TECP: Token-Entropy Conformal Prediction for LLMs

2025/08/30 by Bing Xu, Xu, Beining, Y. Q. Lu +1
Engineering · #Advancements in Photolithography Techniques #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · pdf · doi:10.48550/arxiv.2509.00461

openalex publication_date 2025/08/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Uncertainty quantification (UQ) for open-ended language generation remains a critical yet underexplored challenge, especially under black-box constraints where internal model signals are inaccessible. In this paper, we introduce Token-Entropy Conformal Prediction (TECP), a novel framework that leverages token-level entropy as a logit-free, reference-free uncertainty measure and integrates it into a split conformal prediction (CP) pipeline to construct prediction sets with formal coverage guarantees. Unlike existing approaches that rely on semantic consistency heuristics or white-box features, TECP directly estimates epistemic uncertainty from the token entropy structure of sampled generations and calibrates uncertainty thresholds via CP quantiles to ensure provable error control. Empirical evaluations across six large language models and two benchmarks (CoQA and TriviaQA) demonstrate that TECP consistently achieves reliable coverage and compact prediction sets, outperforming prior self-consistency-based UQ methods. Our method provides a principled and efficient solution for trustworthy generation in black-box LLM settings.

Citations

• MV-Debate: Multi-view Agent Debate with Dynamic Reflection Gating for Multimodal Harmful Content Detection in Social Media
• COIN: Uncertainty-Guarding Selective Question Answering for Foundation Models with Provable Risk Guarantees
• ASCD: Attention-Steerable Contrastive Decoding for Reducing Hallucination in MLLM
• Backdoor Cleaning without External Guidance in MLLM Fine-tuning
• Conformal Prediction with Cellwise Outliers: A Detect-then-Impute Approach
• Conformal Prediction as Bayesian Quadrature
• Knowledge Graphs, Large Language Models, and Hallucinations: An NLP Perspective
• Sample then Identify: A General Framework for Risk Control and Assessment in Multimodal Large Language Models
• ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees
• CoMT: Chain-of-Medical-Thought Reduces Hallucination in Medical Report Generation
• Detecting and Evaluating Medical Hallucinations in Large Vision Language Models
• PediatricsGPT: Large Language Models as Chinese Medical Assistants for Pediatric Applications
• Can LLMs' Tuning Methods Work in Medical Multimodal Domain?
• Word-Sequence Entropy: Towards Uncertainty Estimation in Free-Form Medical Question Answering Applications and Beyond
• SPOT! Revisiting Video-Language Models for Event Understanding
• A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions
• Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of Free-Form Large Language Models
• Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation
• Language Models (Mostly) Know What They Know
• Conformal prediction beyond exchangeability
• TruthfulQA: Measuring How Models Mimic Human Falsehoods

Related

• ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees
• COPU: Conformal Prediction for Uncertainty Quantification in Natural Language Generation
• API Is Enough: Conformal Prediction for Large Language Models Without Logit-Access
• Addressing Uncertainty in LLMs to Enhance Reliability in Generative AI
• Grammars of Formal Uncertainty: When to Trust LLMs in Automated Reasoning Tasks