Papperoni

2025

Why Language Models Hallucinate

Edwin Zhang

Open PDF Google Scholar

citations

Cite Score

1

AI summary

This paper demystifies hallucinations in language models, attributing them to training/evaluation procedures that reward guessing over uncertainty acknowledgment. It connects generative errors to binary classification errors and proposes modifying existing benchmarks to penalize uncertainty less, aiming for more trustworthy AI systems.

Main Contributions

Identifies that language models hallucinate due to training and evaluation procedures rewarding guessing over acknowledging uncertainty.
Analyzes hallucinations as errors in binary classification, showing they arise from natural statistical pressures in pretrained language models.
Argues that the persistence of hallucinations is due to most evaluations grading models as good test-takers, where guessing when uncertain improves performance.
Proposes a socio-technical mitigation: modifying the scoring of existing misaligned benchmarks to stop penalizing uncertain responses, rather than introducing new hallucination evaluations.
Demonstrates how a novel connection between supervised and unsupervised learning demystifies the origin of hallucinations, even with IDK in training data.

Abstract

Like students facing hard exam questions, large language models sometimes guess when uncertain, producing plausible yet incorrect statements instead of admitting uncertainty. Such "hallucinations" persist even in state-of-the-art systems and undermine trust. We argue that language models hallucinate because the training and evaluation procedures reward guessing over acknowledging uncertainty, and we analyze the statistical causes of hallucinations in the modern training pipeline. Hallucinations need not be mysterious—they originate simply as errors in binary classification. If incorrect statements cannot be distinguished from facts, then hallucinations in pretrained language models will arise through natural statistical pressures. We then argue that hallucinations persist due to the way most evaluations are graded—language models are optimized to be good test-takers, and guessing when uncertain improves test performance. This "epidemic" of penalizing uncertain responses can only be addressed through a socio-technical mitigation: modifying the scoring of existing benchmarks that are misaligned but dominate leaderboards, rather than introducing additional hallucination evaluations. This change may steer the field toward more trustworthy AI systems.

Citation Graph

Loading graph...

References [94]

Sort:

Filter:

[1]Training Language Models to Follow Instructions With Human Feedback

Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, C. Wainwright, Pamela Mishkin, Chiyuan Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul F. Christiano, Jan Leike, Ryan Lowe - 2022

11 papers in library cite

No new research here. Only true contribution is scaling RLHF to GPT 3.

[2]Deepseek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Deepseek Ai - 2025

2 papers in library cite

Finally the RLVR paper!

[3]Direct Preference Optimization: Your Language Model Is Secretly a Reward Model

Rafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D. Manning, Chelsea Finn - 2023

3 papers in library cite

Very nice idea and probably very revolutionary but paper is a bit boring.

[4]Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

P. Lewis, Ethan Perez, A. Piktus, F. Petroni, V. Karpukhin, N. Goyal, H. Kuttler, Martha Lewis, W. T. Yih, Tim Rocktaschel, Sebastian Riedel, K. Douwe - 2020

5 papers in library cite

It coined the term "RAG" but really it was a very different concept.

[5]Measuring Mathematical Problem Solving With the MATH Dataset

Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, Jacob Steinhardt - 2021

8 papers in library cite

Good idea in creating the dataset but it's just a regular and uninteresting dataset description paper

[6]Constitutional AI: Harmlessness From AI Feedback

Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron Mckinnon, C. C. Chen, Catherine Olsson, Christopher Olah, Danny Hernandez, Dawn Drain, Deep Ganguli, Dustin Li, Eli Tran Johnson, Ethan Perez, Jamie Kerr, Jared Mueller, Jeffrey Ladish, Joshua Landau, Kamal Ndousse, Kamile Lukosuite, Liane Lovitt, Michael Sellitto, Nelson Elhage, Nicholas Schiefer, Noemi Mercado, Nova Dassarma, Robert Lasenby, Robin Larson, Sam Ringer, Scott Johnston, Shauna Kravec, Sheer El Showk, Stanislav Fort, Tamera Lanham, Timothy Telleen Lawton, Tom Conerly, Tom Henighan, Tristan Hume, Samuel R. Bowman, Zac Hatfield Dodds, Benjamin Mann, Dario Amodei, Nicholas Joseph, Sam McCandlish, Tom B. Brown, Jared Kaplan - 2022

2 papers in library cite

Nice idea, but it's just a new spin on RLHF. Also, too long.

