring with Uncertainty Quantification
• The Art of Scaling Test-Time Compute for Large Language Models
• Latent Debate: A Surrogate Framework for Interpreting LLM Thinking
• Zero-Overhead Introspection for Adaptive Test-Time Compute
• ViRectify: A Challenging Benchmark for Video Reasoning Correction with Multimodal Large Language Models
• RoboDriveVLM: A Novel Benchmark and Baseline towards Robust Vision-Language Models for Autonomous Driving
• Mode-Conditioning Unlocks Superior Test-Time Scaling
• IndiMathBench: Autoformalizing Mathematical Reasoning Problems with a Human Touch
• ART: Adaptive Response Tuning Framework -- A Multi-Agent Tournament-Based Approach to LLM Response Optimization
• Large Language Model based Smart Contract Auditing with LLMBugScanner
• Augmented Runtime Collaboration for Self-Organizing Multi-Agent Systems: A Hybrid Bi-Criteria Routing Approach
• SCALE: Selective Resource Allocation for Overcoming Performance Bottlenecks in Mathematical Test-time Scaling
• ML-Tool-Bench: Tool-Augmented Planning for ML Tasks
• Aligning Probabilistic Beliefs under Informative Missingness: LLM Steerability in Clinical Reasoning
• Evaluating LLMs for One-Shot Patching of Real and Artificial Vulnerabilities
• Toward a Safe Internet of Agents
• Video-R2: Reinforcing Consistent and Grounded Reasoning in Multimodal Language Models
• Multi-chain Graph Refinement and Selection for Reliable Reasoning in Large Language Models
• Experts are all you need: A Composable Framework for Large Language Model Inference
• ORION: Teaching Language Models to Reason Efficiently in the Language of Thought
• Adversarial Training for Process Reward Models
• AgroCoT: A Chain-of-Thought Benchmark for Evaluating Reasoning in Vision-Language Models for Agriculture
• Analyzing Image Beyond Visual Aspect: Image Emotion Classification via Multiple-Affective Captioning
• VeriDispatcher: Multi-Model Dispatching through Pre-Inference Difficulty Prediction for RTL Generation Optimization
• Rethinking Test Time Scaling for Flow-Matching Generative Models
• On the Limits of Innate Planning in Large Language Models
• EWE: An Agentic Framework for Extreme Weather Analysis
• Scaling Generative Verifiers For Natural Language Mathematical Proof Verification And Selection
• GEM: Generative Entropy-Guided Preference Modeling for Few-shot Alignment of LLMs
• BRIDGE: Building Representations In Domain Guided Program Verification
• Aragog: Just-in-Time Model Routing for Scalable Serving of Agentic Workflows
• Complex QA and language models hybrid architectures, Survey
• DRAFT-RL: Multi-Agent Chain-of-Draft Reasoning for Reinforcement Learning-Enhanced LLMs
• Agent0-VL: Exploring Self-Evolving Agent for Tool-Integrated Vision-Language Reasoning
• VeriSciQA: An Auto-Verified Dataset for Scientific Visual Question Answering
• ThreadWeaver: Adaptive Threading for Efficient Parallel Reasoning in Language Models
• Cross Domain Evaluation of Multimodal Chain-of-Thought Reasoning of different datasets into the Amazon CoT Framework
• LAST: LeArning to Think in Space and Time for Generalist Vision-Language Models
• GraphMind: Theorem Selection and Conclusion Generation Framework with Dynamic GNN for LLM Reasoning
• Majority of the Bests: Improving Best-of-N via Bootstrapping
• Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens
• DELTA: Language Diffusion-based EEG-to-Text Architecture
• A²Flow: Automating Agentic Workflow Generation via Self-Adaptive Abstraction Operators
• Reuse, Don't Recompute: Efficient Large Reasoning Model Inference via Memory Orchestration
• LLMs as Firmware Experts: A Runtime-Grown Tree-of-Agents Framework
• MedRule-KG: A Knowledge-Graph--Steered Scaffold for Reliable Mathematical and Biomedical Reasoning
• Enhancing Automated Program Repair via Faulty Token Localization and Quality-Aware Patch Refinement
• SPINE: Token-Selective Test-Time Reinforcement Learning with Entropy-Band Regularization
• Measuring the Impact of Lexical Training Data Coverage on Hallucination Detection in Large Language Models
• AutoLink: Autonomous Schema Exploration and Expansion for Scalable Schema Linking in Text-to-SQL at Scale
• FireScope: Wildfire Risk Raster Prediction with a Chain-of-Thought Oracle
• Hallucinate Less by Thinking More: Aspect-Based Causal Abstention for Large Language Models
• HUMORCHAIN: Theory-Guided Multi-Stage Reasoning for Interpretable Multimodal Humor Generation
• Do Vision-Language Models Understand Visual Persuasiveness?
• Budget-Aware Tool-Use Enables Effective Agent Scaling
• V-ReasonBench: Toward Unified Reasoning Benchmark Suite for Video Generation Models
• Goal-Directed Search Outperforms Goal-Agnostic Memory Compression in Long-Context Memory Tasks
• Cognitive Foundations for Reasoning and Their Manifestation in LLMs
• SDA: Steering-Driven Distribution Alignment for Open LLMs without Fine-Tuning
• Pass@k Metric for RLVR: A Diagnostic Tool of Exploration, But Not an Objective
• JudgeBoard: Benchmarking and Enhancing Small Language Models for Reasoning Evaluation
• Step-Audio-R1 Technical Report
• When to Think and When to Look: Uncertainty-Guided Lookback
• ConInstruct: Evaluating Large Language Models on Conflict Detection and Resolution in Instructions
• Efficiency Will Not Lead to Sustainable Reasoning AI
• SkinGPT-R1: Adapter-Only Dual Distillation for Efficient Dermatology Reasoning
• OEMA: Ontology-Enhanced Multi-Agent Collaboration Framework for Zero-Shot Clinical Named Entity Recognition
• SafeRBench: A Comprehensive Benchmark for Safety Assessment in Large Reasoning Models
• ProRAC: A Neuro-symbolic Method for Reasoning about Actions with LLM-based Progression
• Reasoning via Video: The First Evaluation of Video Models' Reasoning Abilities through Maze-Solving Tasks
• RB-FT: Rationale-Bootstrapped Fine-Tuning for Video Classification
• SVBRD-LLM: Self-Verifying Behavioral Rule Discovery for Autonomous Vehicle Identification
• GPS: General Per-Sample Prompter
• Extending Test-Time Scaling: A 3D Perspective with Context, Batch, and Turn
• Collaborative QA using Interacting LLMs. Impact of Network Structure, Node Capability and Distributed Data
• Dynamic Template Selection for Output Token Generation Optimization: MLP-Based and Transformer Approaches
• Quantifying and Mitigating Selection Bias in LLMs: A Transferable LoRA Fine-Tuning and Efficient Majority Voting Approach
• MedDCR: Learning to Design Agentic Workflows for Medical Coding
• Think with Self-Decoupling and Self-Verification: Automated RTL Design with Backtrack-ToT
• From Perception to Reasoning: Deep Thinking Empowers Multimodal Large Language Models
• Scaling Graph Chain-of-Thought Reasoning: A Multi-Agent Framework with Efficient LLM Serving
• MedSumGraph: enhancing GraphRAG for medical QA with summarization and optimized prompts
• R²Seg: Training-Free OOD Medical Tumor Segmentation via Anatomical Reasoning and Statistical Rejection
• Knots: A Large-Scale Multi-Agent Enhanced Expert-Annotated Dataset and LLM Prompt Optimization for NOTAM Semantic Parsing
• ARCHE: A Novel Task to Evaluate LLMs on Latent Reasoning Chain Extraction
• Optimal Self-Consistency for Efficient Reasoning with Large Language Models
• Learning to Refine: An Agentic RL Approach for Iterative SPARQL Query Construction
• An Analysis of Architectural Impact on LLM-based Abstract Visual Reasoning: A Systematic Benchmark on RAVEN-FAIR
• Scaling Open-Weight Large Language Models for Hydropower Regulatory Information Extraction: A Systematic Analysis
• On the Notion that Language Models Reason
• iMAD: Intelligent Multi-Agent Debate for Efficient and Accurate LLM Inference
• Hindsight Distillation Reasoning with Knowledge Encouragement Preference for Knowledge-based Visual Question Answering
• Binary Verification for Zero-Shot Vision
• Socrates-Mol: Self-Oriented Cognitive Reasoning through Autonomous Trial-and-Error with Empirical-Bayesian Screening for Molecules
• ProgRAG: Hallucination-Resistant Progressive Retrieval and Reasoning over Knowledge Graphs
• Efficient Thought Space Exploration Through Strategic Intervention
• Enhancing the Outcome Reward-based RL Training of MLLMs with Self-Consistency Sampling
• Modeling Uncertainty Trends for Timely Retrieval in Dynamic RAG
• Seer Self-Consistency: Advance Budget Estimation for Adaptive Test-Time Scaling
• Towards Trustworthy Dermatology MLLMs: A Benchmark and Multimodal Evaluator for Diagnostic Narratives
• Automatic Minds: Cognitive Parallels Between Hypnotic States and Large Language Model Processing
• AlphaCast: A Human Wisdom-LLM Intelligence Co-Reasoning Framework for Interactive Time Series Forecasting
