Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding
2024/01/09 by Zilong Wang, Wang, Zilong, Hao Zhang +21 · 2 voices · 78 citations
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Text and Document Classification Technologies #Topic Modeling #cs.CL
paper · pdf · doi:10.48550/arxiv.2401.04398
openalex publication_date 2024/01/09 · arxiv published 2024/01/09 · arxiv updated 2024/01/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Abstract
Table-based reasoning with large language models (LLMs) is a promising direction to tackle many table understanding tasks, such as table-based question answering and fact verification. Compared with generic reasoning, table-based reasoning requires the extraction of underlying semantics from both free-form questions and semi-structured tabular data. Chain-of-Thought and its similar approaches incorporate the reasoning chain in the form of textual context, but it is still an open question how to effectively leverage tabular data in the reasoning chain. We propose the Chain-of-Table framework, where tabular data is explicitly used in the reasoning chain as a proxy for intermediate thoughts. Specifically, we guide LLMs using in-context learning to iteratively generate operations and update the table to represent a tabular reasoning chain. LLMs can therefore dynamically plan the next operation based on the results of the previous ones. This continuous evolution of the table forms a chain, showing the reasoning process for a given tabular problem. The chain carries structured information of the intermediate results, enabling more accurate and reliable predictions. Chain-of-Table achieves new state-of-the-art performance on WikiTQ, FeTaQA, and TabFact benchmarks across multiple LLM choices.
Cited by
- Debugging Tabular Log as Dynamic Graphs
- Evolving from Lessons: Skill-Augmented Table Graph Reasoning for Operation-wise Table Question Answering
- JT-DA: Enhancing Data Analysis with Tool-Integrated Table Reasoning Large Language Models
- Prompting-in-a-Series: Psychology-Informed Contents and Embeddings for Personality Recognition With Decoder-Only Models
- Thucy: An LLM-based Multi-Agent System for Claim Verification across Relational Databases
- Table as a Modality for Large Language Models
- Evidence-Guided Schema Normalization for Temporal Tabular Reasoning
- Multi-Agent Multimodal Large Language Model Framework for Automated Interpretation of Fuel Efficiency Analytics in Public Transportation
- STaR: Towards Cognitive Table Reasoning via Slow-Thinking Large Language Models
- TabRAG: Improving Tabular Document Question Answering for Retrieval Augmented Generation via Structured Representations
- TabDSR: Decompose, Sanitize, and Reason for Complex Numerical Reasoning in Tabular Data
- A Survey of Data Agents: Emerging Paradigm or Overstated Hype?
- A Survey of AI Scientists
- Exploring Generative Process Reward Modeling for Semi-Structured Data: A Case Study of Table Question Answering
- Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table Understanding
- Table Question Answering in the Era of Large Language Models: A Comprehensive Survey of Tasks, Methods, and Evaluation
- Efficient numeracy in language models through single-token number embeddings
- TaTToo: Tool-Grounded Thinking PRM for Test-Time Scaling in Tabular Reasoning
- Training-Free Time Series Classification via In-Context Reasoning with LLM Agents
- SPOGW: a Score-based Preference Optimization method via Group-Wise comparison for workflows
- AutoPK: Leveraging LLMs and a Hybrid Similarity Metric for Advanced Retrieval of Pharmacokinetic Data from Complex Tables and Documents
- When TableQA Meets Noise: A Dual Denoising Framework for Complex Questions and Large-scale Tables
- MontePrep: Monte-Carlo-Driven Automatic Data Preparation without Target Data Instances
- Improving Table Understanding with LLMs and Entity-Oriented Search
- Planning for Success: Exploring LLM Long-term Planning Capabilities in Table Understanding
- TableMind: An Autonomous Programmatic Agent for Tool-Augmented Table Reasoning
- M3TQA: Massively Multilingual Multitask Table Question Answering
- TableZoomer: A Collaborative Agent Framework for Large-scale Table Question Answering
