English
Related papers

Related papers: Inference-Time Reward Hacking in Large Language Mo…

200 papers

While Reinforcement Learning (RL) shows promise in training tool-use Large Language Models (LLMs) using verifiable outcome rewards, existing methods largely overlook the potential of reasoning rewards based on chain-of-thought quality for…

Computation and Language · Computer Science 2026-01-16 Zihan Lin , Xiaohan Wang , Hexiong Yang , Jiajun Chai , Jie Cao , Guojun Yin , Wei Lin , Ran He

Reinforcement learning (RL) has been widely adopted in post-training for large language models (LLMs) at scale. Recently, the incentivization of reasoning capabilities in LLMs from RL indicates that $\textit{proper learning methods could…

Computation and Language · Computer Science 2025-09-26 Zijun Liu , Peiyi Wang , Runxin Xu , Shirong Ma , Chong Ruan , Peng Li , Yang Liu , Yu Wu

Reward models are central to large language model (LLM) post-training. However, past work has shown that they can reward spurious or undesirable attributes such as length, format, hallucinations, and sycophancy. In this work, we introduce…

Machine Learning · Computer Science 2026-02-18 Atticus Wang , Iván Arcuschin , Arthur Conmy

Reinforcement learning with verifiable rewards has been used to elicit impressive performance from language models in many domains. But, broadly beneficial deployments of AI may require us to train models with strong capabilities in…

Machine Learning · Computer Science 2026-05-07 Christine Ye , Joe Benton

Large language model (LLM) alignment relies on complex reward signals that often obscure the specific behaviors being incentivized, creating critical risks of misalignment and reward hacking. Existing interpretation methods typically rely…

Machine Learning · Computer Science 2026-05-22 Edward Chen , Sanmi Koyejo , Carlos Guestrin

Imitation is widely observed in populations of decision-making agents. Using our recent convergence results for asynchronous imitation dynamics on networks, we consider how such networks can be efficiently driven to a desired equilibrium…

Computer Science and Game Theory · Computer Science 2017-04-17 James Riehl , Pouria Ramazi , Ming Cao

We study online learning settings in which experts act strategically to maximize their influence on the learning algorithm's predictions by potentially misreporting their beliefs about a sequence of binary events. Our goal is twofold.…

Machine Learning · Computer Science 2020-07-02 Rupert Freeman , David M. Pennock , Chara Podimata , Jennifer Wortman Vaughan

Reinforcement Learning from Human Feedback (RLHF) aligns language models to human preferences by employing a singular reward model derived from preference data. However, such an approach overlooks the rich diversity of human preferences…

Computation and Language · Computer Science 2024-12-30 Souradip Chakraborty , Jiahao Qiu , Hui Yuan , Alec Koppel , Furong Huang , Dinesh Manocha , Amrit Singh Bedi , Mengdi Wang

Active inference is a probabilistic framework for modelling the behaviour of biological and artificial agents, which derives from the principle of minimising free energy. In recent years, this framework has successfully been applied to a…

Artificial Intelligence · Computer Science 2022-07-13 Lancelot Da Costa , Noor Sajid , Thomas Parr , Karl Friston , Ryan Smith

While numerous works have focused on devising efficient algorithms for reinforcement learning (RL) with uniformly bounded rewards, it remains an open question whether sample or time-efficient algorithms for RL with large state-action space…

Machine Learning · Computer Science 2024-03-08 Jiayi Huang , Han Zhong , Liwei Wang , Lin F. Yang

Aligning Large Language Models (LLMs) to human preferences in content, style, and presentation is challenging, in part because preferences are varied, context-dependent, and sometimes inherently ambiguous. While successful, Reinforcement…

Machine Learning · Computer Science 2024-10-29 Sam Houliston , Alizée Pace , Alexander Immer , Gunnar Rätsch

Aligning multimodal large language models (MLLMs) with human preferences often relies on single-signal, model-based reward methods. Such monolithic rewards often lack confidence calibration across domain-specific tasks, fail to capture…

Artificial Intelligence · Computer Science 2025-10-08 Radha Gulhane , Sathish Reddy Indurthi

Aligning large language models (LLMs) with human preferences has become a critical step in their development. Recent research has increasingly focused on test-time alignment, where additional compute is allocated during inference to enhance…

Computation and Language · Computer Science 2025-09-24 Bolian Li , Yanran Wu , Xinyu Luo , Ruqi Zhang

Reinforcement learning is used to align language models with human preference signals after first pre-training the model to predict the next token of text within a large corpus using likelihood maximization. Before being deployed in a…

Computation and Language · Computer Science 2024-08-30 Alec Solway

Rewards serve as proxies for human preferences and play a crucial role in Reinforcement Learning from Human Feedback (RLHF). However, if these rewards are inherently imperfect, exhibiting various biases, they can adversely affect the…

Machine Learning · Computer Science 2025-11-12 Sheng Ouyang , Yulan Hu , Ge Chen , Qingyang Li , Fuzheng Zhang , Yong Liu

Reward modeling in large language models is susceptible to reward hacking, causing models to latch onto superficial features such as the tendency to generate lists or unnecessarily long responses. In reinforcement learning from human…

Computation and Language · Computer Science 2025-02-19 Taneesh Gupta , Shivam Shandilya , Xuchao Zhang , Rahul Madhavan , Supriyo Ghosh , Chetan Bansal , Huaxiu Yao , Saravan Rajmohan

Hallucinations in Large Language Models (LLMs) -- generations that are plausible but factually unfaithful -- remain a critical barrier to high-stakes deployment. Current detection methods typically rely on computationally expensive external…

Artificial Intelligence · Computer Science 2026-01-23 Manish Bhatt

Commonly in reinforcement learning (RL), rewards are discounted over time using an exponential function to model time preference, thereby bounding the expected long-term reward. In contrast, in economics and psychology, it has been shown…

Machine Learning · Computer Science 2022-12-08 Matthias Schultheis , Constantin A. Rothkopf , Heinz Koeppl

Preference-based fine-tuning has become an important component in training large language models, and the data used at this stage may contain sensitive user information. A central question is how to design a differentially private pipeline…

Machine Learning · Statistics 2026-03-25 Young Hyun Cho , Will Wei Sun

Learning policies via preference-based reward learning is an increasingly popular method for customizing agent behavior, but has been shown anecdotally to be prone to spurious correlations and reward hacking behaviors. While much prior work…

Machine Learning · Computer Science 2023-03-21 Jeremy Tien , Jerry Zhi-Yang He , Zackory Erickson , Anca D. Dragan , Daniel S. Brown
‹ Prev 1 8 9 10 Next ›