• Hierarchical Memorization in Large Language Models: Evidence from Citation Generation
• Consensus Sampling for Safer Generative AI
• DynaAct: Large Language Model Reasoning with Dynamic Action Spaces
• Knowledge-Augmented Long-CoT Generation for Complex Biomolecular Reasoning
• Benchmarking Multi-Step Legal Reasoning and Analyzing Chain-of-Thought Effects in Large Language Models
• DeCoRL: Decoupling Reasoning Chains via Parallel Sub-Step Generation and Cascaded Reinforcement for Interpretable and Scalable RLHF
• Last Layer Logits to Logic: Empowering LLMs with Logic-Consistent Structured Knowledge Reasoning
• Judging by the Rules: Compliance-Aligned Framework for Modern Slavery Statement Monitoring
• PCRLLM: Proof-Carrying Reasoning with Large Language Models under Stepwise Logical Constraints
• C3PO: Optimized Large Language Model Cascades with Probabilistic Cost Constraints for Reasoning
• Voice-Interactive Surgical Agent for Multimodal Patient Data Control
• Think Consistently, Reason Efficiently: Energy-Based Calibration for Implicit Chain-of-Thought
• CoLM: Collaborative Large Models via A Client-Server Paradigm
• A Meta-Evaluation of Faithfulness Metrics for Long-Form Hospital-Course Summarization
• TabRAG: Tabular Document Retrieval via Structured Language Representations
• SAFENLIDB: A Privacy-Preserving Safety Alignment Framework for LLM-based Natural Language Database Interfaces
• A Low-Rank Method for Vision Language Model Hallucination Mitigation in Autonomous Driving
• Data Trajectory Alignment for LLM Domain Adaptation: A Two-Phase Synthesis Framework for Telecommunications Mathematics
• Self-Evaluating LLMs for Multi-Step Tasks: Stepwise Confidence Estimation for Failure Detection
• ALIGN: A Vision-Language Framework for High-Accuracy Accident Location Inference through Geo-Spatial Neural Reasoning
• Confidence-Guided Stepwise Model Routing for Cost-Efficient Reasoning
• TimeSense:Making Large Language Models Proficient in Time-Series Analysis
• FLEX: Continuous Agent Evolution via Forward Learning from Experience
• Kunlun Anomaly Troubleshooter: Enabling Kernel-Level Anomaly Detection and Causal Reasoning for Large Model Distributed Inference
• In-depth Analysis on Caching and Pre-fetching in Mixture of Experts Offloading
• CoT-X: An Adaptive Framework for Cross-Model Chain-of-Thought Transfer and Optimization
• Culture in Action: Evaluating Text-to-Image Models through Social Activities
• Logit-Entropy Adaptive Stopping Heuristic for Efficient Chain-of-Thought Reasoning
• Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm
• Reusing Pre-Training Data at Test Time is a Compute Multiplier
• Batch Prompting Suppresses Overthinking Reasoning Under Constraint: How Batch Prompting Suppresses Overthinking in Reasoning Models
• PSD2Code: Automated Front-End Code Generation from Design Files via Multimodal Large Language Models
• Revealing AI Reasoning Increases Trust but Crowds Out Unique Human Knowledge
• Efficient Reasoning via Thought-Training and Thought-Free Inference
• GEMMA-SQL: A Novel Text-to-SQL Model Based on Large Language Models
• LFC-DA: Logical Formula-Controlled Data Augmentation for Enhanced Logical Reasoning
• From Five Dimensions to Many: Large Language Models as Precise and Interpretable Psychological Profilers
• Analyzing the Power of Chain of Thought through Memorization Capabilities
• QiMeng-NeuComBack: Self-Evolving Translation from IR to Assembly Code
• CGES: Confidence-Guided Early Stopping for Efficient and Accurate Self-Consistency
• LTD-Bench: Evaluating Large Language Models by Letting Them Draw
• The Sequential Edge: Inverse-Entropy Voting Beats Parallel Self-Consistency at Matched Compute
• Unlocking the Power of Multi-Agent LLM for Reasoning: From Lazy Agents to Deliberation
• LA-MARRVEL: A Knowledge-Grounded, Language-Aware LLM Framework for Clinically Robust Rare Disease Gene Prioritization
• Personalized Decision Modeling: Utility Optimization or Textualized-Symbolic Reasoning
• When Visualizing is the First Step to Reasoning: MIRA, a Benchmark for Visual Chain-of-Thought
• FEval-TTC: Fair Evaluation Protocol for Test-Time Compute
• Self-Harmony: Learning to Harmonize Self-Supervision and Self-Play in Test-Time Reinforcement Learning
• GeoToken: Hierarchical Geolocalization of Images via Next Token Prediction
• Efficient Test-Time Retrieval Augmented Generation
• MARS-SQL: A multi-agent reinforcement learning framework for Text-to-SQL
• ORANGE: An Online Reflection ANd GEneration framework with Domain Knowledge for Text-to-SQL
• Assessing LLM Reasoning Steps via Principal Knowledge Grounding
• Do Math Reasoning LLMs Help Predict the Impact of Public Transit Events?
• How Focused Are LLMs? A Quantitative Study via Repetitive Deterministic Prediction Tasks
• Self-Consistency Is Losing Its Edge: Diminishing Returns and Rising Costs in Modern LLMs
• Inference-Time Chain-of-Thought Pruning with Latent Informativeness Signals
• Test-time Scaling of LLMs: A Survey from A Subproblem Structure Perspective
• DTS: Enhancing Large Reasoning Models via Decoding Tree Sketching
• SpecDiff-2: Scaling Diffusion Drafter Alignment For Faster Speculative Decoding
• Reasoning Planning for Language Models
• TreeQA: Enhanced LLM-RAG with logic tree reasoning for reliable and interpretable multi-hop question answering
• Sherlock: Reliable and Efficient Agentic Workflow Execution
• TempoBench: Evaluating Temporal Causal Reasoning in Large Language Models
• ThoughtProbe: Classifier-Guided LLM Thought Space Exploration via Probing Representations
• A Survey on Generative Recommendation: Data, Model, and Tasks
• Rating Roulette: Self-Inconsistency in LLM-As-A-Judge Frameworks
• LLMs are Overconfident: Evaluating Confidence Interval Calibration with FermiEval
• Cross-Platform Evaluation of Reasoning Capabilities in Foundation Models
• Chopping Trees: Semantic Similarity Based Dynamic Pruning for Tree-of-Thought Reasoning
• Stop Wasting Your Tokens: Towards Efficient Runtime Multi-Agent Systems
• The Geometry of Dialogue: Graphing Language Models to Reveal Synergistic Teams for Multi-Agent Collaboration
• Do LLMs Signal When They're Right? Evidence from Neuron Agreement
• Pragmatic Theories Enhance Understanding of Implied Meanings in LLMs
• Reasoning Path Divergence: A New Metric and Curation Strategy to Unlock LLM Diverse Thinking
• Limits of Generalization in RLVR: Two Case Studies in Mathematical Reasoning
• CAVE: Detecting and Explaining Commonsense Anomalies in Visual Environments
• Generalizing Test-time Compute-optimal Scaling as an Optimizable Graph
• Symbolically Scaffolded Play: Designing Role-Sensitive Prompts for Generative NPC Dialogue
• TextualVerifier: Verify TextGrad Step-by-Step
• LLM-as-a-Verifier: A General-Purpose Verification Framework
• Cache Merging as a Convergent Replicated State for Multi-Agent Latent Reasoning
• Considering the ethics of large machine learning models in the chemical sciences
• Rethinking the Evaluation of Harness Evolution for Agents
• Large Language Models for Software Engineering Diagrams: A Systematic Review of UML and ER modelling
• A Reference-Free Score for Detecting Silent Reasoning Failures in Large Language Models
• Lang-PINN: From Language to Physics-Informed Neural Networks via a Multi-Agent Framework
• LLM-TSFD: An industrial time series human-in-the-loop fault diagnosis method based on a large language model
• Truth-Aware Decoding: A Program-Logic Approach to Factual Language Generation
• The Architecture of Errors: From Universal Impossibility to Patch-Local LLM Reliability
• STAR-PólyaMath: Multi-Agent Reasoning under Persistent Meta-Strategic Supervision
• Metacognition Should Be the Scientific Framework for Bounded and Effective Self-Governance in Generative AI
• Lattice Deduction Transformers
• The Consensus Trap: Rescuing Multi-Agent LLMs from Adversarial Majorities via Token-Level Collaboration
• GroupRAG: Cognitively Inspired Group-Aware Retrieval and Reasoning via Knowledge-Driven Problem Structuring
• When Agents Disagree With Themselves: Behavioral Consistency as an Uncertainty Signal for LLM Agents
• Large language models encode clinical knowledge
• How well LLM-based test generation techniques perform with newer LLM versions?
• TrajSelector: Harnessing Latent Representations for Efficient and Effective Best-of-N in Large Reasoning Model
• Parrot: A Training Pipeline Enhances Both Program CoT and Natural Language CoT for Reasoning
• Can LLMs Estimate Cognitive Complexity of Reading Comprehension Items?