- Utilizing Training Data to Improve LLM Reasoning for Tabular Understanding
- Structuring the Unstructured: A Systematic Review of Text-to-Structure Generation for Agentic AI with a Universal Evaluation Framework
- Chain-of-Query: Unleashing the Power of LLMs in SQL-Aided Table Understanding via Multi-Agent Collaboration
- PanelTR: Zero-Shot Table Reasoning Framework Through Multi-Agent Scientific Discussion
- Tabular Data Understanding with LLMs: A Survey of Recent Advances and Challenges
- Enabling Few-Shot Alzheimer's Disease Diagnosis on Biomarker Data with Tabular LLMs
- Beyond Natural Language Plans: Structure-Aware Planning for Query-Focused Table Summarization
- Assemble Your Crew: Automatic Multi-agent Communication Topology Design via Autoregressive Graph Generation
- AraTable: Benchmarking LLMs' Reasoning and Understanding of Arabic Tabular Data
- MRT at IberLEF-2025 PRESTA Task: Maximizing Recovery from Tables with Multiple Steps
- Evolving Roles of LLMs in Scientific Innovation: Assistant, Collaborator, Scientist, and Evaluator
- Improved LLM Agents for Financial Document Question Answering
- MasHost Builds It All: Autonomous Multi-Agent System Directed by Reinforcement Learning
- ExpliCIT-QA: Explainable Code-Based Image Table Question Answering
- Toward Real-World Table Agents: Capabilities, Workflows, and Design Principles for LLM-based Table Intelligence
- TableDreamer: Progressive and Weakness-guided Data Synthesis from Scratch for Table Instruction Tuning
- TableReasoner: Advancing Table Reasoning Framework with Large Language Models
- What to Keep and What to Drop: Adaptive Table Filtering Framework
- A Modular Multitask Reasoning Framework Integrating Spatio-temporal Models and LLMs
- MDBench: A Synthetic Multi-Document Reasoning Benchmark Generated with Knowledge Guidance
- Document-Level Tabular Numerical Cross-Checking: A Coarse-to-Fine Approach
- No Universal Prompt: Unifying Reasoning through Adaptive Prompting for Temporal Table Reasoning
- TableRAG: A Retrieval Augmented Generation Framework for Heterogeneous Document Reasoning
- Table-Text Alignment: Explaining Claim Verification Against Tables in Scientific Papers
- Table-r1: Self-supervised and Reinforcement Learning for Program-based Table Reasoning in Small Language Models
- LLM-Symbolic Integration for Robust Temporal Tabular Reasoning
- MAPLE: Multi-Agent Adaptive Planning with Long-Term Memory for Table Reasoning
- Multimodal Tabular Reasoning with Privileged Structured Information
- TableEval: A Real-World Benchmark for Complex, Multilingual, and Multi-Structured Table Question Answering
- Reasoning-Table: Exploring Reinforcement Learning for Table Reasoning
- Speculative Reward Model Boosts Decision Making Ability of LLMs Cost-Effectively
- LLM Inference Enhanced by External Knowledge: A Survey
- Fortune: Formula-Driven Reinforcement Learning for Symbolic Table Reasoning in Language Models
- MRT at SemEval-2025 Task 8: Maximizing Recovery from Tables with Multiple Steps
- Position: Foundation Models for Tabular Data within Systemic Contexts Need Grounding
- Weaver: Interweaving SQL and LLM for Table Reasoning
- Training with Pseudo-Code for Instruction Following
- Text-to-Pipeline: Bridging Natural Language and Data Preparation Pipelines
- RoT: Enhancing Table Reasoning with Iterative Row-Wise Traversals
- From Automation to Autonomy: A Survey on Large Language Models in Scientific Discovery
- Table-R1: Region-based Reinforcement Learning for Table Understanding
- HALO: Hierarchical Autonomous Logic-Oriented Orchestration for Multi-Agent LLM Systems
- GeoGrid-Bench: Can Foundation Models Understand Multimodal Gridded Geo-Spatial Data?
- EcoTable: Cost-effective Table Integration in Data Lakes for Natural Language Queries
- Understanding LLM Scientific Reasoning through Promptings and Model's Explanation on the Answers
- 100 Days After DeepSeek-R1: A Survey on Replication Studies and More Directions for Reasoning Language Models
- Tabular LLMs for Interpretable Few-Shot Alzheimer's Disease Prediction with Multimodal Biomedical Data
- Safety in Large Reasoning Models: A Survey
- EdgeLM: Edge Demonstrations for Language Models' Table Understanding
- Science Hierarchography: Hierarchical Organization of Science Literature
Discussions
Related