相关论文: Reliability and Learnability of Human Bandit Feedb…

We consider the problem of fitting a reinforcement learning (RL) model to some given behavioral data under a multi-armed bandit environment. These models have received much attention in recent years for characterizing human and animal…

How can we make models robust to even imperfect human feedback? In reinforcement learning from human feedback (RLHF), human preferences over model outputs are used to train a reward model that assigns scalar values to responses. Because…

Designing an effective reward function has long been a challenge in reinforcement learning, particularly for complex tasks in unstructured environments. To address this, various learning paradigms have emerged that leverage different forms…

The sequential nature of decision-making in financial asset trading aligns naturally with the reinforcement learning (RL) framework, making RL a common approach in this domain. However, the low signal-to-noise ratio in financial markets…

The majority of language model training builds on imitation learning. It covers pretraining, supervised fine-tuning, and affects the starting conditions for reinforcement learning from human feedback (RLHF). The simplicity and scalability…

Reinforcement learning from human feedback (RLHF) is a variant of reinforcement learning (RL) that learns from human feedback instead of relying on an engineered reward function. Building on prior work on the related setting of…

In modern ML Ops environments, model deployment is a critical process that traditionally relies on static heuristics such as validation error comparisons and A/B testing. However, these methods require human intervention to adapt to…

Reinforcement learning from human feedback (RLHF) is the mainstream paradigm used to align large language models (LLMs) with human preferences. Yet existing RLHF heavily relies on accurate and informative reward models, which are vulnerable…

Reinforcement learning from human feedback (RLHF) has emerged as a key technique for aligning the output of large language models (LLMs) with human preferences. To learn the reward function, most existing RLHF algorithms use the…

We present the first real-world application of methods for improving neural machine translation (NMT) with human reinforcement, based on explicit and implicit user feedback collected on the eBay e-commerce platform. Previous work has been…

Reinforcement Learning (RL) heavily relies on the careful design of the reward function. However, accurately assigning rewards to each state-action pair in Long-Term Reinforcement Learning (LTRL) tasks remains a significant challenge. As a…

We present an approach to structured prediction from bandit feedback, called Bandit Structured Prediction, where only the value of a task loss function at a single predicted point, instead of a correct structure, is observed in learning. We…

Faithfulness, expressiveness, and elegance is the constant pursuit in machine translation. However, traditional metrics like \textit{BLEU} do not strictly align with human preference of translation quality. In this paper, we explore…

The complexity of designing reward functions has been a major obstacle to the wide application of deep reinforcement learning (RL) techniques. Describing an agent's desired behaviors and properties can be difficult, even for experts. A new…

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…

Reinforcement learning with verifiable rewards can improve LLM reasoning, but learning remains sample-inefficient when terminal rewards are sparse. This has motivated a growing line of work on RL with textual feedback, where a critic model…

Training large language models with reinforcement learning (RL) against verifiable rewards significantly enhances their reasoning abilities, yet remains computationally expensive due to inefficient uniform prompt sampling. We introduce…

Reinforcement Learning with Verifiable reward (RLVR) on preference data has become the mainstream approach for training Generative Reward Models (GRMs). Typically in pairwise rewarding tasks, GRMs generate reasoning chains ending with…

Reinforcement Learning (RL) plays an important role in the robotic manipulation domain since it allows self-learning from trial-and-error interactions with the environment. Still, sample efficiency and reward specification seriously limit…

We apply preference modeling and reinforcement learning from human feedback (RLHF) to finetune language models to act as helpful and harmless assistants. We find this alignment training improves performance on almost all NLP evaluations,…