• Aligning Large Language Models with Procedural Rules: An Autoregressive State-Tracking Prompting for In-Game Trading
• ParallelMuse: Agentic Parallel Thinking for Deep Information Seeking
• LoRA-DA: Data-Aware Initialization for Low-Rank Adaptation via Asymptotic Analysis
• Quantum Combinatorial Reasoning for Large Language Models
• Critique-RL: Training Language Models for Critiquing through Two-Stage Reinforcement Learning
• Verifying Large Language Models' Reasoning Paths via Correlation Matrix Rank
• Auto prompting without training labels: An LLM cascade for product quality assessment in e-commerce catalogs
• ProofSketch: Efficient Verified Reasoning for Large Language Models
• From Stochasticity to Signal: A Bayesian Latent State Model for Reliable Measurement with LLMs
• SSL4RL: Revisiting Self-supervised Learning as Intrinsic Reward for Visual-Language Reasoning
• Compiler.next: A Search-Based Compiler to Power the AI-Native Future of Software Engineering
• TDFlow: Agentic Workflows for Test Driven Software Engineering
• Think Twice: Branch-and-Rethink Reasoning Reward Model
• Unleashing Diverse Thinking Modes in LLMs through Multi-Agent Collaboration
• BrowseConf: Confidence-Guided Test-Time Scaling for Web Agents
• Incentivizing Agentic Reasoning in LLM Judges via Tool-Integrated Reinforcement Learning
• Can Language Models Compose Skills In-Context?
• Reasoning About Reasoning: Towards Informed and Reflective Use of LLM Reasoning in HCI
• Scalable Supervising Software Agents with Patch Reasoner
• Multi-Modal Fact-Verification Framework for Reducing Hallucinations in Large Language Models
• S-Chain: Structured Visual Chain-of-Thought For Medicine
• Scalable Oversight via Partitioned Human Supervision
• Agent-GSPO: Communication-Efficient Multi-Agent Systems via Group Sequence Policy Optimization
• CHOIR: Collaborative Harmonization fOr Inference Robustness
• Chitchat with AI: Understand the supply chain carbon disclosure of companies worldwide through Large Language Model
• You Don't Need Prompt Engineering Anymore: The Prompting Inversion
• Controllable Mathematical Reasoning via Self-Optimizing Thought Vectors
• Foundation of Intelligence: Review of Math Word Problems from Human Cognition Perspective
• RETuning: Upgrading Inference-Time Scaling for Stock Movement Prediction with Large Language Models
• Boosting Accuracy and Efficiency of Budget Forcing in LLMs via Reinforcement Learning for Mathematical Reasoning
• Co-Sight: Enhancing LLM-Based Agents via Conflict-Aware Meta-Verification and Trustworthy Reasoning with Structured Facts
• Unifying Large Language Models and Knowledge Graphs: A Roadmap
• String Seed of Thought: Prompting LLMs for Distribution-Faithful and Diverse Generation
• Large Language Models Meet Text-Attributed Graphs: A Survey of Integration Frameworks and Applications
• The Virtues of Brevity: Avoid Overthinking in Parallel Test-Time Reasoning
• Reasoning's Razor: Reasoning Improves Accuracy but Can Hurt Recall at Critical Operating Points in Safety and Hallucination Detection
• Towards Scalable Oversight with Collaborative Multi-Agent Debate in Error Detection
• Language Ranker: A Lightweight Ranking framework for LLM Decoding
• GeoThought: A Dataset for Enhancing Mathematical Geometry Reasoning in Vision-Language Models
• Neural Diversity Regularizes Hallucinations in Small Models
• Systematic Evaluation of Uncertainty Estimation Methods in Large Language Models
• The Shape of Reasoning: Topological Analysis of Reasoning Traces in Large Language Models
• Robust Preference Alignment via Directional Neighborhood Consensus
• Automatic Prompt Engineering for Automatic Scoring
• TriAgent: Automated Biomarker Discovery with Deep Research Grounding for Triage in Acute Care by LLM-Based Multi-Agent Collaboration
• Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table Understanding
• Rethinking Cross-lingual Gaps from a Statistical Viewpoint
• Enhancing visual-LLM for construction site safety compliance via prompt engineering and Bi-stage retrieval-augmented generation
• SCoPE VLM: Selective Context Processing for Efficient Document Navigation in Vision-Language Models
• Teaming LLMs to Detect and Mitigate Hallucinations
• Automated HIV Screening on Dutch Electronic Health Records with Large Language Models
• AgenticMath: Enhancing LLM Reasoning via Agentic-based Math Data Generation
• The Zero-Step Thinking: An Empirical Study of Mode Selection as Harder Early Exit in Reasoning Models
• SheetBrain: A Neuro-Symbolic Agent for Accurate Reasoning over Complex and Large Spreadsheets
• DiSRouter: Distributed Self-Routing for LLM Selections
• What Makes a Good Curriculum? Disentangling the Effects of Data Ordering on LLM Mathematical Reasoning
• Robust Driving QA through Metadata-Grounded Context and Task-Specific Prompts
• Prompt Decorators: A Declarative and Composable Syntax for Reasoning, Formatting, and Control in LLMs
• Large language models in medicine
• PlanU: Large Language Model Reasoning through Planning under Uncertainty
• 3D Optimization for AI Inference Scaling: Balancing Accuracy, Cost, and Latency
• DelvePO: Direction-Guided Self-Evolving Framework for Flexible Prompt Optimization
• Exemplar-Guided Planing: Enhanced LLM Agent for KGQA
• OPTAGENT: Optimizing Multi-Agent LLM Interactions Through Verbal Reinforcement Learning for Enhanced Reasoning
• Foundational Automatic Evaluators: Scaling Multi-Task Generative Evaluator Training for Reasoning-Centric Domains
• LLM-as-a-Prophet: Understanding Predictive Intelligence with Prophet Arena
• Certified Self-Consistency: Statistical Guarantees and Test-Time Training for Reliable Reasoning in LLMs
• Strengthening LLMs for Tabular Prediction with Structural Priors
• StreamingThinker: Large Language Models Can Think While Reading
• CLAWS:Creativity detection for LLM-generated solutions using Attention Window of Sections
• Investigating Thinking Behaviours of Reasoning-Based Language Models for Social Bias Mitigation
• QueST: Incentivizing LLMs to Generate Difficult Problems
• I-RAVEN-X: Benchmarking Generalization and Robustness of Analogical and Mathematical Reasoning in Large Language and Reasoning Models
• Reasoning Distillation and Structural Alignment for Improved Code Generation
• DAG-Math: Graph-of-Thought Guided Mathematical Reasoning in LLMs
• Investigating the Impact of Rationales for LLMs on Natural Language Understanding
• Planner and Executor: Collaboration between Discrete Diffusion And Autoregressive Models in Reasoning
• The Spark Effect: On Engineering Creative Diversity in Multi-Agent AI Systems
• Rewiring Experts on the Fly:Continuous Rerouting for Better Online Adaptation in Mixture-of-Expert models
• A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning
• CarBoN: Calibrated Best-of-N Sampling Improves Test-time Reasoning
• LSPO: Length-aware Dynamic Sampling for Policy Optimization in LLM Reasoning
• Fusion-Augmented Large Language Models: Boosting Diagnostic Trustworthiness via Model Consensus
• Composition-Grounded Instruction Synthesis for Visual Reasoning
• Harmonizing Diverse Models: A Layer-wise Merging Strategy for Consistent Generation
• Mapping Smarter, Not Harder: A Test-Time Reinforcement Learning Agent That Improves Without Labels or Model Updates
• You May Speak Freely: Improving the Fine-Grained Visual Recognition Capabilities of Multimodal Large Language Models with Answer Extraction
• Boosting Instruction Following at Scale
• Finding Answers in Thought Matters: Revisiting Evaluation on Large Language Models with Reasoning
• COIG-Writer: A High-Quality Dataset for Chinese Creative Writing with Thought Processes
• ToolPRM: Fine-Grained Inference Scaling of Structured Outputs for Function Calling
• LLM Agents Beyond Utility: An Open-Ended Perspective
• Suicidal Comment Tree Dataset: Enhancing Risk Assessment and Prediction Through Contextual Analysis
• CURE: Confidence-driven Unified Reasoning Ensemble Framework for Medical Question Answering
• Automated Extraction of Protocol State Machines from 3GPP Specifications with Domain-Informed Prompts and LLM Ensembles
• Metacognitive Self-Correction for Multi-Agent System via Prototype-Guided Next-Execution Reconstruction
• Where to Search: Measure the Prior-Structured Search Space of LLM Agents
• Budget-aware Test-time Scaling via Discriminative Verification
• Stop Reducing Responsibility in LLM-Powered Multi-Agent Systems to Local Alignment
• Classifying and Addressing the Diversity of Errors in Retrieval-Augmented Generation Systems
• Big Reasoning with Small Models: Instruction Retrieval at Inference Time
• In-Distribution Steering: Balancing Control and Coherence in Language Model Generation
• Stable LLM Ensemble: Interaction between Example Representativeness and Diversity
• Retrieval-in-the-Chain: Bootstrapping Large Language Models for Generative Retrieval
• Towards Human-Centric Intelligent Treatment Planning for Radiation Therapy
• Higher Satisfaction, Lower Cost: A Technical Report on How LLMs Revolutionize Meituan's Intelligent Interaction Systems
• ESI: Epistemic Uncertainty Quantification via Semantic-preserving Intervention for Large Language Models
• Optimal Aggregation of LLM and PRM Signals for Efficient Test-Time Scaling
• Adaptive Reasoning Executor: A Collaborative Agent System for Efficient Reasoning
• Max It or Miss It: Benchmarking LLM On Solving Extremal Problems
• Schema for In-Context Learning
• Benefits and Limitations of Communication in Multi-Agent Reasoning
• Multi-Agent Debate for LLM Judges with Adaptive Stability Detection
• PromptFlow: Training Prompts Like Neural Networks
• Evolution of meta's llama models and parameter-efficient fine-tuning of large language models: a survey
• Towards Inference-time Scaling for Continuous Space Reasoning
• A Survey on Parallel Reasoning
• HoneyBee: Data Recipes for Vision-Language Reasoners
• LLM Reasoning for Machine Translation: Synthetic Data Generation over Thinking Tokens
• DMAS-Forge: A Framework for Transparent Deployment of AI Applications as Distributed Systems
• OneRec-Think: In-Text Reasoning for Generative Recommendation
• ReLook: Vision-Grounded RL with a Multimodal LLM Critic for Agentic Web Coding
• LLM-Specific Utility: A New Perspective for Retrieval-Augmented Generation
• EAGER: Entropy-Aware GEneRation for Adaptive Inference-Time Scaling
• CoSPED: Consistent Soft Prompt Targeted Data Extraction and Defense
• Enhancing LLM Reasoning via Non-Human-Like Reasoning Path Preference Optimization
• Enhancing Large Language Model Reasoning via Selective Critical Token Fine-Tuning
• Unlocking Exploration in RLVR: Uncertainty-aware Advantage Shaping for Deeper Reasoning
• D3MAS: Decompose, Deduce, and Distribute for Enhanced Knowledge Sharing in Multi-Agent Systems
• Rethinking RL Evaluation: Can Benchmarks Truly Reveal Failures of RL Methods?