[7]Beyond the Imitation Game: Quantifying and Extrapolating the Capabilities of Language Models

Aarohi Srivastava, Abhinav Rastogi, Abhishek Rao, Abu Awal Md Shoeb, Abubakar Abid, Adam Fisch, Adam R. Brown, Adam Santoro, Aman Gupta, Adria Garriga Alonso - 2022

4 papers in library cite

Nice initiative but too much focus on models and performance rather than the bench itself. Also, saturated right after

[8]WebGPT: Browser-Assisted Question-Answering With Human Feedback

Reiichiro Nakano, Jacob Hilton, Suchir Balaji, Jeffrey Wu, Long Ouyang, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, Xu Jiang, Karl Cobbe, Tyna Eloundou, Gretchen Krueger, Kevin Button, Matthew Knight, Benjamin Chess, John Schulman - 2021

7 papers in library cite

TBH the nicest thing about this is the idea of using a text-based web browser. Other than that, just another application of RLHF + PPO.

[9]Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them

Mirac Suzgun, Nathan Scales, Nathanael Scharli, Sebastian Gehrmann, Yi Tay, Hyung Won Chung, Aakanksha Chowdhery, Quoc V. Le, Ed H. Chi, Denny Zhou, Jason Wei - 2022

4 papers in library cite

Good read and good analysis but no interesting research. They just distilled the benchmark and applied CoT. Good for showing that it works but not exciting.

[10]GPT-4 Technical Report

Openai - 2023

6 papers in library cite

GPT 4

[11]TruthfulQA: Measuring How Models Mimic Human Falsehoods

Stephen Lin, Jacob Hilton, Owain Evans - 2022

4 papers in library cite

[12]GPQA: A Graduate-Level Google-Proof Q&A Benchmark

D. Rein, B. L. Hou, A. C. Stickland, J. Petty, R. Y. Pang, J. Dirani, J. Michael, Samuel R. Bowman - 2024

3 papers in library cite

[13]On Faithfulness and Factuality in Abstractive Summarization

J. Maynez, Shashi Narayan, B. Bohnet, R. Mcdonald - 2020

6 papers in library cite

[14]SWE-bench: Can Language Models Resolve Real-World GitHub Issues?

C. E. Jimenez, Jihan Yang, A. Wettig, S. Yao, K. Pei, O. Press, K. R. Narasimhan - 2024

2 papers in library cite

[15]MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark

Yuzhi Wang, X. Ma, G. Zhang, Yuan Ni, A. Chandra, S. Guo, W. Ren, A. Arulraj, X. He, Zhejun Jiang, Tao Li, M. Ku, K. Wang, A. Zhuang, R. Fan, Xiang Yue, Weizhu Chen - 2024

3 papers in library cite

[16]Retrieval Augmentation Reduces Hallucination in Conversation

K. Shuster, S. Poff, Mark Chen, Douwe Kiela, Jason Weston - 2021

3 papers in library cite

[17]Instruction-Following Evaluation for Large Language Models

Jingren Zhou, T. Lu, Swaroop Mishra, S. Brahma, S. Basu, Y. Luan, Denny Zhou, L. Hou - 2023

2 papers in library cite

[18]Dataset Shift in Machine Learning

J. Q. Candela, M. Sugiyama, A. Schwaighofer, N. D. Lawrence - 2009

4 papers in library cite

Book, 250 pages

[19]Language Models (Mostly) Know What They Know

Saurav Kadavath, Tom Conerly, Amanda Askell, Tom Henighan, Dawn Drain, Ethan Perez, Nicholas Schiefer, Zac Hatfield Dodds, Nova Dassarma, Eli Tran Johnson, Scott Johnston, Sheer El Showk, Andy Jones, Nelson Elhage, Tristan Hume, Anna Chen, Yuntao Bai, S. Bowman, Stanislav Fort, Deep Ganguli, Danny Hernandez, J. Jacobson, Jackson Kernion, Shauna Kravec, Liane Lovitt, Kamal Ndousse, Catherine Olsson, Sam Ringer, Dario Amodei, Tom B. Brown, Jack Clark, Nicholas Joseph, Benjamin Mann, Sam McCandlish, Christopher Olah, Jared Kaplan - 2022

