# Online Intrinsic Rewards for Decision Making Agents from Large Language Model Feedback

arXiv: https://arxiv.org/abs/2410.23022 · HTML: https://mlsanity.com/paper/2410.23022/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use.
- Builds on 41 of its 57 references that are in the map.
- First posted on arXiv 30 Oct 2024.

## Placed in

- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)
- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 30 of 41)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [PaLM-E: An Embodied Multimodal Language Model](https://mlsanity.com/paper/2303.03378/) (2023)
- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [Curiosity-driven Exploration by Self-supervised Prediction](https://mlsanity.com/paper/1705.05363/) (2017)
- [Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)](https://mlsanity.com/paper/1511.07289/) (2015)
- [Exploration by Random Network Distillation](https://mlsanity.com/paper/1810.12894/) (2018)
- [Unifying Count-Based Exploration and Intrinsic Motivation](https://mlsanity.com/paper/1606.01868/) (2016)
- [Eureka: Human-Level Reward Design via Coding Large Language Models](https://mlsanity.com/paper/2310.12931/) (2023)
- [MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge](https://mlsanity.com/paper/2206.08853/) (2022)
- [Language to Rewards for Robotic Skill Synthesis](https://mlsanity.com/paper/2306.08647/) (2023)
- [Reward Design with Language Models](https://mlsanity.com/paper/2303.00001/) (2023)
- [Go-Explore: a New Approach for Hard-Exploration Problems](https://mlsanity.com/paper/1901.10995/) (2019)
- [The NetHack Learning Environment](https://mlsanity.com/paper/2006.13760/) (2020)
- [Guiding Pretraining in Reinforcement Learning with Large Language Models](https://mlsanity.com/paper/2302.06692/) (2023)
- [Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning](https://mlsanity.com/paper/2310.12921/) (2023)
- [Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models](https://mlsanity.com/paper/1507.00814/) (2015)
- [RIDE: Rewarding Impact-Driven Exploration for Procedurally-Generated Environments](https://mlsanity.com/paper/2002.12292/) (2020)
- [Survey on Large Language Model-Enhanced Reinforcement Learning: Concept, Taxonomy, and Methods](https://mlsanity.com/paper/2404.00282/) (2024)
- [Model-Based Active Exploration](https://mlsanity.com/paper/1810.12162/) (2018)
- [Motif: Intrinsic Motivation from Artificial Intelligence Feedback](https://mlsanity.com/paper/2310.00166/) (2023)
- [Sample Factory: Egocentric 3D Control from Pixels at 100000 FPS with Asynchronous Reinforcement Learning](https://mlsanity.com/paper/2006.11751/) (2020)
- [PC-PG: Policy Cover Directed Exploration for Provable Policy Gradient Learning](https://mlsanity.com/paper/2007.08459/) (2020)
- [The Laplacian in RL: Learning Representations with Efficient Approximations](https://mlsanity.com/paper/1810.04586/) (2018)
- [Auto MC-Reward: Automated Dense Reward Design with Large Language Models for Minecraft](https://mlsanity.com/paper/2312.09238/) (2023)

## Built on by (most cited first, 10 of 10)

- [Diversity-Incentivized Exploration for Versatile Reasoning](https://mlsanity.com/paper/2509.26209/) (2025)
- [From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms](https://mlsanity.com/paper/2605.06716/) (2026)
- [Toward Efficient Exploration by Large Language Model Agents](https://mlsanity.com/paper/2504.20997/) (2025)
- [Meta-Thinking in LLMs via Multi-Agent Reinforcement Learning: A Survey](https://mlsanity.com/paper/2504.14520/) (2025)
- [Scalable Option Learning in High-Throughput Environments](https://mlsanity.com/paper/2509.00338/) (2025)
- [ARM-FM: Automated Reward Machines via Foundation Models for Compositional Reinforcement Learning](https://mlsanity.com/paper/2510.14176/) (2025)
- [PCGRLLM: Large Language Model-Driven Reward Design for Procedural Content Generation Reinforcement Learning](https://mlsanity.com/paper/2502.10906/) (2025)
- [Leveraging LLMs for reward function design in reinforcement learning control tasks](https://mlsanity.com/paper/2511.19355/) (2025)
- [Goal-Guided Efficient Exploration via Large Language Model in Reinforcement Learning](https://mlsanity.com/paper/2509.22008/) (2025)
- [Hierarchical Behaviour Spaces](https://mlsanity.com/paper/2604.24558/) (2026)

## Abstract (arXiv)

Automatically synthesizing dense rewards from natural language descriptions is a promising paradigm in reinforcement learning (RL), with applications to sparse reward problems, open-ended exploration, and hierarchical skill design. Recent works have made promising steps by exploiting the prior knowledge of large language models (LLMs). However, these approaches suffer from important limitations: they are either not scalable to problems requiring billions of environment samples, due to requiring LLM annotations for each observation, or they require a diverse offline dataset, which may not exist or be impossible to collect. In this work, we address these limitations through a combination of algorithmic and systems-level contributions. We propose ONI, a distributed architecture that simultaneously learns an RL policy and an intrinsic reward function using LLM feedback. Our approach annotates the agent's collected experience via an asynchronous LLM server, which is then distilled into an intrinsic reward model. We explore a range of algorithmic choices for reward modeling with varying complexity, including hashing, classification, and ranking models. Our approach achieves state-of-the-art performance across a range of challenging tasks from the NetHack Learning Environment, while removing the need for large offline datasets required by prior work. We make our code available at https://github.com/facebookresearch/oni.