Related papers: CREAM: Consistency Regularized Self-Rewarding Lang…

Existing benchmarks for large language models (LLMs) predominantely focus on assessing their capabilities through verifiable tasks. Such objective and static benchmarks offer limited utility for practical LLM selection, making it difficult…

Existing alignment methods directly use the reward model learned from user preference data to optimize an LLM policy, subject to KL regularization with respect to the base policy. This practice is suboptimal for maximizing user's utility…

Recent advances in aligning Large Language Models with human preferences have benefited from larger reward models and better preference data. However, most of these methodologies rely on the accuracy of the reward model. The reward models…

Proximal Policy Optimization (PPO) is commonly used in Reinforcement Learning from Human Feedback to align large language models (LLMs) with downstream tasks. This paper investigates the feasibility of using PPO for direct reinforcement…

Mathematical reasoning in large language models has improved substantially with reinforcement learning using verifiable rewards, where final answers can be checked automatically and converted into reliable training signals. Most such…

Human feedback plays a pivotal role in aligning large language models (LLMs) with human preferences. However, such feedback is often noisy or inconsistent, which can degrade the quality of reward models and hinder alignment. While various…

Traditionally, reward models used for reinforcement learning from human feedback (RLHF) are trained to directly predict preference scores without leveraging the generation capabilities of the underlying large language model (LLM). This…

Preference alignment via reward models helps build safe, helpful, and reliable large language models (LLMs). However, subjectivity in preference judgments and the lack of representative sampling in preference data collection can introduce…

Direct Preference Optimization (DPO) and its variants have become the de facto standards for aligning large language models (LLMs) with human preferences or specific goals. However, DPO requires high-quality preference data and suffers from…

In reinforcement learning from human feedback, preference-based reward models play a central role in aligning large language models to human-aligned behavior. However, recent studies show that these models are prone to reward hacking and…

Reward models (RMs) are central to aligning large language models (LLMs) with human values but have received less attention than pretrained and post-trained LLMs themselves. Because RMs are initialized from LLMs, they inherit…

Reward models are pivotal for aligning Large Language Models (LLMs) with human preferences. Existing approaches face two key limitations: Discriminative reward models require large-scale annotated data, as they cannot exploit the preference…

Test-time scaling (TTS) for large language models (LLMs) has thus far fallen into two largely separate paradigms: (1) reinforcement learning (RL) methods that optimize sparse outcome-based rewards, yet suffer from instability and low sample…

We tackle the problem of aligning pre-trained large language models (LMs) with human preferences. If we view text generation as a sequential decision-making problem, reinforcement learning (RL) appears to be a natural conceptual framework.…

Alignment of large language models (LLMs) with human preferences typically relies on supervised reward models or external judges that demand abundant annotations. However, in fields that rely on professional knowledge, such as medicine and…

Recent Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) increasingly use Reinforcement Learning (RL) for post-pretraining, such as RL with Verifiable Rewards (RLVR) for objective tasks and RL from Human Feedback (RLHF)…

In reinforcement learning, specification gaming occurs when AI systems learn undesired behaviors that are highly rewarded due to misspecified training goals. Specification gaming can range from simple behaviors like sycophancy to…

Reward models (RMs) are essential for aligning large language models (LLM) with human expectations. However, existing RMs struggle to capture the stochastic and uncertain nature of human preferences and fail to assess the reliability of…

Reinforcement Learning from Human Feedback (RLHF) involves training policy models (PMs) and reward models (RMs) to align language models with human preferences. Instead of focusing solely on PMs and RMs independently, we propose to examine…

Reward models (RMs) are central to the alignment of language models (LMs). An RM often serves as a proxy for human preferences to guide downstream LM behavior. However, our understanding of RM behavior is limited. Our work (i) formalizes a…