相关论文: SAFER: Probing Safety in Reward Models with Sparse…

As large language models (LLMs) are increasingly deployed in diverse user facing applications, aligning them with real user preferences becomes essential. Existing methods like Reinforcement Learning from Human Feedback (RLHF) rely on…

Sparse autoencoders (SAEs) provide a powerful mechanism for decomposing the dense representations produced by Large Language Models (LLMs) into interpretable latent features. We posit that SAEs constitute a natural foundation for Learned…

Building neural reward models from human preferences is a pivotal component in reinforcement learning from human feedback (RLHF) and large language model alignment research. Given the scarcity and high cost of human annotation, how to…

While Reinforcement Learning from Human Feedback (RLHF) is widely used to align Large Language Models (LLMs) with human preferences, it typically assumes homogeneous preferences across users, overlooking diverse human values and minority…

Addressing the critical need for robust safety in Large Language Models (LLMs), particularly against adversarial attacks and in-distribution errors, we introduce Reinforcement Learning with Backtracking Feedback (RLBF). This framework…

We study the challenge of achieving theoretically grounded feature recovery using Sparse Autoencoders (SAEs) for the interpretation of Large Language Models. Existing SAE training algorithms often lack rigorous mathematical guarantees and…

Reinforcement Learning from Human Feedback (RLHF) has achieved considerable success in aligning large language models (LLMs) by modeling human preferences with a learnable reward model and employing a reinforcement learning algorithm to…

Reinforcement learning from human feedback (RLHF) has been an effective technique for aligning AI systems with human values, with remarkable successes in fine-tuning large-language models recently. Most existing RLHF paradigms make the…

Current RLHF methods such as PPO and DPO typically reduce human preferences to binary labels, which are costly to obtain and too coarse to reflect individual variation. We observe that expressions of satisfaction and dissatisfaction follow…

Reinforcement learning with verifiable rewards (RLVR) scales the reasoning ability of large language models (LLMs) but remains bottlenecked by limited labeled samples for continued data scaling. Reinforcement learning with intrinsic rewards…

The correct specification of reward models is a well-known challenge in reinforcement learning. Hand-crafted reward functions often lead to inefficient or suboptimal policies and may not be aligned with user values. Reinforcement learning…

Large Language Models (LLMs) fine-tuned via Reinforcement Learning from Human Feedback (RLHF) and Reinforcement Learning with Verifiable Rewards (RLVR) significantly improve the alignment of human-AI values, further raising the upper bound…

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…

The trustworthiness of Large Language Models (LLMs) refers to the extent to which their outputs are reliable, safe, and ethically aligned, and it has become a crucial consideration alongside their cognitive performance. In practice,…

Reinforcement Learning from Human Feedback (RLHF) leverages human preference data to train language models to align more closely with human essence. These human preference data, however, are labeled at the sequence level, creating a…

Sparse autoencoders (SAEs) are a recent technique for decomposing neural network activations into human-interpretable features. However, in order for SAEs to identify all features represented in frontier models, it will be necessary to…

Reward Models (RMs) are crucial for online alignment of language models (LMs) with human preferences. However, RM-based preference-tuning is vulnerable to reward hacking, whereby LM policies learn undesirable behaviors from flawed RMs. By…

Reward models (RMs) are essential for aligning Large Language Models (LLMs) with human preferences. However, they often struggle with capturing complex human preferences and generalizing to unseen data. To address these challenges, we…

Large language models are typically aligned with human preferences by optimizing $\textit{reward models}$ (RMs) fitted to human feedback. However, human preferences are multi-faceted, and it is increasingly common to derive reward from a…

Pre-trained large-scale language models (LLMs) excel at producing coherent articles, yet their outputs may be untruthful, toxic, or fail to align with user expectations. Current approaches focus on using reinforcement learning with human…