3 papers in library cite

[20]Humanity's Last Exam

L. Phan, A. Gatti, Z. Han, N. Li, Jiaxi Hu, Haowei Zhang, Chen Bo Calvin Zhang, M. Shaaban, J. Ling, Sherry Shi, M. Choi, A. Agrawal, A. Chopra, A. Khoja, R. Kim, R. Ren, J. Hausenloy, Oliver Zhang, Mantas Mazeika, D. Dodonov, T. N. Nguyen, Jaehoon Lee, A. 1. Others - 2025

1 paper in library cites

[21]Artificial Intelligence: A Modern Approach

S. Russell, P. Norvig - 1995

4 papers in library cite

[22]BBQ: A Hand-Built Bias Benchmark for Question Answering

Alicia Parrish, Anna Chen, Nikita Nangia, Vishakh Padmakumar, Jason Phang, Jana Thompson, Phu Mon Htut, S. Bowman - 2022

4 papers in library cite

[23]On the Uniform Convergence of Relative Frequencies of Events to Their Probabilities

V. N. Vapnik, A. Y. Chervonenkis - 1971

3 papers in library cite

[24]The Language Model Evaluation Harness

Leo Gao, J. Tow, B. Abbasi, Stella Biderman, S. Black, A. Dipofi, C. Foster, L. Golding, J. Hsu, A. L. Noac'h, H. Li, Kyle Mcdonell, Niklas Muennighoff, C. Ociepa, Jason Phang, Laria Reynolds, H. Schoelkopf, A. Skowron, L. Sutawika, Eric Tang, A. Thite, B. Wang, K. Wang, Andy Zou - 2024

3 papers in library cite

[25]Holistic Evaluation of Language Models

Percy Liang, R. Bommasani, Teddy Lee, D. Tsipras, Dilara Soylu, Michihiro Yasunaga, Y. Z. Zhang, D. Narayanan, Yonghui Wu, A. Kumar, B. Newman, B. Yuan, B. Yan, Chiyuan Zhang, C. Cosgrove, Christopher D. Manning, C. Re, D. A. Navas, D. A. Hudson, E. Zelikman, Esin Durmus, Faisal Ladhak, Frieda Rong, H. Ren, H. Yao, J. Wang, K. Santhanam, L. Orr, L. Zheng, M. Yuksekgonul, Mirac Suzgun, N. Kim, N. Guha, N. S. Chatterji, Omar Khattab, P. Henderson, Q. Huang, R. A. Chi, S. M. Xie, S. Santurkar, Surya Ganguli, Tatsunori Hashimoto, T. Icard, Tong Zhang, V. Chaudhary, Wenyi Wang, Xiang Lisa Li, Y. Mai, Y. Z. Zhang, Y. Koreeda - 2023

2 papers in library cite

[26]On the Method of Bounded Differences

C. Mcdiarmid - 1989

2 papers in library cite

[27]Open LLM Leaderboard v2

C. Fourrier, N. Habib, A. Lozovskaya, K. Szafer, Thomas Wolf - 2024

2 papers in library cite

[28]The Population Frequencies of Species and the Estimation of Population Parameters

I. J. Good - 1953

2 papers in library cite

[29]A Few Useful Things to Know About Machine Learning

P. Domingos - 2012

1 paper in library cites

[30]A Unifying View on Dataset Shift in Classification

J. G. M. Torres, T. Raeder, R. A. Rodriguez, N. V. Chawla, F. Herrera - 2012

1 paper in library cites

[31]Actively Avoiding Nonsense in Generative Models

S. Hanneke, A. T. Kalai, G. Kamath, C. Tzamos - 2018

1 paper in library cites

[32]AI Chatbots Will Never Stop Hallucinating

L. Leffer - 2024

1 paper in library cites

[33]AI Hallucinations Can't Be Stopped but These Techniques Can Limit Their Damage

