Pith. sign in

REVIEW 5 cited by

Sources of Hallucination by Large Language Models on Inference Tasks

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

arxiv 2305.14552 v2 pith:KLMBXVYK submitted 2023-05-23 cs.CL cs.AI

classification cs.CLcs.AI
keywords datalanguagellmsmodelsbehaviorbiaseshallucinationhypothesis
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Large Language Models (LLMs) are claimed to be capable of Natural Language Inference (NLI), necessary for applied tasks like question answering and summarization. We present a series of behavioral studies on several LLM families (LLaMA, GPT-3.5, and PaLM) which probe their behavior using controlled experiments. We establish two biases originating from pretraining which predict much of their behavior, and show that these are major sources of hallucination in generative LLMs. First, memorization at the level of sentences: we show that, regardless of the premise, models falsely label NLI test samples as entailing when the hypothesis is attested in training data, and that entities are used as ``indices'' to access the memorized data. Second, statistical patterns of usage learned at the level of corpora: we further show a similar effect when the premise predicate is less frequent than that of the hypothesis in the training data, a bias following from previous studies. We demonstrate that LLMs perform significantly worse on NLI test samples which do not conform to these biases than those which do, and we offer these as valuable controls for future LLM evaluation.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 20 citations worldwide. Full citation record

  1. Synthetic Heuristic Evaluation: A Comparison between AI- and Human-Powered Usability Evaluation

    cs.HC 2025-07 reject novelty 6.0 of 10

    An LLM prompted to conduct heuristic evaluation reported more usability issues on two apps than five human experts, but the ground truth included the LLM's own findings.

  2. Conformal Prediction Beyond the Seen: A Missing Mass Perspective for Uncertainty Quantification in Generative Models

    cs.LG 2025-06 reject novelty 6.0 of 10

    CPQ builds conformal prediction sets for black-box LLMs by stopping queries when the estimated missing-mass derivative is small and thresholding a Good-Turing based score, with a fallback EE label for unseen correct answers.

  3. Are Large Language Models Good Temporal Graph Learners?

    cs.CL 2025-06 conditional novelty 6.0 of 10

    TGTalker prompts LLMs with the most recent edges and neighbor history of a temporal graph, achieving link prediction accuracy competitive with specialized temporal graph neural networks on five real-world datasets.

  4. Small Data Explainer -- The impact of small data methods in everyday life

    cs.CY 2025-07 conditional novelty 3.0 of 10

    A review and explainer that frames small data methods through the recurring challenges of similarity, transfer, and uncertainty and maps them to application areas and technical approaches.

  5. The Scales of Justitia: A Comprehensive Survey on Safety Evaluation of LLMs

    cs.CL 2025-06 conditional novelty 3.0 of 10

    A structured survey of LLM safety evaluation that proposes a why/what/where/how taxonomy and catalogs metrics, datasets, benchmarks, evaluators, and frameworks.

Pith tools