相关论文: Transforming and Combining Rewards for Aligning La…

Large language models (LLMs) are often sycophantic, prioritizing agreement with their users over accurate or objective statements. This problematic behavior becomes more pronounced during reinforcement learning from human feedback (RLHF),…

Reinforcement learning from human feedback (RLHF) has emerged as the primary method for aligning large language models (LLMs) with human preferences. The RLHF process typically starts by training a reward model (RM) using human preference…

The effectiveness of reinforcement learning (RL) agents in continuous control robotics tasks is mainly dependent on the design of the underlying reward function, which is highly prone to reward hacking. A misalignment between the reward…

Multi-objective preference alignment in language models often encounters a challenging trade-off: optimizing for one human preference (e.g., helpfulness) frequently compromises others (e.g., harmlessness) due to the inherent conflicts…

Learning from human feedback has shown success in aligning large, pretrained models with human values. Prior works have mostly focused on learning from high-level labels, such as preferences between pairs of model outputs. On the other…

Reinforcement Learning (RL) has proven highly effective in aligning Large Language Models (LLMs) with human preferences. Typical RL methods optimize under an overall sequence reward, which can lead to a suboptimal learning process. This…

Reward models (RMs) are a crucial component in the alignment of large language models' (LLMs) outputs with human values. RMs approximate human preferences over possible LLM responses to the same prompt by predicting and comparing reward…

Despite the general capabilities of pre-trained large language models (LLMs), they still need further adaptation to better serve practical applications. In this paper, we demonstrate the interchangeability of three popular and distinct…

Large Language Models (LLMs) have revolutionized natural language processing, yet aligning these models with human values and preferences using RLHF remains a significant challenge. This challenge is characterized by various instabilities,…

The alignment of Large Language Models (LLMs) is critically dependent on reward models trained on costly human preference data. While recent work explores bypassing this cost with AI feedback, these methods often lack a rigorous theoretical…

We consider the problem of multi-objective alignment of foundation models with human preferences, which is a critical step towards helpful and harmless AI systems. However, it is generally costly and unstable to fine-tune large foundation…

ChatGLM is a free-to-use AI service powered by the ChatGLM family of large language models (LLMs). In this paper, we present the ChatGLM-RLHF pipeline -- a reinforcement learning from human feedback (RLHF) system -- designed to enhance…

We introduce a large language model (LLM) capable of processing speech inputs and show that tuning it further with reinforcement learning on human preference (RLHF) enables it to adapt better to disordered speech than traditional…

Humans use social context to specify preferences over behaviors, i.e. their reward functions. Yet, algorithms for inferring reward models from preference data do not take this social learning view into account. Inspired by pragmatic human…

Reward models have become a staple in modern NLP, serving as not only a scalable text evaluator, but also an indispensable component in many alignment recipes and inference-time algorithms. However, while recent reward models increase…

Misalignment between model predictions and intended usage can be detrimental for the deployment of computer vision models. The issue is exacerbated when the task involves complex structured outputs, as it becomes harder to design procedures…

Alignment via reinforcement learning from human feedback (RLHF) has become the dominant paradigm for controlling the quality of outputs from large language models (LLMs). However, existing theories do not provide strong justification for…

Reward models are a key component of large language model alignment, serving as proxies for human preferences during training. However, existing evaluations focus primarily on broad instruction-following benchmarks, providing limited…

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…

As large language models (LLMs) become increasingly embedded in everyday applications, ensuring their alignment with the diverse preferences of individual users has become a critical challenge. Currently deployed approaches typically assume…