N. Jones - 2025

1 paper in library cites

[34]An Introduction to Computational Learning Theory

M. J. Kearns, U. V. Vazirani - 1994

1 paper in library cites

[35]Answer, Refuse, or Guess? Investigating Risk-Aware Decision Making in Language Models

C. K. Wu, Z. R. Tam, Chin Yew Lin, Y. N. Chen, H. Y. Lee - 2025

1 paper in library cites

[36]Artificial Intelligence Index Report 2025

N. Maslej, L. Fattorini, R. Perrault, Y. Gil, V. Parli, N. Kariuki, E. Capstick, A. Reuel, E. Brynjolfsson, J. Etchemendy, K. Ligett, T. Lyons, J. Manyika, J. C. Niebles, Y. Shoham, R. Wald, T. Walsh, A. Hamrah, L. Santarlasci, J. B. Lotufo, A. Rome, A. Shi, S. Oak - 2025

1 paper in library cites

[37]Beyond Binary Rewards: Training LMs to Reason About Their Uncertainty

M. Damani, I. Puri, S. Slocum, I. Shenfeld, L. Choshen, Yoon Kim, Jacob Andreas - 2025

1 paper in library cites

[38]Calibrated Language Models Must Hallucinate

A. T. Kalai, S. S. Vempala - 2024

1 paper in library cites

[39]Concentration Inequalities for the Missing Mass and for histogram Rule Error

D. Mcallester, L. Ortiz - 2003

1 paper in library cites

[40]Dario Amodei: Anthropic CEO on Claude, AGI & the Future of AI & Humanity

Dario Amodei, L. Fridman - 2024

1 paper in library cites

[41]Detecting Hallucinations in Large Language Models Using Semantic Entropy

S. Farquhar, J. Kossen, L. Kuhn, Yarin Gal - 2024

1 paper in library cites

[42]Do Language Models Know When They're Hallucinating References?

A. Agrawal, Mirac Suzgun, L. Mackey, A. Kalai - 2024

1 paper in library cites

[43]Do Large Language Models Know What They Don't Know?

Z. Yin, Q. Sun, Q. Guo, Jeffrey Wu, X. Qiu, X. Huang - 2023

1 paper in library cites

[44]Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?

Z. Gekhman, G. Yona, R. Aharoni, M. Eyal, A. Feder, R. Reichart, J. Herzig - 2024

1 paper in library cites

[45]Durably Reducing Conspiracy Beliefs Through Dialogues With AI

T. H. Costello, G. Pennycook, D. G. Rand - 2024

1 paper in library cites

[46]Factuality Enhanced Language Models for Open-Ended Text Generation

Nayeon Lee, W. Ping, P. Xu, M. Patwary, Pascale Fung, M. Shoeybi, Bryan Catanzaro - 2022

1 paper in library cites

[47]Fine-Tuning Language Models for Factuality

K. Tian, Eric Mitchell, H. Yao, Christopher D. Manning, Chelsea Finn - 2024

1 paper in library cites

[48]Foundations of Cryptography: Volume 1, Basic Tools

O. Goldreich - 2001

1 paper in library cites

[49]Gemini 2.5 Pro Model Card

G. Deepmind - 2025

1 paper in library cites

[50]GPT-5 System Card

Openai - 2025

1 paper in library cites

[51]GPT4All: Training an Assistant-Style Chatbot With Large-Scale Data Distillation From GPT-3.5-Turbo

Y. Anand, Z. Nussbaum, B. Duderstadt, B. Schmidt, A. Mulyar - 2023

1 paper in library cites

[52]GraphArena: Evaluating and Exploring Large Language Models on Graph Computation

Jie Tang, Q. Zhang, Yiwei Li, N. Chen, Jeffrey Li - 2025

1 paper in library cites

[53]Hallucination Is Inevitable: An Innate Limitation of Large Language Models

Zhiwei Xu, Shantanu Jain, M. Kankanhalli - 2024

1 paper in library cites

[54]Hallucination Mitigation for Retrieval-Augmented Large Language Models: A Review

Wenxuan Zhang, J. Zhang - 2025

1 paper in library cites

[55]Hallucination, Monofacts, and Miscalibration: An Empirical Investigation

M. M. Miao, M. Kearns - 2025

1 paper in library cites

[56]HalluLens: LLM Hallucination Benchmark

Y. Bang, Z. Ji, A. Schelten, A. Hartshorn, T. Fowler, Chiyuan Zhang, N. Cancedda, Pascale Fung - 2025

1 paper in library cites

[57]Head-to-Tail: How Knowledgeable Are Large Language Models (LLM)? AKA Will LLMs Replace Knowledge Graphs?

