English
Related papers

Related papers: The Hydra Effect: Emergent Self-repair in Language…

200 papers

A recent study (Kuribayashi et al., 2025) has shown that human sentence processing behavior, typically measured on syntactically unchallenging constructions, can be effectively modeled using surprisal from early layers of large language…

Computation and Language · Computer Science 2026-04-21 Tatsuki Kuribayashi , Alex Warstadt , Yohei Oseki , Ethan Gotlieb Wilcox

We show that language models hallucinate not because they fail to detect uncertainty, but because of a failure to integrate it into output generation. Across architectures, uncertain inputs are reliably identified, occupying…

Artificial Intelligence · Computer Science 2026-03-17 Valeria Ruscio , Keiran Thompson

Large Language Models (LLMs) are increasingly deployed in high-stakes financial domains, yet they suffer from specific, reproducible hallucinations when performing arithmetic operations. Current mitigation strategies often treat the model…

Computation and Language · Computer Science 2025-12-01 Soham Mirajkar

Large language models are becoming the go-to solution for the ever-growing number of tasks. However, with growing capacity, models are prone to rely on spurious correlations stemming from biases and stereotypes present in the training data.…

Computation and Language · Computer Science 2024-05-30 Tomasz Limisiewicz , David Mareček , Tomáš Musil

Large language models are increasingly used as behavioral simulators, but it remains unclear when their outputs reflect human-like cognitive mechanisms rather than prompt-sensitive surface patterns. We study this question through the…

Artificial Intelligence · Computer Science 2026-05-26 Ciarán Walsh , Emilio Barkett

Large Language Models are growing in size, and we expect them to continue to do so, as larger models train quicker. However, this increase in size will severely impact inference costs. Therefore model compression is important, to retain the…

Machine Learning · Computer Science 2024-04-10 Georgy Tyukin

Diffusion language models (DLMs) have recently emerged as competitive alternatives to autoregressive (AR) language models, yet differences in their activation dynamics remain poorly understood. We characterize these dynamics in LLaDA-8B and…

Machine Learning · Computer Science 2026-05-12 Alexander Conzelmann , Albert Catalan-Tatjer , Shiwei Liu

Large language models improve at math after instruction tuning, reinforcement learning, or knowledge distillation. We ask whether these gains come from major changes in the transformer layers or from smaller adjustments that keep the…

Machine Learning · Computer Science 2025-11-06 Aadim Nepal , Safal Shrestha , Anubhav Shrestha , Minwu Kim , Jalal Naghiyev , Ravid Shwartz-Ziv , Keith Ross

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…

Computation and Language · Computer Science 2025-09-08 Adam Tauman Kalai , Ofir Nachum , Santosh S. Vempala , Edwin Zhang

We ask whether multilingual language models trained on unbalanced, English-dominated corpora use English as an internal pivot language -- a question of key importance for understanding how language models function and the origins of…

Computation and Language · Computer Science 2024-06-12 Chris Wendler , Veniamin Veselovsky , Giovanni Monea , Robert West

Language is not only a tool for communication but also a medium for human cognition and reasoning. If, as linguistic relativity suggests, the structure of language shapes cognitive patterns, then large language models (LLMs) trained on…

Computation and Language · Computer Science 2025-06-23 Chenxi Wang , Yixuan Zhang , Lang Gao , Zixiang Xu , Zirui Song , Yanbo Wang , Xiuying Chen

Large language model (LLM)-driven AI systems may exhibit an inference failure mode we term `neural howlround,' a self-reinforcing cognitive loop where certain highly weighted inputs become dominant, leading to entrenched response patterns…

Computation and Language · Computer Science 2025-04-14 Seth Drake

Mathematical reasoning is an increasingly important indicator of large language model (LLM) capabilities, yet we lack understanding of how LLMs process even simple mathematical tasks. To address this, we reverse engineer how three mid-sized…

Artificial Intelligence · Computer Science 2025-02-04 Subhash Kantamneni , Max Tegmark

Hallucination is often regarded as a major impediment for using large language models (LLMs), especially for knowledge-intensive tasks. Even when the training corpus consists solely of true statements, language models still generate…

Computation and Language · Computer Science 2024-07-12 Yuji Zhang , Sha Li , Jiateng Liu , Pengfei Yu , Yi R. Fung , Jing Li , Manling Li , Heng Ji

Attention mechanisms, especially self-attention, have played an increasingly important role in deep feature representation for visual tasks. Self-attention updates the feature at each position by computing a weighted sum of features using…

Computer Vision and Pattern Recognition · Computer Science 2021-06-01 Meng-Hao Guo , Zheng-Ning Liu , Tai-Jiang Mu , Shi-Min Hu

Hallucinations in vision-language models (VLMs) hinder reliability and real-world applicability, usually stemming from distribution shifts between pretraining data and test samples. Existing solutions, such as retraining or fine-tuning on…

Multimedia · Computer Science 2025-06-10 Fei Zhao , Chengcui Zhang , Runlin Zhang , Tianyang Wang , Xi Li

In the quest to give a formal compositional semantics to natural languages, semanticists have started turning their attention to phenomena that have been also considered as parts of pragmatics (e.g., discourse anaphora and presupposition…

Computation and Language · Computer Science 2015-06-19 Jiri Marsik , Maxime Amblard

Large vision-language models (VLMs) are highly capable, yet often hallucinate by favoring textual prompts over visual evidence. We study this failure mode in a controlled object-counting setting, where the prompt overstates the number of…

Computer Vision and Pattern Recognition · Computer Science 2026-04-20 William Rudman , Michal Golovanevsky , Dana Arad , Yonatan Belinkov , Ritambhara Singh , Carsten Eickhoff , Kyle Mahowald

Language models continue to hallucinate despite increases in parameters, compute, and data. We propose neural diversity -- decorrelated parallel representations -- as a principled mechanism that reduces hallucination rates at fixed…

Computation and Language · Computer Science 2025-12-11 Kushal Chakrabarti , Nirmal Balachundhar

This paper analyzes the predictions of image captioning models with attention mechanisms beyond visualizing the attention itself. We develop variants of layer-wise relevance propagation (LRP) and gradient-based explanation methods, tailored…

Computer Vision and Pattern Recognition · Computer Science 2021-08-03 Jiamei Sun , Sebastian Lapuschkin , Wojciech Samek , Alexander Binder