• Tracing the Traces: Latent Temporal Signals for Efficient and Accurate Reasoning
• LLM-Friendly Knowledge Representation for Customer Support
• Agro-Consensus: Semantic Self-Consistency in Vision-Language Models for Crop Disease Management in Developing Countries
• MatryoshkaThinking: Recursive Test-Time Scaling Enables Efficient Reasoning
• Audit-of-Understanding: Posterior-Constrained Inference for Mathematical Reasoning in Language Models
• Unilaw-R1: A Large Language Model for Legal Reasoning with Reinforcement Learning and Iterative Inference
• SwarmSys: Decentralized Swarm-Inspired Agents for Scalable and Adaptive Reasoning
• Failure-Driven Workflow Refinement
• Enhancing Large Language Model Reasoning with Reward Models: An Analytical Survey
• Ontology-Guided Neuro-Symbolic Inference: Grounding Language Models with Mathematical Domain Knowledge
• PromptGuard at BLP-2025 Task 1: A Few-Shot Classification Framework Using Majority Voting and Keyword Similarity for Bengali Hate Speech Detection
• Prompting Test-Time Scaling Is A Strong LLM Reasoning Data Augmentation
• SIMBA UQ: Similarity-Based Aggregation for Uncertainty Quantification in Large Language Models
• Mitigating Overthinking through Reasoning Shaping
• Agentic Systems in Radiology: Design, Applications, Evaluation, and Challenges
• Logit Arithmetic Elicits Long Reasoning Capabilities Without Training
• Safety Game: Balancing Safe and Informative Conversations with Blackbox Agentic AI using LP Solvers
• MEC³O: Multi-Expert Consensus for Code Time Complexity Prediction
• Repairing Regex Vulnerabilities via Localization-Guided Instructions
• MASA: LLM-Driven Multi-Agent Systems for Autoformalization
• Fundamentals of Building Autonomous LLM Agents
• Mind-Paced Speaking: A Dual-Brain Approach to Real-Time Reasoning in Spoken Language Models
• FOR-Prompting: From Objection to Revision via an Asymmetric Prompting Protocol
• RefGrader: Automated Grading of Mathematical Competition Proofs using Agentic Workflows
• Q-Router: Agentic Video Quality Assessment with Expert Model Routing and Artifact Localization
• CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewards
• Think Just Enough: Sequence-Level Entropy as a Confidence Signal for LLM Reasoning
• Dynamic Generation of Multi-LLM Agents Communication Topologies with Graph Diffusion Models
• GCPO: When Contrast Fails, Go Gold
• Parallel Test-Time Scaling for Latent Reasoning Models
• Selection, Reflection and Self-Refinement: Revisit Reasoning Tasks via a Causal Lens
• xRouter: Training Cost-Aware LLMs Orchestration System via Reinforcement Learning
• MOSAIC: Multi-agent Orchestration for Task-Intelligent Scientific Coding
• LOGicalThought: Logic-Based Ontological Grounding of LLMs for High-Assurance Reasoning
• TS-Agent: Understanding and Reasoning Over Raw Time Series via Iterative Insight Gathering
• CARPAS: Towards Content-Aware Refinement of Provided Aspects for Summarization in Large Language Models
• Reasoning for Hierarchical Text Classification: The Case of Patents
• On Randomness in Agentic Evals
• CLUE: Non-parametric Verification from Experience via Hidden-State Clustering
• Think Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression
• Towards Interpretable and Inference-Optimal COT Reasoning with Sparse Autoencoder-Guided Generation
• Do LLMs Know They Are Being Tested? Evaluation Awareness and Incentive-Sensitive Failures in GPT-OSS-20B
• Efficient numeracy in language models through single-token number embeddings
• GPT-5 Model Corrected GPT-4V's Chart Reading Errors, Not Prompting
• Foundations of LLM Knowledge Materialization: Termination, Reproducibility, Robustness
• Iterative LLM-Based Generation and Refinement of Distracting Conditions in Math Word Problems
• On the Role of Temperature Sampling in Test-Time Scaling
• Optimal Stopping vs Best-of-N for Inference Time Optimization
• Auto-Prompt Ensemble for LLM Judge
• Adaptive Tool Generation with Models as Tools and Reinforcement Learning
• Get RICH or Die Scaling: Profitably Trading Inference Compute for Robustness
• AsyncSpade: Efficient Test-Time Scaling with Asynchronous Sparse Decoding
• Self-signals Driven Multi-LLM Debate for Efficient and Accurate Reasoning
• Valid Stopping for LLM Generation via Empirical Dynamic Formal Lift
• Belief-Calibrated Multi-Agent Consensus Seeking for Complex NLP Tasks
• Pushing Test-Time Scaling Limits of Deep Search with Asymmetric Verification
• Towards Label-Free Biological Reasoning Synthetic Dataset Creation via Uncertainty Filtering
• ARM: Discovering Agentic Reasoning Modules for Generalizable Multi-Agent Systems
• AQA-TTRL: Self-Adaptation in Audio Question Answering with Test-Time Reinforcement Learning
• Mission Impossible: Feedback-Guided Dynamic Interactive Planning for Improving Reasoning on LLMs
• Sample Smart, Not Hard: Correctness-First Decoding for Better Reasoning in LLMs
• Classical AI vs. LLMs for Decision-Maker Alignment in Health Insurance Choices
• RareAgent: Self-Evolving Reasoning for Drug Repurposing in Rare Diseases
• Verifier-free Test-Time Sampling for Vision-Language-Action Models
• When Thinking Drifts: Evidential Grounding for Robust Video Reasoning
• AgentRouter: A Knowledge-Graph-Guided LLM Router for Collaborative Multi-Agent Question Answering
• Let it Calm: Exploratory Annealed Decoding for Verifiable Reinforcement Learning
• Slm-mux: Orchestrating small language models for reasoning
• SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs
• RLAD: Training LLMs to Discover Abstractions for Solving Reasoning Problems
• AI-to-AI Feedback: Amplified Intelligence — Prior Art and Governance Implications for Multi-Model Advisory Architectures
• Modeling Student Learning with 3.8 Million Program Traces
• Test-Time Scaling in Diffusion LLMs via Hidden Semi-Autoregressive Experts
• Resource-Efficient Fine-Tuning of LLaMA-3.2-3B for Medical Chain-of-Thought Reasoning
• Bridging Reasoning to Learning: Unmasking Illusions using Complexity Out of Distribution Generalization
• Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models
• Multi-Agent Collaborative Intelligence: Dual-Dial Control for Reliable LLM Reasoning
• MedCLM: Learning to Localize and Reason via a CoT-Curriculum in Medical Vision-Language Models
• Natural Language Edge Labelling: Decoupling Intent from Execution in Structured LM Reasoning
• Learning on the Job: Test-Time Curricula for Targeted Reinforcement Learning
• AlphaApollo: Orchestrating Foundation Models and Professional Tools into a Self-Evolving System for Deep Agentic Reasoning
• Open Agent Specification (Agent Spec): A Unified Representation for AI Agents
• PatternKV: Flattening KV Representation Expands Quantization Headroom
• Large Language Models Hallucination: A Comprehensive Survey
• SPOGW: a Score-based Preference Optimization method via Group-Wise comparison for workflows
• Increasing LLM response trustworthiness using voting ensembles
• Chronological Thinking in Full-Duplex Spoken Dialogue Language Models
• Searching Meta Reasoning Skeleton to Guide LLM Reasoning
• Don't Pass@k: A Bayesian Framework for Large Language Model Evaluation
• Less Diverse, Less Safe: The Indirect But Pervasive Risk of Test-Time Scaling in Large Language Models
• On Using Large Language Models to Enhance Clinically-Driven Missing Data Recovery Algorithms in Electronic Health Records
• MITS: Enhanced Tree Search Reasoning for LLMs via Pointwise Mutual Information
• GuidedSampling: Steering LLMs Towards Diverse Candidate Solutions at Inference-Time
• Efficient Test-Time Scaling for Small Vision-Language Models
• Sample, Align, Synthesize: Graph-Based Response Synthesis with ConGrs
• Best-of-Majority: Minimax-Optimal Strategy for Pass@k Inference Scaling
• Lateral Tree-of-Thoughts Surpasses ToT by Incorporating Logically-Consistent, Low-Utility Candidates
• Self-Reflective Generation at Test Time
• RoiRL: Efficient, Self-Supervised Reasoning with Offline Iterative Reinforcement Learning