Ke Sun, Y. E. Xu, H. Zha, Yibo Liu, X. L. Dong - 2023

1 paper in library cites

[58]HELM Capabilities: Evaluating LMs Capability by Capability

Jiacheng Xu, Y. Mai, Percy Liang - 2025

1 paper in library cites

[59]Hijacking Context in Large Multi-Modal Models

J. Jeong - 2024

1 paper in library cites

[60]How Language Model Hallucinations Can Snowball

Mingchuan Zhang, O. Press, W. Merrill, A. Liu, Noah A. Smith - 2023

1 paper in library cites

[61]How Pre-Trained Language Models Capture Factual Knowledge? A Causal-Inspired Analysis

Shanda Li, Xiang Lisa Li, L. Shang, Zhikang Dong, C. Sun, Bing Liu, Z. Ji, Xu Jiang, Qian Liu - 2022

1 paper in library cites

[62]How to Do Things With Words

J. L. Austin - 1962

1 paper in library cites

[63]Improving Mathematical Reasoning With Process Supervision

Openai - 2023

1 paper in library cites

[64]Introducing Deep Research

Openai - 2025

1 paper in library cites

[65]Introducing GPT-4.1 in the API

Openai - 2025

1 paper in library cites

[66]Investigating Memorization of Conspiracy Theories in Text Generation

S. Levy, M. Saxon, W. Y. Wang - 2021

1 paper in library cites

[67]Language Generation in the Limit

J. Kleinberg, S. Mullainathan - 2024

1 paper in library cites

[68]Learning Reductions That Really Work

A. Beygelzimer, H. D. Iii, John Langford, P. Mineiro - 2016

1 paper in library cites

[69]Learning to Reason With LLMs

Openai - 2024

1 paper in library cites

[70]Logic and Conversation

H. P. Grice - 1975

1 paper in library cites

[71]Medical Large Language Models Are Vulnerable to Data-Poisoning Attacks

D. A. Alber, Zhilin Yang, A. Alyakin, E. Yang, S. Rai, A. A. Valliani - 2025

1 paper in library cites

[72]MuSR: Testing the Limits of Chain-of-Thought With Multistep Soft Reasoning

Z. Sprague, X. Ye, K. Bostrom, S. Chaudhuri, G. Durrett - 2024

1 paper in library cites

[73]NPHardEval: Dynamic Benchmark on Reasoning Ability of Large Language Models via Complexity Classes

L. Fan, W. Hua, Lei Li, H. Ling, Y. Z. Zhang - 2024

1 paper in library cites

[74]Omni-MATH: A Universal Olympiad Level Mathematic Benchmark for Large Language Models

B. Gao, Francis Song, Zhilin Yang, Zhipeng Cai, Y. Miao, Q. Dong, Lei Li, C. Ma, L. C. Chen, Runxin Xu, Z. Tang, B. Wang, D. Zan, S. Quan, G. Zhang, L. Sha, Y. Z. Zhang, Xiang Ren, T. Liu, B. Chang - 2024

1 paper in library cites

[75]On the Convergence Rate of Good-Turing Estimators

D. A. Mcallester, R. E. Schapire - 2000

1 paper in library cites

[76]On the Limits of Language Generation: Trade-Offs Between Hallucination and Mode-Collapse

A. Kalavasis, A. Mehrotra, G. Velegkas - 2025

1 paper in library cites

[77]Open-LLM-Leaderboard: From Multi-Choice to Open-Style Questions for LLMS Evaluation, Benchmark, and Arena

