English
Related papers

Related papers: Language Models Learn to Mislead Humans via RLHF

200 papers

Numerous previous studies have sought to determine to what extent language models, pretrained on natural language text, can serve as useful models of human cognition. In this paper, we are interested in the opposite question: whether we can…

Computation and Language · Computer Science 2024-10-18 Samuel Kiegeland , Ethan Gotlieb Wilcox , Afra Amini , David Robert Reich , Ryan Cotterell

Foundation models, specifically Large Language Models (LLMs), have lately gained wide-spread attention and adoption. Reinforcement Learning with Human Feedback (RLHF) involves training a reward model to capture desired behaviors, which is…

Computation and Language · Computer Science 2024-01-25 Will LeVine , Benjamin Pikus , Anthony Chen , Sean Hendryx

While recent advances have boosted LM proficiency in linguistic benchmarks, LMs consistently struggle to reason correctly on complex tasks like mathematics. We turn to Reinforcement Learning from Human Feedback (RLHF) as a method with which…

Computation and Language · Computer Science 2023-11-13 Sarah Pan , Vladislav Lialin , Sherin Muckatira , Anna Rumshisky

Confirmation bias, the tendency to seek evidence that supports rather than challenges one's belief, hinders one's reasoning ability. We examine whether large language models (LLMs) exhibit confirmation bias by adapting the rule-discovery…

Computation and Language · Computer Science 2026-04-06 Ayush Rajesh Jhaveri , Anthony GX-Chen , Ilia Sucholutsky , Eunsol Choi

This paper studies reinforcement learning from human feedback (RLHF) for aligning large language models with human preferences. While RLHF has demonstrated promising results, many algorithms are highly sensitive to misspecifications in the…

Machine Learning · Computer Science 2025-10-30 Erhan Xu , Kai Ye , Hongyi Zhou , Luhan Zhu , Francesco Quinzan , Chengchun Shi

While Reinforcement Learning from Human Feedback (RLHF) has shown promise in aligning generative AI, we present empirical evidence that it can also cause severe, systematic misalignment. We hypothesize that this stems from evaluator…

Machine Learning · Computer Science 2025-06-11 Kaiqu Liang , Haimin Hu , Ryan Liu , Thomas L. Griffiths , Jaime Fernández Fisac

Recent strides in large language models (LLMs) have yielded remarkable performance, leveraging reinforcement learning from human feedback (RLHF) to significantly enhance generation and alignment capabilities. However, RLHF encounters…

Computation and Language · Computer Science 2024-05-31 Kuo Liao , Shuang Li , Meng Zhao , Liqun Liu , Mengge Xue , Zhenyu Hu , Honglin Han , Chengguo Yin

Learning from human feedback has been shown to be effective at aligning language models with human preferences. Past work has often relied on Reinforcement Learning from Human Feedback (RLHF), which optimizes the language model using reward…

Computation and Language · Computer Science 2023-05-18 Yao Zhao , Rishabh Joshi , Tianqi Liu , Misha Khalman , Mohammad Saleh , Peter J. Liu

Reinforcement Learning from Human Feedback aligns the outputs of Large Language Models with human values and preferences. Central to this process is the reward model (RM), which translates human feedback into training signals for optimising…

Artificial Intelligence · Computer Science 2025-09-23 Zeyu Huang , Zihan Qiu , Zili Wang , Edoardo M. Ponti , Ivan Titov

Reinforcement learning from human feedback (RLHF) has proven effective in enhancing the instruction-following capabilities of large language models; however, it remains underexplored in the cross-modality domain. As the number of modalities…

Traditional software fault injection methods, while foundational, face limitations in adequately representing real-world faults, offering customization, and requiring significant manual effort and expertise. This paper introduces a novel…

Software Engineering · Computer Science 2024-04-12 Domenico Cotroneo , Pietro Liguori

Aligning large language models (LLMs) with human preferences has proven to drastically improve usability and has driven rapid adoption as demonstrated by ChatGPT. Alignment techniques such as supervised fine-tuning (SFT) and reinforcement…

We find that language models have difficulties generating fallacious and deceptive reasoning. When asked to generate deceptive outputs, language models tend to leak honest counterparts but believe them to be false. Exploiting this…

Computation and Language · Computer Science 2025-05-26 Yue Zhou , Henry Peng Zou , Barbara Di Eugenio , Yang Zhang

Accurately handling the underlying support values in sentences is crucial for understanding the speaker's tendencies, yet it poses a challenging task in natural language understanding (NLU). In this article, we explore the potential of…

Computation and Language · Computer Science 2024-03-18 Pingwei Sun

A key challenge in training Large Language Models (LLMs) is properly aligning them with human preferences. Reinforcement Learning with Human Feedback (RLHF) uses pairwise comparisons from human annotators to train reward functions and has…

Machine Learning · Computer Science 2025-01-17 Ariel D. Procaccia , Benjamin Schiffer , Shirley Zhang

When we read, we make predictions about upcoming words; these predictions influence our reading behavior. The success of large language models (LLMs), which, like humans, make predictions about upcoming words, has motivated their use as…

Computation and Language · Computer Science 2026-05-27 Byung-Doh Oh , Tal Linzen

Past analyses of reinforcement learning from human feedback (RLHF) assume that the human evaluators fully observe the environment. What happens when human feedback is based only on partial observations? We formally define two failure cases:…

Machine Learning · Computer Science 2024-11-19 Leon Lang , Davis Foote , Stuart Russell , Anca Dragan , Erik Jenner , Scott Emmons

Modern language models can imitate complex patterns through few-shot learning, enabling them to complete challenging tasks without fine-tuning. However, imitation can also lead models to reproduce inaccuracies or harmful content if present…

Machine Learning · Computer Science 2024-03-13 Danny Halawi , Jean-Stanislas Denain , Jacob Steinhardt

Large Language Models (LLMs) have demonstrated remarkable capabilities in open-ended text generation tasks. However, the inherent open-ended nature of these tasks implies that there is always room for improvement in the quality of model…

Computation and Language · Computer Science 2024-09-16 Ziqi Wang , Le Hou , Tianjian Lu , Yuexin Wu , Yunxuan Li , Hongkun Yu , Heng Ji

Robust, faithful and harm-free pronoun use for individuals is an important goal for language model development as their use increases, but prior work tends to study only one or two of these characteristics at a time. To measure progress…

Computation and Language · Computer Science 2024-10-08 Vagrant Gautam , Eileen Bingert , Dawei Zhu , Anne Lauscher , Dietrich Klakow