• NCV: A Node-Wise Consistency Verification Approach for Low-Cost Structured Error Localization in LLM Reasoning
• Knowledge Graph-Guided Multi-Agent Distillation for Reliable Industrial Question Answering with Datasets
• Retrv-R1: A Reasoning-Driven MLLM Framework for Universal and Efficient Multimodal Retrieval
• AutoMaAS: Self-Evolving Multi-Agent Architecture Search for Large Language Models
• StepChain GraphRAG: Reasoning Over Knowledge Graphs for Multi-Hop Question Answering
• Typed Chain-of-Thought: A Curry-Howard Framework for Verifying LLM Reasoning
• Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifiers
• Hearing the Order: Investigating Selection Bias in Large Audio-Language Models
• When Silence Matters: The Impact of Irrelevant Audio on Text Reasoning in Large Audio-Language Models
• JoyAgent-JDGenie: Technical Report on the GAIA
• Curiosity-Driven LLM-as-a-judge for Personalized Creative Judgment
• Generalized Parallel Scaling with Interdependent Generations
• Rethinking Thinking Tokens: LLMs as Improvement Operators
• Making, not Taking, the Best of N
• CodeChemist: Functional Knowledge Transfer for Low-Resource Code Generation via Test-Time Scaling
• Stochastic Self-Organization in Multi-Agent Systems
• Beyond Majority Voting: LLM Aggregation by Leveraging Higher-Order Information
• Training Large Language Models To Reason In Parallel With Global Forking Tokens
• Judging with Confidence: Calibrating Autoraters to Preference Distributions
• BiasFreeBench: a Benchmark for Mitigating Bias in Large Language Model Responses
• GRPO-λ: Credit Assignment improves LLM Reasoning
• TAMA: Tool-Augmented Multimodal Agent for Procedural Activity Understanding
• Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models
• Probing the Critical Point (CritPt) of AI Reasoning: a Frontier Physics Research Benchmark
• Go with Your Gut: Scaling Confidence for Autoregressive Image Generation
• Memory-Driven Self-Improvement for Decision Making with Large Language Models
• Latent Thinking Optimization: Your Latent Reasoning Language Model Secretly Encodes Reward Signals in Its Latent Thoughts
• Reinforced Strategy Optimization for Conversational Recommender Systems via Network-of-Experts
• DyFlow: Dynamic Workflow Framework for Agentic Reasoning
• Boosting Process-Correct CoT Reasoning by Modeling Solvability of Multiple-Choice QA
• Chain-in-Tree: Back to Sequential Reasoning in LLM Tree Search
• Self-Evolving Vision-Language Models for Image Quality Assessment via Voting and Ranking
• TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning
• SING-SQL: A Synthetic Data Generation Framework for In-Domain Text-to-SQL Translation
• Nudging the Boundaries of LLM Reasoning
• Uncertainty-Aware Answer Selection for Improved Reasoning in Multi-LLM Systems
• Explainable Fault Localization for Programming Assignments via LLM-Guided Annotation
• Automated Model Discovery via Multi-modal & Multi-step Pipeline
• Planner-R1: Reward Shaping Enables Efficient Agentic RL with Smaller LLMs
• Voice Evaluation of Reasoning Ability: Diagnosing the Modality-Induced Performance Gap
• SimulRAG: Simulator-based RAG for Grounding LLMs in Long-form Scientific QA
• DeepSearch: Overcome the Bottleneck of Reinforcement Learning with Verifiable Rewards via Monte Carlo Tree Search
• Adaptive Test-Time Reasoning via Reward-Guided Dual-Phase Search
• From Perception to Cognition: A Survey of Vision-Language Interactive Reasoning in Multimodal Large Language Models
• PixelCraft: A Multi-Agent System for High-Fidelity Visual Reasoning on Structured Images
• Rethinking Entropy Regularization in Large Reasoning Models
• SecInfer: Preventing Prompt Injection via Inference-time Scaling
• The Dialogue That Heals: A Comprehensive Evaluation of Doctor Agents' Inquiry Capability
• Intra-request branch orchestration for efficient LLM reasoning
• Retro*: Optimizing LLMs for Reasoning-Intensive Document Retrieval
• A Hierarchical Error Framework for Reliable Automated Coding in Communication Research: Applications to Health and Political Communication
• KnowGuard: Knowledge-Driven Abstention for Multi-Round Clinical Reasoning
• LatentEvolve: Self-Evolving Test-Time Scaling in Latent Space
• From Ambiguity to Verdict: A Semiotic-Grounded Multi-Perspective Agent for LLM Logical Reasoning
• CDT: A Comprehensive Capability Framework for Large Language Models Across Cognition, Domain, and Task
• Plan before Solving: Problem-Aware Strategy Routing for Mathematical Reasoning with LLMs
• From Static to Dynamic: Adaptive Monte Carlo Search for Mathematical Process Supervision
• MAS²: Self-Generative, Self-Configuring, Self-Rectifying Multi-Agent Systems
• MedMMV: A Controllable Multimodal Multi-Agent Framework for Reliable and Verifiable Clinical Reasoning
• SimuHome: A Temporal- and Environment-Aware Benchmark for Smart Home LLM Agents
• Learning to Ponder: Adaptive Reasoning in Latent Space
• Can Large Language Models Express Uncertainty Like Human?
• Dynamic Policy Induction for Adaptive Prompt Optimization: Bridging the Efficiency-Accuracy Gap via Lightweight Reinforcement Learning
• Think Twice, Generate Once: Safeguarding by Progressive Self-Reflection
• ContextPRM: Leveraging Contextual Coherence for multi-domain Test-Time Scaling
• Calibrating Verbalized Confidence with Self-Generated Distractors
• Not Wrong, But Untrue: LLM Overconfidence in Document-Based Queries
• Expanding Computation Spaces of LLMs at Inference Time
• Agentar-Scale-SQL: Advancing Text-to-SQL through Orchestrated Test-Time Scaling
• ReasonCACHE: Teaching LLMs To Reason Without Weight Updates
• Beyond Magic Words: Sharpness-Aware Prompt Evolving for Robust Large Language Models with TARE
• Explore-Execute Chain: Towards an Efficient Structured Reasoning Paradigm
• Fast Thinking for Large Language Models
• MASH: A Multiplatform and Multimodal Annotated Dataset for Societal Impact of Hurricane
• Reasoning Scaffolding: Distilling the Flow of Thought from LLMs
• Do Repetitions Matter? Strengthening Reliability in LLM Evaluations
• EAPO: Enhancing Policy Optimization with On-Demand Expert Assistance
• Cognition-of-Thought Elicits Social-Aligned Reasoning in Large Language Models
• Language, Culture, and Ideology: Personalizing Offensiveness Detection in Political Tweets with Reasoning LLMs
• Scaling LLM Test-Time Compute with Mobile NPU on Smartphones
• Decoupling Reasoning and Perception: An LLM-LMM Framework for Faithful Visual Reasoning
• Learning How to Use Tools, Not Just When: Pattern-Aware Tool-Integrated Reasoning
• Scaling Policy Compliance Assessment in Language Models with Policy Reasoning Traces
• MathBode: Measuring the Stability of LLM Reasoning using Frequency Response
• Tracing Uncertainty in Language Model "Reasoning"
• Agentic AI Reasoning for Mobile Edge General Intelligence: Fundamentals, Approaches, and Directions
• From Harm to Help: Turning Reasoning In-Context Demos into Assets for Reasoning LMs
• AutoEP: LLMs-Driven Automation of Hyperparameter Evolution for Metaheuristic Algorithms
• Diagnose, Localize, Align: A Full-Stack Framework for Reliable LLM Multi-Agent Systems under Instruction Conflicts
• PonderLM-2: Pretraining LLM with Latent Thoughts in Continuous Space
• LAGEA: Language Guided Embodied Agents for Robotic Manipulation
• Critique to Verify: Accurate and Honest Test-Time Scaling with RL-Trained Verifiers
• Training Vision-Language Process Reward Models for Test-Time Scaling in Multimodal Reasoning: Key Insights and Lessons Learned
• Protocode: Prototype-Driven Interpretability for Code Generation in LLMs
• Taming Variability: Randomized and Bootstrapped Conformal Risk Control for LLMs
• Semantic Voting: A Self-Evaluation-Free Approach for Efficient LLM Self-Improvement on Unverifiable Open-ended Tasks
• HEART: Emotionally-driven test-time scaling of Language Models
• UML-CoT: Structured Reasoning and Planning with Unified Modeling Language for Robotic Room Cleaning
• Dynamic Experts Search: Enhancing Reasoning in Mixture-of-Experts LLMs at Test Time
• InfiAgent: Self-Evolving Pyramid Agent Framework for Infinite Scenarios
• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• PRIME: Planning and Retrieval-Integrated Memory for Enhanced Reasoning
• From Long to Lean: Performance-aware and Adaptive Chain-of-Thought Compression via Multi-round Refinement
• R-Capsule: Compressing High-Level Plans for Efficient Large Language Model Reasoning
• Multi-Agent Path Finding via Offline RL and LLM Collaboration
• Think Right, Not More: Test-Time Scaling for Numerical Claim Verification
• Reinforcement Learning-Guided Chain-of-Draft for Token-Efficient Code Generation
• Fuzzy Reasoning Chain (FRC): An Innovative Reasoning Framework from Fuzziness to Clarity
• A2R: An Asymmetric Two-Stage Reasoning Framework for Parallel Reasoning
• Teaching Transformers to Solve Combinatorial Problems through Efficient Trial & Error
• Benchmarking and Mitigating Sycophancy in Medical Vision Language Models
• Mixture-of-Visual-Thoughts: Exploring Context-Adaptive Reasoning Mode Selection for General Visual Reasoning
• Retrieval-of-Thought: Efficient Reasoning via Reusing Thoughts
• Thinking in Many Modes: How Composite Reasoning Elevates Large Language Model Performance with Limited Data
• Why Chain of Thought Fails in Clinical Text Understanding
• Unsupervised Conformal Inference: Bootstrapping and Alignment to Control LLM Uncertainty
• Towards Transparent AI: A Survey on Explainable Language Models
• Correct Reasoning Paths Visit Shared Decision Pivots
• Plan2Evolve: LLM Self-Evolution for Improved Planning Capability via Automated Domain Generation
• Dual-Head Reasoning Distillation: Improving Classifier Accuracy with Train-Time-Only Reasoning
• A State-of-the-Art SQL Reasoning Model using RLVR
• VideoJudge: Bootstrapping Enables Scalable Supervision of MLLM-as-a-Judge for Video Understanding
• Bounds of Chain-of-Thought Robustness: Reasoning Steps, Embed Norms, and Beyond
• Best-of-∞ -- Asymptotic Performance of Test-Time Compute
• Verification Limits Code LLM Training
• Distilling Many-Shot In-Context Learning into a Cheat Sheet
• LogReasoner: Empowering LLMs with Expert-like Coarse-to-Fine Reasoning for Automated Log Analysis
• TyphoonMLA: A Mixed Naive-Absorb MLA Kernel For Shared Prefix
• A Fano-Style Accuracy Upper Bound for LLM Single-Pass Reasoning in Multi-Hop QA
• Difference-Guided Reasoning: A Temporal-Spatial Framework for Large Language Models
• RJE: A Retrieval-Judgment-Exploration Framework for Efficient Knowledge Graph Question Answering with LLMs
• Mixture of Thoughts: Learning to Aggregate What Experts Think, Not Just What They Say
• MARS: toward more efficient multi-agent collaboration for LLM reasoning
• Instruction Boundary: Quantifying Biases in LLM Reasoning under Various Coverage
• SIM-CoT: Supervised Implicit Chain-of-Thought
• SKYLENAGE Technical Report: Mathematical Reasoning and Contest-Innovation Benchmarks for Multi-Level Math Evaluation
• LOCA: Logical Chain Augmentation for Scientific Corpus Cleaning
• Mamba Modulation: On the Length Generalization of Mamba
• STARQA: A Question Answering Dataset for Complex Analytical Reasoning over Structured Databases
• What makes prompts a graph: necessary and sufficient conditions for prompt graph engineering
• Σ-Mem: An Online Reliability Memory for LLM-based Multi-Agent Systems
• LEDGERMIND: Provenance-Constrained Multimodal Agentic Reasoning with a Structured Evidence Ledger
• Distilling Answer Set Programming Theories from Large Language Models
• Open Security Benchmark: Towards Autonomous Enterprise Cyber Defense
• Leveraging Trajectory Graphs for Pre-Execution Error Diagnosis in Agentic LLM Systems
• Bridging Inference-Time Scaling and Episodic Memory with Action-Centric Graphs
• Sympathetic Framing: Evaluating AI Alignment across Sociodemographic Groups
• LASAR: Latent Adaptive Semantic Aligned Reasoning for Generative Recommendation
• Can neural networks do arithmetic? A survey on the elementary numerical skills of state-of-the-art deep learning models
• RELISH: LLM REgression with a Latent Iterative State Head
• Verbalizing LLM's Higher-order Uncertainty via Imprecise Probabilities
• Geometric Risk Control for Vision-Language Model OCR
• Language Models Coupled with Metacognition Can Outperform Reasoning Models
• Large language model consensus substantially improves the cell type annotation accuracy for scRNA-seq data
• Pandora: Leveraging Code-driven Knowledge Transfer for Unified Structured Knowledge Reasoning
• DRQA: Dynamic Reasoning Quota Allocation for Controlling Overthinking in Reasoning Large Language Models
• Estimating the Self-Consistency of LLMs
• Citrus-V: Advancing Medical Foundation Models with Unified Medical Image Grounding for Clinical Reasoning
• LLM-based Agents Suffer from Hallucinations: A Survey of Taxonomy, Methods, and Directions
• Solving Math Word Problems Using Estimation Verification and Equation Generation
• CCQA: Generating Question from Solution Can Improve Inference-Time Reasoning in SLMs
• Confidence-Aware Routing for Large Language Model Reliability Enhancement: A Multi-Signal Approach to Pre-Generation Hallucination Mitigation
• When TableQA Meets Noise: A Dual Denoising Framework for Complex Questions and Large-scale Tables
• Correlation or Causation: Analyzing the Causal Structures of LLM and LRM Reasoning Process
• MSCoRe: A Benchmark for Multi-Stage Collaborative Reasoning in LLM Agents
• Mitigating Strategy-Selection Bias in Reasoning for More Effective Test-Time Scaling
• Variation in Verification: Understanding Verification Dynamics in Large Language Models
• Can Agents Judge Systematic Reviews Like Humans? Evaluating SLRs with LLM-based Multi-Agent System
• Localizing Malicious Outputs from CodeLLM
• K-DeCore: Facilitating Knowledge Transfer in Continual Structured Knowledge Reasoning via Knowledge Decoupling
• MoEs Are Stronger than You Think: Hyper-Parallel Inference Scaling with RoE
• V-CECE: Visual Counterfactual Explanations via Conceptual Edits
• A Novel Differential Feature Learning for Effective Hallucination Detection and Classification
• SCAN: Self-Denoising Monte Carlo Annotation for Robust Process Reward Learning
• SalaMAnder: Shapley-based Mathematical Expression Attribution and Metric for Chain-of-Thought Reasoning
• From Scores to Steps: Diagnosing and Improving LLM Performance in Evidence-Based Medical Calculations
• Best-of-L: Cross-Lingual Reward Modeling for Mathematical Reasoning
• Spectral Logit Sculpting: Adaptive Low-Rank Logit Transformation for Controlled Text Generation
• GPO: Learning from Critical Steps to Improve LLM Reasoning
• Temporal Reasoning with Large Language Models Augmented by Evolving Knowledge Graphs
• LLM Agents at the Roundtable: A Multi-Perspective and Dialectical Reasoning Framework for Essay Scoring
• The NazoNazo Benchmark: A Cost-Effective and Extensible Test of Insight-Based Reasoning in LLMs
• (P)rior(D)yna(F)low: A Priori Dynamic Workflow Construction via Multi-Agent Collaboration
• Large Language Models in Operations Research: Methods, Applications, and Challenges
• Fleming-R1: Toward Expert-Level Medical Reasoning via Reinforcement Learning
• From Pixels to Urban Policy-Intelligence: Recovering Legacy Effects of Redlining with a Multimodal LLM
• Self-Improvement of Language Models by Post-Training on Multi-Agent Debate
• Class-Invariant Test-Time Augmentation for Domain Generalization
• A Taxonomy of Prompt Defects in LLM Systems
• Compute as Teacher: Turning Inference Compute Into Reference-Free Supervision
• MICA: Multi-Agent Industrial Coordination Assistant
• Enhancing Multi-Agent Debate System Performance via Confidence Expression
• Speech-Based Cognitive Screening: A Systematic Evaluation of LLM Adaptation Strategies
• Thinking in a Crowd: How Auxiliary Information Shapes LLM Reasoning
• DSCC-HS: A Dynamic Self-Reinforcing Framework for Hallucination Suppression in Large Language Models
• Early Stopping Chain-of-thoughts in Large Language Models
• Slim-SC: Thought Pruning for Efficient Scaling with Self-Consistency