A. Myrzakhan, S. M. Bsharat, Z. Shen - 2024

1 paper in library cites

[78]OpenAI 03 and 04-Mini System Card

Openai - 2025

1 paper in library cites

[79]Pragmatics in the Era of Large Language Models: A Survey on Datasets, Evaluation, Opportunities and Challenges

B. Ma, Yiwei Li, W. Zhou, Z. Gong, Y. J. Liu, K. Jasinskaja, A. Friedrich, J. Hirschberg, F. Kreuter, Barbara Plank - 2025

1 paper in library cites

[80]Probabilistic and on-Line Methods in Machine Learning

A. Kalai - 2001

1 paper in library cites

[81]Reducing Conversational Agents' Overconfidence Through Linguistic Calibration

S. J. Mielke, A. Szlam, E. Dinan, Y. Lan Boureau - 2022

1 paper in library cites

[82]Scheduled Sampling for Sequence Prediction With Recurrent Neural Networks

Samy Bengio, Oriol Vinyals, Navdeep Jaitly, Noam Shazeer - 2015

1 paper in library cites

[83]SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models

P. Manakul, A. Liusie, M. Gales - 2023

1 paper in library cites

[84]Simple Synthetic Data Reduces Sycophancy in Large Language Models

Jason Wei, Dong Huang, Y. Lu, Denny Zhou, Quoc V. Le - 2023

1 paper in library cites

[85]Survey of Hallucination in Natural Language Generation

Z. Ji, Nayeon Lee, R. Frieske, Tao Yu, D. Su, Yiheng Xu, E. Ishii, Y. Bang, Deli Chen, W. Dai, H. S. Chan, Andrea Madotto, Pascale Fung - 2023

1 paper in library cites

[86]Teaching Models to Express Their Uncertainty in Words

Stephen Lin, Jacob Hilton, Owain Evans - 2022

1 paper in library cites

[87]The Hallucinations Leaderboard - An Open Effort to Measure Hallucinations in Large Language Models

G. Hong, A. P. Gema, R. Saxena, X. Du, P. Nie, Y. Zhao, Laura Perez-Beltrachini, M. Ryabinin, X. He, C. Fourrier, P. Minervini - 2024

1 paper in library cites

[88]The Reversal Curse: LLMs Trained on "A Is B" Fail to Learn "B Is A"

L. Berglund, Meg Tong, M. Kaufmann, M. Balesni, A. C. Stickland, Tomasz Korbak, Owain Evans - 2024

1 paper in library cites

[89]The Well-Calibrated Bayesian

A. P. Dawid - 1982

1 paper in library cites

[90]Toward Efficient Agnostic Learning

M. J. Kearns, R. E. Schapire, L. M. Sellie - 1994

1 paper in library cites

[91]Verify When Uncertain: Beyond Self-Consistency in Black Box Hallucination Detection

Y. Xue, K. Greenewald, Y. Mroueh, B. Mirzasoleiman - 2025

1 paper in library cites

[92]Vicuna: An Open-Source Chatbot Impressing GPT-4 With 90%* ChatGPT Quality

Wei-Lin Chiang, Zhiyuan Li, Zongyu Lin, Y. Sheng, Ziyi Wu, Haowei Zhang, L. Zheng, S. Zhuang, Y. Zhuang, Joseph E. Gonzalez, Ion Stoica, E. P. Xing - 2023

1 paper in library cites

[93]Why and How LLMs Hallucinate: Connecting the Dots With Subsequence Associations

Y. S. Sun, Y. Gai, L. C. Chen, A. Ravichander, Yejin Choi, Dawn Song - 2025

1 paper in library cites

[94]WildBench: Benchmarking LLMs With Challenging Tasks From Real Users in the Wild

Bill Yuchen Lin, Y. Deng, K. Chandu, A. Ravichander, V. Pyatkin, N. Dziri, R. L. Bras, Yejin Choi - 2025

1 paper in library cites

Cited by

0

papers in your library

Cites

20

papers in your library

Read

on October 5, 2025

Very nice explanation on why LLMs hallucinate (but the problem isn't solved yet), but more still, a VERY nice push to change evaluation methods.

Tags

Paper Aliases

No aliases