REVIEW 18 cited by
Mitigating LLM Hallucinations via Conformal Abstention
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Mitigating LLM Hallucinations via Conformal Abstention
read the original abstract
We develop a principled procedure for determining when a large language model (LLM) should abstain from responding (e.g., by saying "I don't know") in a general domain, instead of resorting to possibly "hallucinating" a non-sensical or incorrect answer. Building on earlier approaches that use self-consistency as a more reliable measure of model confidence, we propose using the LLM itself to self-evaluate the similarity between each of its sampled responses for a given query. We then further leverage conformal prediction techniques to develop an abstention procedure that benefits from rigorous theoretical guarantees on the hallucination rate (error rate). Experimentally, our resulting conformal abstention method reliably bounds the hallucination rate on various closed-book, open-domain generative question answering datasets, while also maintaining a significantly less conservative abstention rate on a dataset with long responses (Temporal Sequences) compared to baselines using log-probability scores to quantify uncertainty, while achieveing comparable performance on a dataset with short answers (TriviaQA). To evaluate the experiments automatically, one needs to determine if two responses are equivalent given a question. Following standard practice, we use a thresholded similarity function to determine if two responses match, but also provide a method for calibrating the threshold based on conformal prediction, with theoretical guarantees on the accuracy of the match prediction, which might be of independent interest.
Forward citations
Cited by 18 Pith papers
-
OpenHalDet: A Unified Benchmark for Hallucination Detection across Diverse Generation Scenarios
OpenHalDet creates a standardized benchmark and open codebase for comparing hallucination detectors across diverse LLM generation scenarios and access settings.
-
From Scalars to Tensors: Declared Losses Recover Epistemic Distinctions That Neutrosophic Scalars Cannot Express
Declared losses recover epistemic distinctions collapsed by scalar neutrosophic T/I/F values in LLM evaluations.
-
Two Axes of LLM Abstention: Answer Correctness and Question Answerability
Answer correctness and question answerability are separate axes: ordinary confidence tracks the first while hidden probes track the second, and a factorized dual-threshold policy certifies both risk budgets at higher ...
-
Integrating Local and Global Entropy for Uncertainty Quantification in LLMs
GLU is a single-pass unsupervised uncertainty score for LLMs formed by multiplying global hidden-state geometric entropy with local token entropy, shown to match or beat baselines on three model families and six bench...
-
Empirical Bayes Conformal Prediction for Vision and Language Models
Empirical Bayes conformal prediction converts score variability into r-value nonconformity scores that preserve target coverage while reducing inclusion of high-variance false candidates in image classification, CLIP ...
-
Pause and Reflect: Conformal Aggregation for Chain-of-Thought Reasoning
A conformal procedure for CoT replaces majority voting with weighted aggregation and calibrates abstention to guarantee low confident-error rates, achieving 90.1% selective accuracy on GSM8K by abstaining on under 5% ...
-
PASC: Pipeline-Aware Conformal Prediction with Joint Coverage Guarantees for Multi-Stage NLP and LLM Pipelines
PASC converts multi-stage joint coverage into a single scalar conformal problem on the joint max nonconformity score, delivering finite-sample distribution-free guarantees and higher empirical coverage than Bonferroni...
-
Causal Evidence that Language Models use Confidence to Drive Behavior
Language models deploy multidimensional internal confidence representations and threshold-based policies to control abstention behavior, with causal support from activation steering experiments.
-
Understanding New-Knowledge-Induced Factual Hallucinations in LLMs: Analysis and Interpretation
Fine-tuning on new knowledge induces propagating hallucinations in LLMs by weakening attention to key entities, with mitigation via reintroducing known knowledge during later training stages.
-
Auto-DSM Under the Lens: A Black-Box Evaluation Framework for LLM-Based DSM Generation
A black-box evaluation framework with multi-run metrics reveals that LLM-based DSM generation is sensitive to phrasing, exhibits hallucination, and fails to consistently acknowledge knowledge boundaries.
-
Strategic Decision Support for AI Agents
The paper introduces an optimization framework for AI agents to strategically seek support, proving a threshold policy on support value and providing an online algorithm to control missed-support error without distrib...
-
Conformal Certification of Reasoning Trace Prefixes
CROP applies conformal prediction to certify the longest contiguous prefix of an LLM reasoning trace that is guaranteed to contain no annotated errors under an exchangeability assumption.
-
Position: Uncertainty Quantification in LLMs is Just Unsupervised Clustering
Mainstream UQ for LLMs reduces to unsupervised clustering of internal generation consistency and therefore cannot detect confident hallucinations or provide reliable safety signals.
-
Margin-Adaptive Confidence Ranking for Reliable LLM Judgement
Introduces a margin-adaptive confidence ranking method that learns an estimator from simulated diversity and derives margin-dependent generalization bounds for use in fixed-sequence testing of LLM-human agreement.
-
Learning from AVA: Early Lessons from a Curated and Trustworthy Generative AI for Policy and Development Research
AVA is a specialized GenAI platform for development policy research that provides verifiable syntheses from World Bank reports and is associated with 2.4-3.9 hours of weekly time savings in a large-scale user evaluation.
-
Margin-Adaptive Confidence Ranking for Reliable LLM Judgement
Develops a margin-adaptive learned confidence estimator for LLMs with generalization guarantees to improve agreement rates with human judgments over heuristic baselines.
-
Foundations of Reliable Inference: Reliability-Efficiency Co-Design
A unified framework is developed for co-designing reliability and efficiency to enable efficient reliable inference with trustworthy uncertainty quantification in AI models.
-
Breaking the Chains of Probability: Neutrosophic Logic as a New Framework for Epistemic Uncertainty in Large Language Models
Empirical tests on four GPT models across five uncertainty types found hyper-truth states (T+I+F>1) in 35% of cases, mostly under ethical contradictions and paradoxes.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.