• FVDebug: An LLM-Driven Debugging Assistant for Automated Root Cause Analysis of Formal Verification Failures
• Single-stream Policy Optimization
• REFINER: Reasoning Feedback on Intermediate Representations
• The LLM Already Knows: Estimating LLM-Perceived Question Difficulty via Hidden Representations
• LTA-thinker: Latent Thought-Augmented Training Framework for Large Language Models on Complex Reasoning
• Leveraging Large Language Models to Effectively Generate Visual Data for Canine Musculoskeletal Diagnoses
• Routing Distilled Knowledge via Mixture of LoRA Experts for Large Language Model based Bundle Generation
• Analogy-Driven Financial Chain-of-Thought (AD-FCoT): A Prompting Approach for Financial Sentiment Analysis
• Shaping Explanations: Semantic Reward Modeling with Encoder-Only Transformers for GRPO
• Large Language Model-Based Automatic Formulation for Stochastic Optimization Models
• Selective Risk Certification for LLM Outputs via Information-Lift Statistics: PAC-Bayes, Robustness, and Skeleton Design
• A Fine-Grained 3D Radio Map Construction Paradigm with Ultra-Low Sampling Rates by Large Generative Models
• Audited Reasoning Refinement: Fine-Tuning Language Models via LLM-Guided Step-Wise Evaluation and Correction
• HalluDetect: Detecting, Mitigating, and Benchmarking Hallucinations in Conversational Systems in the Legal Domain
• The Prompt Engineering Report Distilled: Quick Start Guide for Life Sciences
• Predictable Compression Failures: Why Language Models Actually Hallucinate
• Free-MAD: Consensus-Free Multi-Agent Debate
• PersonaX: Multimodal Datasets with LLM-Inferred Behavior Traits
• Improving Table Understanding with LLMs and Entity-Oriented Search
• Reasoning Under Uncertainty: Exploring Probabilistic Reasoning Capabilities of LLMs
• Planning for Success: Exploring LLM Long-term Planning Capabilities in Table Understanding
• DeepDive: Advancing Deep Search Agents with Knowledge Graphs and Multi-Turn RL
• RefactorCoderQA: Benchmarking LLMs for Multi-Domain Coding Question Solutions in Cloud and Edge Deployment
• Who Decides How Knowing Becomes Doing? Redistributing Authority in Human-AI Music Co-Creation
• On LLM-Based Scientific Inductive Reasoning Beyond Equations
• Large Language Models Meet Legal Artificial Intelligence: A Survey
• Smart Trial: Evaluating the Use of Large Language Models for Recruiting Clinical Trial Participants via Social Media
• Unsupervised Hallucination Detection by Inspecting Reasoning Processes
• Limited Reference, Reliable Generation: A Two-Component Framework for Tabular Data Generation in Low-Data Regimes
• Enhancing LLM-based Specification Generation via Program Slicing and Logical Deletion
• XAgents: A Unified Framework for Multi-Agent Cooperation via IF-THEN Rules and Multipolar Task Processing Graph
• CDE: Curiosity-Driven Exploration for Efficient Reinforcement Learning in Large Language Models
• Visual Programmability: A Guide for Code-as-Thought in Chart Understanding
• Jupiter: Enhancing LLM Data Analysis Capabilities via Notebook and Inference-Time Value-Guided Search
• Quality Assessment of Tabular Data using Large Language Models and Code Generation
• Latency and Token-Aware Test-Time Compute
• PromptGuard: An Orchestrated Prompting Framework for Principled Synthetic Text Generation for Vulnerable Populations using LLMs with Enhanced Safety, Fairness, and Controllability
• Evaluating LLMs Without Oracle Feedback: Agentic Annotation Evaluation Through Unsupervised Consistency Signals
• Leveraging AI Agents for Autonomous Networks: A Reference Architecture and Empirical Studies
• Improving MLLM Historical Record Extraction with Test-Time Image
• CM-Align: Consistency-based Multilingual Alignment for Large Language Models
• AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning
• Amortized Latent Steering: Low-Cost Alternative to Test-Time Optimization
• XML Prompting as Grammar-Constrained Interaction: Fixed-Point Semantics, Convergence Guarantees, and Human-AI Protocols
• Parallel-R1: Towards Parallel Thinking via Reinforcement Learning
• Video Parallel Scaling: Aggregating Diverse Frame Subsets for VideoLLMs
• HealthSLM-Bench: Benchmarking Small Language Models for Mobile and Wearable Healthcare Monitoring
• RAFFLES: Reasoning-based Attribution of Faults for LLM Systems
• MachineLearningLM: Scaling Many-shot In-context Learning via Continued Pretraining
• Test-Time Scaling in Reasoning Models Is Not Effective for Knowledge-Intensive Tasks Yet
• From Implicit Exploration to Structured Reasoning: Leveraging Guideline and Refinement for LLMs
• The Majority is not always right: RL training for solution aggregation
• Benchmarking Information Retrieval Models on Complex Retrieval Tasks
• X-SQL: Expert Schema Linking and Understanding of Text-to-SQL with Multi-LLMs
• From Long to Short: LLMs Excel at Trimming Own Reasoning Chains
• Mitigating Spurious Correlations Between Question and Answer via Chain-of-Thought Correctness Perception Distillation
• Cross-Question Method Reuse in Large Language Models: From Word-Level Prediction to Rational Logical-Layer Reasoning
• Self-Aligned Reward: Towards Effective and Efficient Reasoners
• Sticker-TTS: Learn to Utilize Historical Experience with a Sticker-driven Test-Time Scaling Framework
• Less is More Tokens: Efficient Math Reasoning via Difficulty-Aware Chain-of-Thought Distillation
• DreamPRM-1.5: Unlocking the Potential of Each Instance for Multimodal Process Reward Model Training
• Talk Isn't Always Cheap: Understanding Failure Modes in Multi-Agent Debate
• FLAMES: Improving LLM Math Reasoning via a Fine-Grained Analysis of the Data Synthesis Pipeline
• Psychologically Enhanced AI Agents
• CoT-Space: A Theoretical Framework for Internal Slow-Thinking via Reinforcement Learning
• Quantized Large Language Models in Biomedical Natural Language Processing: Evaluation and Recommendation
• A Comprehensive Survey on Trustworthiness in Reasoning with Large Language Models
• A Probabilistic Inference Scaling Theory for LLM Self-Correction
• Beyond Interpretability: Exploring the Comprehensibility of Adaptive Video Streaming through Large Language Models
• Explainable Knowledge Graph Retrieval-Augmented Generation (KG-RAG) with KG-SMILE
• Learning When to Plan: Efficiently Allocating Test-Time Compute for LLM Agents
• Easier Painting Than Thinking: Can Text-to-Image Models Set the Stage, but Not Direct the Play?
• Knowledge Integration for Physics-informed Symbolic Regression Using Pre-trained Large Language Models
• ReCode: Improving LLM-based Code Repair with Fine-Grained Retrieval-Augmented Generation
• Scale, Don't Fine-tune: Guiding Multimodal LLMs for Efficient Visual Place Recognition at Test-Time
• LLMs for LLMs: A Structured Prompting Methodology for Long Legal Documents
• Avoidance Decoding for Diverse Multi-Branch Story Generation
• Inducing Faithfulness in Structured Reasoning via Counterfactual Sensitivity
• Trusted Uncertainty in Large Language Models: A Unified Framework for Confidence Calibration and Risk-Controlled Refusal
• GradeSQL: Test-Time Inference with Outcome Reward Models for Text-to-SQL Generation from Large Language Models
• Vis-CoT: A Human-in-the-Loop Framework for Interactive Visualization and Intervention in LLM Chain-of-Thought Reasoning
• Aligning Reasoning LLMs for Materials Discovery with Physics-aware Rejection Sampling
• GIER: Gap-Driven Self-Refinement for Large Language Models
• Scaling Up, Speeding Up: A Benchmark of Speculative Decoding for Efficient LLM Test-Time Scaling
• Illuminating Patterns of Divergence: DataDios SmartDiff for Large-Scale Data Difference Analysis
• ParaThinker: Native Parallel Thinking as a New Paradigm to Scale LLM Test-time Compute
• PiCSAR: Probabilistic Confidence Selection And Ranking for Reasoning Chains
• Reasoning-Intensive Regression
• How Well Do Vision--Language Models Understand Cities? A Comparative Study on Spatial Reasoning from Street-View Images
• Can Multiple Responses from an LLM Reveal the Sources of Its Uncertainty?
