English
Related papers

Related papers: Confidence Regulation Neurons in Language Models

200 papers

Large language models (LLMs) frequently generate hallucinations -- plausible but factually incorrect outputs -- undermining their reliability. While prior work has examined hallucinations from macroscopic perspectives such as training data…

Artificial Intelligence · Computer Science 2025-12-03 Cheng Gao , Huimin Chen , Chaojun Xiao , Zhiyi Chen , Zhiyuan Liu , Maosong Sun

Uncertainty estimation is crucial for evaluating Large Language Models (LLMs), particularly in high-stakes domains where incorrect answers result in significant consequences. Numerous approaches consider this problem, while focusing on a…

Computation and Language · Computer Science 2025-03-04 Petr Sychev , Andrey Goncharov , Daniil Vyazhev , Edvard Khalafyan , Alexey Zaytsev

Large language models (LLMs) generate not only reasoning text, but also token-level confidence trajectories that record how uncertainty evolves during inference. Whether these trajectories are relevant to reasoning correctness remains…

Machine Learning · Computer Science 2026-05-19 Shuo Liu , Ding Liu , Shi-Ju Ran

Large language models (LLMs) have demonstrated remarkable performance across diverse tasks by encoding vast amounts of factual knowledge. However, they are still prone to hallucinations, generating incorrect or misleading information, often…

Computation and Language · Computer Science 2025-11-13 Xiaomin Li , Zhou Yu , Ziji Zhang , Yingying Zhuang , Swair Shah , Narayanan Sadagopan , Anurag Beniwal

Language confusion -- where large language models (LLMs) generate unintended languages against the user's need -- remains a critical challenge, especially for English-centric models. We present the first mechanistic interpretability (MI)…

Computation and Language · Computer Science 2025-09-19 Ercong Nie , Helmut Schmid , Hinrich Schütze

In long-horizon tasks, recent agents based on Large Language Models (LLMs) face a significant challenge that sparse, outcome-based rewards make it difficult to assign credit to intermediate steps. Previous methods mainly focus on creating…

Machine Learning · Computer Science 2025-09-12 Jiawei Wang , Jiacai Liu , Yuqian Fu , Yingru Li , Xintao Wang , Yuan Lin , Yu Yue , Lin Zhang , Yang Wang , Ke Wang

As large language models (LLMs) are increasingly used in high-stakes domains, accurately assessing their confidence is crucial. Humans typically express confidence through epistemic markers (e.g., "fairly confident") instead of numerical…

Computation and Language · Computer Science 2026-04-14 Jiayu Liu , Qing Zong , Weiqi Wang , Yangqiu Song

We analyze a family of large language models in such a lightweight manner that can be done on a single GPU. Specifically, we focus on the OPT family of models ranging from 125m to 66b parameters and rely only on whether an FFN neuron is…

Computation and Language · Computer Science 2023-09-12 Elena Voita , Javier Ferrando , Christoforos Nalmpantis

Recent studies show that LLM hidden states encode reward-related information, such as answer correctness and model confidence. However, existing approaches typically fit black-box probes on the full hidden states, offering little insight…

Computation and Language · Computer Science 2026-05-12 Guowei Xu , Mert Yuksekgonul , James Zou

Decoding strategies play a central role in shaping the reasoning ability of large language models (LLMs). Traditional methods such as greedy decoding and beam search often suffer from error propagation, while sampling-based approaches…

This paper investigates the relationship between large language models' (LLMs) ability to recognize repetitive input patterns and their performance on in-context learning (ICL). In contrast to prior work that has primarily focused on…

Computation and Language · Computer Science 2025-11-12 Nhi Hoai Doan , Tatsuya Hiraoka , Kentaro Inui

Large Language Models (LLMs) suffer significant performance degradation in multi-turn conversations when information is presented incrementally. Given that multi-turn conversations characterize everyday interactions with LLMs, this…

Computation and Language · Computer Science 2025-11-04 Haziq Mohammad Khalid , Athikash Jeyaganthan , Timothy Do , Yicheng Fu , Sean O'Brien , Vasu Sharma , Kevin Zhu

Large Language Models (LLMs) typically have billions of parameters and are thus often difficult to interpret in their operation. In this work, we demonstrate that it is possible to decode neuron weights directly into token probabilities…

Computation and Language · Computer Science 2025-03-03 Harry J Davies

Large Language Models (LLMs) are increasingly attracting attention in various applications. Nonetheless, there is a growing concern as some users attempt to exploit these models for malicious purposes, including the synthesis of controlled…

Artificial Intelligence · Computer Science 2026-01-22 Chongwen Zhao , Yutong Ke , Kaizhu Huang

Many recent studies have found evidence for emergent reasoning capabilities in large language models (LLMs), but debate persists concerning the robustness of these capabilities, and the extent to which they depend on structured reasoning…

Computation and Language · Computer Science 2025-06-09 Yukang Yang , Declan Campbell , Kaixuan Huang , Mengdi Wang , Jonathan Cohen , Taylor Webb

This paper investigates the information encoded in the embeddings of large language models (LLMs). We conduct simulations to analyze the representation entropy and discover a power law relationship with model sizes. Building upon this…

Machine Learning · Computer Science 2024-02-07 Zhiquan Tan , Chenghai Li , Weiran Huang

Large Language Models (LLMs) have demonstrated impressive performance across various tasks, with different models excelling in distinct domains and specific abilities. Effectively combining the predictions of multiple LLMs is crucial for…

Computation and Language · Computer Science 2025-08-01 Jizhou Guo

Growing evidence suggests that large language models do not use their depth uniformly, yet we still lack a fine-grained understanding of their layer-wise prediction dynamics. In this paper, we trace the intermediate representations of…

Computation and Language · Computer Science 2026-03-03 Akshat Gupta , Jay Yeung , Gopala Anumanchipalli , Anna Ivanova

Recent work has sought to quantify large language model uncertainty to facilitate model control and modulate user trust. Previous works focus on measures of uncertainty that are theoretically grounded or reflect the average overt behavior…

Computation and Language · Computer Science 2025-03-18 Kyle Moore , Jesse Roberts , Daryl Watson , Pamela Wisniewski

The design of safety-critical agents based on large language models (LLMs) requires more than simple prompt engineering. This paper presents a comprehensive information-theoretic analysis of how rule encodings in system prompts influence…

Artificial Intelligence · Computer Science 2025-10-10 Joachim Diederich