paper-with-me

Papers

Incoherent Probability Judgments in Large Language Models

2024-01-30 · Jian-Qiao Zhu, Thomas L. Griffiths

Autoregressive Large Language Models (LLMs) trained for next-word prediction have demonstrated remarkable proficiency at producing coherent text. But are they equally adept at forming coherent probability judgments? We use probabilistic identities and repeated judgments to assess the coherence of probability judgments made by LLMs. Our results show that the judgments produced by these models are often incoherent, displaying human-like systematic deviations from the rules of probability theory. Moreover, when prompted to judge the same event, the mean-variance relationship of probability judgments produced by LLMs shows an inverted-U-shaped like that seen in humans. We propose that these deviations from rationality can be explained by linking autoregressive LLMs to implicit Bayesian inference and drawing parallels with the Bayesian Sampler model of human probability judgments.

📄 PDF Abstract BibTeX arXiv:2401.16646

Code (0)

등록된 구현이 없습니다.

Tasks

Bayesian Inference

Similar Papers 제목 키워드 기반

Safety Measurements for Fine-tuned LLMs Should be Grounded in Capability

2026-06-02 · Krishnapriya Vishnubhotla, Hillary Dawkins, Isar Nejadgholi, Svetlana Kiritchenko arxiv

Adapting foundation large language models to a user's task or preferred style through fine-tuning can result in compromising the model's safety. Previous works examined the effects of fine-tuning on model safety in limit…

When Context Misleads: Surprisal, Energy and Attention Entropy as Metrics of Coherence Illusions in LLMs

2026-06-19 · Ece Takmaz, Nitin Kumar, Li Kloostra, Jakub Dotlacil arxiv

Psycholinguistics studies show that human readers fall for coherence illusions: an incoherent discourse can seem coherent simply because a distractor matches what comes next. We investigate whether Dutch language models …

DEAM: Dialogue Coherence Evaluation using AMR-based Semantic Manipulations

2022-03-18 · ACL 2022 5 · Sarik Ghazarian, Nuan Wen, Aram Galstyan, Nanyun Peng

Automatic evaluation metrics are essential for the rapid development of open-domain dialogue systems as they facilitate hyper-parameter tuning and comparison between models. Although recently proposed trainable conversat…

Abstract Meaning RepresentationCoherence EvaluationDialogue Evaluation

Automatic Detection of Incoherent Speech for Diagnosing Schizophrenia

2018-06-01 · WS 2018 6 · Dan Iter, Jong Yoon, Dan Jurafsky

Schizophrenia is a mental disorder which afflicts an estimated 0.7{\%} of adults world wide. It affects many areas of mental function, often evident from incoherent speech. Diagnosing schizophrenia relies on subjective j…

SentenceSentence EmbeddingSentence-EmbeddingWord Embeddings

If Probable, Then Acceptable? Understanding Conditional Acceptability Judgments in Large Language Models

2025-10-09 · Jasmin Orth, Philipp Mondorf, Barbara Plank arxiv

Conditional acceptability refers to how plausible a conditional statement is perceived to be. It plays an important role in communication and reasoning, as it influences how individuals interpret implications, assess arg…