English
Related papers

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

200 papers

Inference-time alignment enhances the performance of large language models without requiring additional training or fine-tuning but presents challenges due to balancing computational efficiency with high-quality output. Best-of-N (BoN)…

Computation and Language · Computer Science 2025-09-04 Jiahao Qiu , Yifu Lu , Yifan Zeng , Jiacheng Guo , Jiayi Geng , Chenhao Zhu , Xinzhe Juan , Ling Yang , Huazheng Wang , Kaixuan Huang , Yue Wu , Mengdi Wang

Existing approaches to reward inference from behavior typically assume that humans provide demonstrations according to specific models of behavior. However, humans often indicate their goals through a wide range of behaviors, from actions…

Machine Learning · Computer Science 2025-02-26 Will Schwarzer , Jordan Schneider , Philip S. Thomas , Scott Niekum

Numerous algorithms have been proposed to $\textit{align}$ language models to remove undesirable behaviors. However, the challenges associated with a very large state space and creating a proper reward function often result in various…

Computation and Language · Computer Science 2024-06-06 Suraj Anand , David Getzen

Reinforcement learning from human feedback (RLHF) is a standard approach for fine-tuning large language models to follow instructions. As part of this process, learned reward models are used to approximately model human preferences.…

Machine Learning · Computer Science 2024-03-12 Thomas Coste , Usman Anwar , Robert Kirk , David Krueger

The demand for regulating potentially risky behaviors of large language models (LLMs) has ignited research on alignment methods. Since LLM alignment heavily relies on reward models for optimization or evaluation, neglecting the quality of…

Computation and Language · Computer Science 2024-10-01 Yan Liu , Xiaoyuan Yi , Xiaokang Chen , Jing Yao , Jingwei Yi , Daoguang Zan , Zheng Liu , Xing Xie , Tsung-Yi Ho

Reinforcement learning (RL) has shown extraordinary potential in aligning diffusion models to downstream tasks, yet most of them still suffer from significant reward hacking, which degrades generative diversity and quality by inducing…

Machine Learning · Computer Science 2026-05-14 Jiaming Li , Chenyu Zhu , Nanxi Yi , Youjun Bao , Li Sun , Quanying Lv , Xiang Fang , Daizong Liu , Jianjun Li , Kun He , Bowen Zhou , Zhiyuan Ma

Large language models (LLMs) are known for their exceptional performance across a range of natural language processing tasks, but their deployment comes at a high computational and financial cost. On the other hand, smaller language models…

Computation and Language · Computer Science 2024-09-24 Adarsh MS , Jithin VG , Ditto PS

Reinforcement Learning from Human Feedback (RLHF) enables powerful LLM alignment but can introduce reward hacking - models exploit spurious correlations in proxy rewards without genuine alignment. Compounding this, the objectives…

Artificial Intelligence · Computer Science 2026-02-24 Mohammad Beigi , Ming Jin , Junshan Zhang , Jiaxin Zhang , Qifan Wang , Lifu Huang

Autonomous agents optimize the reward function we give them. What they don't know is how hard it is for us to design a reward function that actually captures what we want. When designing the reward, we might think of some specific training…

Artificial Intelligence · Computer Science 2020-10-08 Dylan Hadfield-Menell , Smitha Milli , Pieter Abbeel , Stuart Russell , Anca Dragan

We seek to align agent behavior with a user's objectives in a reinforcement learning setting with unknown dynamics, an unknown reward function, and unknown unsafe states. The user knows the rewards and unsafe states, but querying the user…

Computers and Society · Computer Science 2021-03-26 Siddharth Reddy , Anca D. Dragan , Sergey Levine , Shane Legg , Jan Leike

Large Language Models (LLMs) demonstrate transformative potential, yet their reasoning remains inconsistent and unreliable. Reinforcement learning (RL)-based fine-tuning is a key mechanism for improvement, but its effectiveness is…

Machine Learning · Computer Science 2026-02-11 Pei-Chi Pan , Yingbin Liang , Sen Lin

In many sequential decision making tasks, it is challenging to design reward functions that help an RL agent efficiently learn behavior that is considered good by the agent designer. A number of different formulations of the reward-design…

Artificial Intelligence · Computer Science 2018-06-25 Zeyu Zheng , Junhyuk Oh , Satinder Singh

In reinforcement learning (RL), agents continually interact with the environment and use the feedback to refine their behavior. To guide policy optimization, reward models are introduced as proxies of the desired objectives, such that when…

Machine Learning · Computer Science 2025-06-19 Rui Yu , Shenghua Wan , Yucen Wang , Chen-Xiao Gao , Le Gan , Zongzhang Zhang , De-Chuan Zhan

As long-horizon coding agents produce more code than any developer can review, oversight collapses onto a single surface: the automated test suite. Reward hacking naturally arises in this setup, as the agent optimizes for passing tests…

Software Engineering · Computer Science 2026-05-21 Bingchen Zhao , Dhruv Srikanth , Yuxiang Wu , Zhengyao Jiang

Recent years have seen considerable advancements in multi-step reasoning with Large Language Models (LLMs). The previous studies have elucidated the merits of integrating feedback or search mechanisms during model inference to improve the…

Computation and Language · Computer Science 2023-10-17 Qianli Ma , Haotian Zhou , Tingkai Liu , Jianbo Yuan , Pengfei Liu , Yang You , Hongxia Yang

Estimating how well a machine learning model performs during inference is critical in a variety of scenarios (for example, to quantify uncertainty, or to choose from a library of available models). However, the standard accuracy estimate of…

Computer Vision and Pattern Recognition · Computer Science 2021-10-07 Xuechen Zhang , Samet Oymak , Jiasi Chen

Large Language Models (LLMs) often exhibit strong linguistic abilities while remaining unreliable on multi-step reasoning tasks, particularly when deployed without additional training or fine-tuning. In this work, we study inference-time…

Computation and Language · Computer Science 2026-03-24 Vinay Sharma , Manish Jain

Aligning LLMs with user preferences is crucial for real-world use but often requires costly fine-tuning or expensive inference, forcing trade-offs between alignment quality and computational cost. Existing inference-time methods typically…

Machine Learning · Computer Science 2025-08-08 Mason Nakamura , Saaduddin Mahmud , Kyle H. Wray , Hamed Zamani , Shlomo Zilberstein

Meta-Black-Box Optimization (MetaBBO) is an emerging avenue within Optimization community, where algorithm design policy could be meta-learned by reinforcement learning to enhance optimization performance. So far, the reward functions in…

Machine Learning · Computer Science 2026-01-30 Zechuan Huang , Zhiguang Cao , Hongshu Guo , Yue-Jiao Gong , Zeyuan Ma

We deploy large language models (LLMs) as business development (BD) agents for persuasive price negotiation in online travel agencies (OTAs). The agent must follow a multi-stage Standard Operating Procedure (SOP) and strict guardrails (no…

Computation and Language · Computer Science 2026-04-30 Xia Zeng , Yihan Chen , Luhui Liu , Chao Luo , Ye Chen , Zhuoran Zhuang