# Online Preference-based Reinforcement Learning with Self-augmented Feedback from Large Language Model

arXiv: https://arxiv.org/abs/2412.16878 · HTML: https://mlsanity.com/paper/2412.16878/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Reward modeling and RLHF, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 21 of its 39 references that are in the map.
- First posted on arXiv 22 Dec 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 21 of 21)

- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning](https://mlsanity.com/paper/1910.10897/) (2019)
- [Vision-Language Models for Vision Tasks: A Survey](https://mlsanity.com/paper/2304.00685/) (2023)
- [PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training](https://mlsanity.com/paper/2106.05091/) (2021)
- [A Survey of Inverse Reinforcement Learning: Challenges, Methods and Progress](https://mlsanity.com/paper/1806.06877/) (2018)
- [Principled Reinforcement Learning with Human Feedback from Pairwise or K-wise Comparisons](https://mlsanity.com/paper/2301.11270/) (2023)
- [RoboGen: Towards Unleashing Infinite Data for Automated Robot Learning via Generative Simulation](https://mlsanity.com/paper/2311.01455/) (2023)
- [LIV: Language-Image Representations and Rewards for Robotic Control](https://mlsanity.com/paper/2306.00958/) (2023)
- [RL-VLM-F: Reinforcement Learning from Vision Language Foundation Model Feedback](https://mlsanity.com/paper/2402.03681/) (2024)
- [LLM Critics Help Catch LLM Bugs](https://mlsanity.com/paper/2407.00215/) (2024)
- [RoboCLIP: One Demonstration is Enough to Learn Robot Policies](https://mlsanity.com/paper/2310.07899/) (2023)
- [B-Pref: Benchmarking Preference-Based Reinforcement Learning](https://mlsanity.com/paper/2111.03026/) (2021)
- [Unpacking Reward Shaping: Understanding the Benefits of Reward Engineering on Sample Complexity](https://mlsanity.com/paper/2210.09579/) (2022)
- [Code as Reward: Empowering Reinforcement Learning with VLMs](https://mlsanity.com/paper/2402.04764/) (2024)
- [PlanAgent: A Multi-modal Large Language Agent for Closed-loop Vehicle Motion Planning](https://mlsanity.com/paper/2406.01587/) (2024)
- [AlignDiff: Aligning Diverse Human Preferences via Behavior-Customisable Diffusion Model](https://mlsanity.com/paper/2310.02054/) (2023)
- [FuRL: Visual-Language Models as Fuzzy Rewards for Reinforcement Learning](https://mlsanity.com/paper/2406.00645/) (2024)
- [Data Scaling Laws for Imitation Learning-Based End-to-End Autonomous Driving](https://mlsanity.com/paper/2412.02689/) (2024)
- [Text-Aware Diffusion for Policy Learning](https://mlsanity.com/paper/2407.01903/) (2024)

## Built on by (most cited first, 7 of 7)

- [Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RL](https://mlsanity.com/paper/2505.10832/) (2025)
- [ReasonPlan: Unified Scene Prediction and Decision Reasoning for Closed-loop Autonomous Driving](https://mlsanity.com/paper/2505.20024/) (2025)
- [AlphaDecay: Module-wise Weight Decay for Heavy-Tailed Balancing in LLMs](https://mlsanity.com/paper/2506.14562/) (2025)
- [AED: Automatic Discovery of Effective and Diverse Vulnerabilities for Autonomous Driving Policy with Large Language Models](https://mlsanity.com/paper/2503.20804/) (2025)
- [Evaluating Feature Dependent Noise in Preference-based Reinforcement Learning](https://mlsanity.com/paper/2601.01904/) (2026)
- [How well can LLMs provide planning feedback in grounded environments?](https://mlsanity.com/paper/2509.09790/) (2025)
- [Multi-Objective Reward and Preference Optimization: Theory and Algorithms](https://mlsanity.com/paper/2512.10601/) (2025)

## Abstract (arXiv)

Preference-based reinforcement learning (PbRL) provides a powerful paradigm to avoid meticulous reward engineering by learning rewards based on human preferences. However, real-time human feedback is hard to obtain in online tasks. Most work suppose there is a "scripted teacher" that utilizes privileged predefined reward to provide preference feedback. In this paper, we propose a RL Self-augmented Large Language Model Feedback (RL-SaLLM-F) technique that does not rely on privileged information for online PbRL. RL-SaLLM-F leverages the reflective and discriminative capabilities of LLM to generate self-augmented trajectories and provide preference labels for reward learning. First, we identify an failure issue in LLM-based preference discrimination, specifically "query ambiguity", in online PbRL. Then LLM is employed to provide preference labels and generate self-augmented imagined trajectories that better achieve the task goal, thereby enhancing the quality and efficiency of feedback. Additionally, a double-check mechanism is introduced to mitigate randomness in the preference labels, improving the reliability of LLM feedback. The experiment across multiple tasks in the MetaWorld benchmark demonstrates the specific contributions of each proposed module in RL-SaLLM-F, and shows that self-augmented LLM feedback can effectively replace the impractical "scripted teacher" feedback. In summary, RL-SaLLM-F introduces a new direction of feedback acquisition in online PbRL that does not rely on any online privileged information, offering an efficient and lightweight solution with LLM-driven feedback.