• An Agile Method for Implementing Retrieval Augmented Generation Tools in Industrial SMEs
• Measuring Reasoning Utility in LLMs via Conditional Entropy Reduction
• CyberSleuth: Autonomous Blue-Team LLM Agent for Web Attack Forensics
• Automated Quality Assessment for LLM-Based Complex Qualitative Coding: A Confidence-Diversity Framework
• PersoNo: Personalised Notification Urgency Classifier in Mixed Reality
• LFD: Layer Fused Decoding to Exploit External Knowledge in Retrieval-Augmented Generation
• SLIM: Subtrajectory-Level Elimination for More Effective Reasoning
• Entropy-Guided Loop: Achieving Reasoning through Uncertainty-Aware Generation
• OmniHuman-1.5: Instilling an Active Mind in Avatars via Cognitive Simulation
• Can Structured Templates Facilitate LLMs in Tackling Harder Tasks? : An Exploration of Scaling Laws by Difficulty
• Beyond the Textual: Generating Coherent Visual Options for MCQs
• Tailored Teaching with Balanced Difficulty: Elevating Reasoning in Multimodal Chain-of-Thought via Prompt Curriculum
• Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning Tasks
• Thinking Before You Speak: A Proactive Test-time Scaling Approach
• VistaWise: Building Cost-Effective Agent with Cross-Modal Knowledge Graph for Minecraft
• MIRAGE: Scaling Test-Time Inference with Parallel Graph-Retrieval-Augmented Reasoning Chains
• InReAcTable: LLM-Powered Interactive Visual Data Story Construction from Tabular Data
• Named Entity Recognition of Historical Text via Large Language Model
• LiveMCP-101: Stress Testing and Diagnosing MCP-enabled Agents on Challenging Queries
• Deep Think with Confidence
• Fine-tuning and prompt engineering for large language models-based code review automation
• Trust but Verify! A Survey on Verification Design for Test-time Scaling
• Semantic Energy: Detecting LLM Hallucination Beyond Entropy
• Automated Optimization Modeling through Expert-Guided Large Language Model Reasoning
• Zero-knowledge LLM hallucination detection and mitigation through fine-grained cross-model consistency
• Your Reward Function for RL is Your Best PRM for Search: Unifying RL and Search-Based TTS
• Prompt Orchestration Markup Language
• Can Large Language Models (LLMs) Describe Pictures Like Children? A Comparative Corpus Study
• Expertise-aware Multi-LLM Recruitment and Collaboration for Medical Decision-Making
• CausalPlan: Empowering Efficient LLM Multi-Agent Collaboration Through Causality-Driven Planning
• Generics and Default Reasoning in Large Language Models
• Input-Time Scaling
• Self-Organizing Agent Network for LLM-based Workflow Automation
• Multimodal Chain of Continuous Thought for Latent-Space Reasoning in Vision-Language Models
• Can Large Models Teach Student Models to Solve Mathematical Problems Like Human Beings? A Reasoning Distillation Method via Multi-LoRA Interaction
• Is GPT-OSS Good? A Comprehensive Evaluation of OpenAI's Latest Open Source Models
• Navigating the Exploration-Exploitation Tradeoff in Inference-Time Scaling of Diffusion Models
• Fast, Slow, and Tool-augmented Thinking for LLMs: A Review
• Improving Pre-Trained Vision-Language-Action Policies with Model-Based Search
• Cognitive Decision Routing in Large Language Models: When to Think Fast, When to Think Slow
• Cost-Aware Contrastive Routing for LLMs
• ReaLM: Reflection-Enhanced Autonomous Reasoning with Small Language Models
• Mitigating Hallucinations in Large Language Models via Causal Reasoning
• AgentMental: An Interactive Multi-Agent Framework for Explainable and Adaptive Mental Health Assessment
• MM-R1: Unleashing the Power of Unified Multimodal Large Language Models for Personalized Image Generation
• ETTRL: Balancing Exploration and Exploitation in LLM Test-Time Reinforcement Learning Via Entropy Mechanism
• HOID-R1: Reinforcement Learning for Open-World Human-Object Interaction Detection Reasoning with Multimodal Large Language Model
• Hard Examples Are All You Need: Maximizing GRPO Post-Training Under Annotation Budgets
• Audio Flamingo Sound-CoT Technical Report: Improving Chain-of-Thought Reasoning in Sound Understanding
• PASS: Probabilistic Agentic Supernet Sampling for Interpretable and Adaptive Chest X-Ray Reasoning
• MAC: A Live Benchmark for Multimodal Large Language Models in Scientific Understanding
• MRFD: Multi-Region Fusion Decoding with Self-Consistency for Mitigating Hallucinations in LVLMs
• Pruning Long Chain-of-Thought of Large Reasoning Models via Small-Scale Preference Optimization
• MEML-GRPO: Heterogeneous Multi-Expert Mutual Learning for RLVR Advancement
• Episodic Memory Representation for Long-form Video Understanding
• ASPD: Unlocking Adaptive Serial-Parallel Decoding by Exploring Intrinsic Parallelism in LLMs
• Leveraging Large Language Models for Rare Disease Named Entity Recognition
• InternBootcamp Technical Report: Boosting LLM Reasoning with Verifiable Task Scaling
• Rational Inverse Reasoning: Few-Shot Imitation by Inferring Intent through Planning
• Privacy-protected Retrieval-Augmented Generation for Knowledge Graph Question Answering
• Train Long, Think Short: Curriculum Learning for Efficient Reasoning
• Time Is a Feature: Exploiting Temporal Dynamics in Diffusion Language Models
• Evaluating Large Language Models as Expert Annotators
• ThinkTuning: Instilling Cognitive Reflections without Distillation
• Towards Theoretical Understanding of Transformer Test-Time Computing: Investigation on In-Context Linear Regression
• Can LLMs Detect Their Confabulations? Estimating Reliability in Uncertainty-Aware Language Models
• Steerable Pluralism: Pluralistic Alignment via Few-Shot Comparative Regression
• Hallucination as a Computational Boundary: A Hierarchy of Inevitability and the Oracle Escape
• EndoCogniAgent: Closed-Loop Agentic Reasoning with Self-Consistency Validation for Endoscopic Diagnosis
• Uncertainty-Aware Semantic Decoding for LLM-Based Sequential Recommendation
• AURA: A Fine-Grained Benchmark and Decomposed Metric for Audio-Visual Reasoning
• Chain-of-Alpha: Unleashing the Power of Large Language Models for Alpha Mining in Quantitative Trading
• Mediator-Guided Multi-Agent Collaboration among Open-Source Models for Medical Decision-Making
• Navigating the Risks of Using Large Language Models for Text Annotation in Social Science Research
• First Ask Then Answer: A Framework Design for AI Dialogue Based on Supplementary Questioning with Large Language Models
• Learning by Teaching: Engaging Students as Instructors of Large Language Models in Computer Science Education
• Test-Time Reinforcement Learning for GUI Grounding via Region Consistency
• Decision-Making with Deliberation: Meta-reviewing as a Document-grounded Dialogue
• Incident Response Planning Using a Lightweight Large Language Model with Reduced Hallucination
• FedCoT: Communication-Efficient Federated Reasoning Enhancement for Large Language Models
• ConfAgents: A Conformal-Guided Multi-Agent Framework for Cost-Efficient Medical Diagnosis
• Charts-of-Thought: Enhancing LLM Visualization Literacy Through Structured Data Extraction
• Hop, Skip, and Overthink: Diagnosing Why Reasoning Models Fumble during Multi-Hop Analysis
• TRAIL: Joint Inference and Refinement of Knowledge Graphs with Large Language Models
• StepFun-Formalizer: Unlocking the Autoformalization Potential of LLMs through Knowledge-Reasoning Fusion
• PersonaEval: Are LLM Evaluators Human Enough to Judge Role-Play?
• Chain of Questions: Guiding Multimodal Curiosity in Language Models
• Deliberative Reasoning Network: An Uncertainty-Driven Paradigm for Belief-Tracked Inference with Pretrained Language Models
• Large Language Model's Multi-Capability Alignment in Biomedical Domain
• GeoSR: Cognitive-Agentic Framework for Probing Geospatial Knowledge Boundaries via Iterative Self-Refinement
• Enhancing Serendipity Recommendation System by Constructing Dynamic User Knowledge Graphs with Large Language Models
• Unravelling the Probabilistic Forest: Arbitrage in Prediction Markets
• GTPO: Stabilizing Group Relative Policy Optimization via Gradient and Entropy Control
• Toward a Trustworthy Optimization Modeling Agent via Verifiable Synthetic Data Generation
• Adaptive Command: Real-Time Policy Adjustment via Language Models in StarCraft II
• CTTS: Collective Test-Time Scaling
• Polymath: A Self-Optimizing Agent with Dynamic Hierarchical Workflow
• XFacta: Contemporary, Real-World Dataset and Evaluation for Multimodal Misinformation Detection with Multimodal LLMs
• CAAD: Context-Aware Adaptive Decoding for Truthful Text Generation
• A Survey on AgentOps: Categorization, Challenges, and Future Directions
• MArgE: Meshing Argumentative Evidence from Multiple Large Language Models for Justifiable Claim Verification
• A Confidence-Diversity Framework for Calibrating AI Judgement in Accessible Qualitative Coding Tasks
• TIBSTC-CoT: A Multi-Domain Instruction Dataset for Chain-of-Thought Reasoning in Language Models
• Test-time Prompt Intervention
• SE-Agent: Self-Evolution Trajectory Optimization in Multi-Step Reasoning with LLM-Based Agents
• Counterfactual Probing for Hallucination Detection and Mitigation in Large Language Models
• Uncertainty-Based Methods for Automated Process Reward Data Construction and Output Aggregation in Mathematical Reasoning
• ReflecSched: Solving Dynamic Flexible Job-Shop Scheduling via LLM-Powered Hierarchical Reflection
• Enhancing Math Reasoning in Small-sized LLMs via Preview Difficulty-Aware Intervention
• LLM-Assisted Model-Based Fuzzing of Protocol Implementations
• Accurate and Consistent Graph Model Generation from Text with Large Language Models
• Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models
• Medical Reasoning in the Era of LLMs: A Systematic Review of Enhancement Techniques and Applications
• Ordinal Folding Index: A Computable Metric for Self-Referential Semantics
• BAR Conjecture: the Feasibility of Inference Budget-Constrained LLM Services with Authenticity and Reasoning
• LENS: Learning Ensemble Confidence from Neural States for Multi-LLM Answer Integration
• Failures Are the Stepping Stones to Success: Enhancing Few-Shot In-Context Learning by Leveraging Negative Samples
• Hate in Plain Sight: On the Risks of Moderating AI-Generated Hateful Illusions
• MetaAgent: Automatically Constructing Multi-Agent Systems Based on Finite State Machines

Related

• Escape Sky-high Cost: Early-stopping Self-Consistency for Multi-step Reasoning
• Make Every Penny Count: Difficulty-Adaptive Self-Consistency for Cost-Efficient Reasoning
• Universal Self-Consistency for Large Language Model Generation
• Semantic Self-Consistency: Enhancing Language Model Reasoning via Semantic Weighting
• Just Ask One More Time! Self-Agreement Improves Reasoning of Language Models in (Almost) All Scenarios