mlsanityData to 14 Oct 2026

Language Instructed Reinforcement Learning for Human-AI Coordination

arXiv 2304.07297 ↗First posted 13 Apr 2023cs.AI cs.CL cs.LGcited by 63

Not on the map. Why?

Abstract (arXiv)

One of the fundamental quests of AI is to produce agents that coordinate well with humans. This problem is challenging, especially in domains that lack high quality human behavioral data, because multi-agent reinforcement learning (RL) often converges to different equilibria from the ones that humans prefer. We propose a novel framework, instructRL, that enables humans to specify what kind of strategies they expect from their AI partners through natural language instructions. We use pretrained large language models to generate a prior policy conditioned on the human instruction and use the prior to regularize the RL objective. This leads to the RL agent converging to equilibria that are aligned with human preferences. We show that instructRL converges to human-like policies that satisfy the given instructions in a proof-of-concept environment as well as the challenging Hanabi benchmark. Finally, we show that knowing the language instruction significantly boosts human-AI coordination performance in human evaluations in Hanabi.

Key facts. Cited by 63 papers in the map (data through 14 Oct 2026). Builds on 28 of its 40 references that are in the map. First posted on arXiv 13 Apr 2023.

Builds on

28 of 40 references.

References · 28

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2022Training language models to follow instructions with human feedbackcited by 15,971
  4. 2017Proximal Policy Optimization Algorithmscited by 15,215
  5. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  6. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  7. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  8. 2022Scaling Instruction-Finetuned Language Modelscited by 3,024
  9. 2016Asynchronous Methods for Deep Reinforcement Learningcited by 2,258
  10. 2019Fine-Tuning Language Models from Human Preferencescited by 2,150
  11. 2022Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agentscited by 1,035
  12. 2017Reinforcement Learning with Deep Energy-Based Policiescited by 669
16 more
  1. 2022MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledgecited by 478
  2. 2019On the Utility of Learning about Humans for Human-AI Coordinationcited by 349
  3. 2023Reward Design with Language Modelscited by 261
  4. 2020"Other-Play" for Zero-Shot Coordinationcited by 198
  5. 2019The Hanabi Challenge: A New Frontier for AI Researchcited by 197
  6. 2017Equivalence Between Policy Gradients and Soft Q-Learningcited by 173
  7. 2021Collaborating with Humans without Human Datacited by 142
  8. 2021Off-Belief Learningcited by 75
  9. 2022Semantic Exploration from Language Abstractions and Pretrained Representationscited by 51
  10. 2022Improving Intrinsic Exploration with Language Abstractionscited by 49
  11. 2022Mastering the Game of No-Press Diplomacy via Human-Regularized Reinforcement Learning and Planningcited by 48
  12. 2021ELLA: Exploration through Learned Language Abstractioncited by 42
  13. 2022K-level Reasoning for Zero-Shot Coordination in Hanabicited by 32
  14. 2022The Boltzmann Policy Distribution: Accounting for Systematic Suboptimality in Human Modelscited by 31
  15. 2021On the Critical Role of Conventions in Adaptive Human-AI Collaborationcited by 27
  16. 2022Using Natural Language and Program Abstractions to Instill Human Inductive Biases in Machinescited by 20

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 17; 2024: 23; 2025: 16; 2026: 7.

63 papers.

Sort:

By area:

  1. 2023VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Modelscited by 750
  2. 2023Language to Rewards for Robotic Skill Synthesiscited by 312
  3. 2023Large Language Models as General Pattern Machinescited by 197
  4. 2023LanguageMPC: Large Language Models as Decision Makers for Autonomous Drivingcited by 190
  5. 2023RoboCLIP: One Demonstration is Enough to Learn Robot Policiescited by 134
  6. 2024Survey on Large Language Model-Enhanced Reinforcement Learning: Concept, Taxonomy, and Methodscited by 116
  7. 2023Toward General-Purpose Robots via Foundation Models: A Survey and Meta-Analysiscited by 103
  8. 2024Large Language Models for Robotics: Opportunities, Challenges, and Perspectivescited by 89
  9. 2023Large Language Models as Generalizable Policies for Embodied Taskscited by 86
  10. 2023Vision-Language Models as a Source of Rewardscited by 66
  11. 2023Imitation Bootstrapped Reinforcement Learningcited by 64
  12. 2023Distilling and Retrieving Generalizable Knowledge for Robot Manipulation via Language Correctionscited by 45
38 more
  1. 2023Self-Refined Large Language Model as Automated Reward Function Designer for Deep Reinforcement Learning in Roboticscited by 36
  2. 2024True Knowledge Comes from Practice: Aligning LLMs with Embodied Environments via Reinforcement Learningcited by 36
  3. 2024PICLe: Eliciting Diverse Behaviors from Large Language Models with Persona In-Context Learningcited by 29
  4. 2023Can ChatGPT Enable ITS? The Case of Mixed Traffic Control via Reinforcement Learningcited by 20
  5. 2024The RL/LLM Taxonomy Tree: Reviewing Synergies Between Reinforcement Learning and Large Language Modelscited by 15
  6. 2026Co-Evolving LLM Decision and Skill Bank Agents for Long-Horizon Taskscited by 15
  7. 2024How Can LLM Guide RL? A Value-Based Approachcited by 13
  8. 2023Asking Before Acting: Gather Information in Embodied Decision Making with Language Modelscited by 12
  9. 2024Multi-Agent, Human-Agent and Beyond: A Survey on Cooperation in Social Dilemmascited by 12
  10. 2025M3HF: Multi-agent Reinforcement Learning from Multi-phase Human Feedback of Mixed Qualitycited by 11
  11. 2023Guide Your Agent with Adaptive Multimodal Rewardscited by 10
  12. 2024Towards Collaborative Intelligence: Propagating Intentions and Reasoning for Multi-Agent Coordination with Large Language Modelscited by 10
  13. 2025The Evolving Landscape of LLM- and VLM-Integrated Reinforcement Learningcited by 10
  14. 2023Toward Grounded Commonsense Reasoningcited by 9
  15. 2024A Survey On Enhancing Reinforcement Learning in Complex Environments: Insights from Human and LLM Feedbackcited by 9
  16. 2024Video2Reward: Generating Reward Function from Videos for Legged Robot Behavior Learningcited by 9
  17. 2025TeViR: Text-to-Video Reward with Diffusion Models for Efficient Reinforcement Learningcited by 8
  18. 2025FOUNDER: Grounding Foundation Models in World Models for Open-Ended Embodied Decision Makingcited by 8
  19. 2023Efficient Human-AI Coordination via Preparatory Language-based Conventioncited by 7
  20. 2024A Survey on Human-AI Collaboration with Large Foundation Modelscited by 7
  21. 2024Collaborative AI Teaming in Unknown Environments via Active Goal Deductioncited by 6
  22. 2024LORD: Large Models based Opposite Reward Design for Autonomous Drivingcited by 6
  23. 2024The Duality of Generative AI and Reinforcement Learning in Robotics: A Reviewcited by 6
  24. 2025VS-Bench: Evaluating VLMs for Strategic Abilities in Multi-Agent Environmentscited by 6
  25. 2024Efficient Language-instructed Skill Acquisition via Reward-Policy Co-Evolutioncited by 5
  26. 2024Safe Reinforcement Learning with Free-form Natural Language Constraints and Pre-Trained Language Modelscited by 4
  27. 2024Policy Learning with a Language Bottleneckcited by 4
  28. 2024MotIF: Motion Instruction Fine-tuningcited by 4
  29. 2025IPCGRL: Language-Instructed Reinforcement Learning for Procedural Level Generationcited by 4
  30. 2025LLM-Explorer: A Plug-in Reinforcement Learning Policy Exploration Enhancement Driven by Large Language Modelscited by 4
  31. 2024On Bits and Bandits: Quantifying the Regret-Information Trade-offcited by 3
  32. 2024Knowing What Not to Do: Leverage Language Model Insights for Action Space Pruning in Multi-agent Reinforcement Learningcited by 3
  33. 2025A Principle of Targeted Intervention for Multi-Agent Reinforcement Learningcited by 3
  34. 2026RDA: Reward Design Agent for Reinforcement Learningcited by 3
  35. 2024Enhancing Q-Learning with Large Language Model Heuristicscited by 2
  36. 2025Training Dialogue Systems by AI Feedback for Improving Overall Dialogue Impressioncited by 2
  37. 2025RoboInspector: Unveiling the Unreliability of Policy Code for LLM-enabled Robotic Manipulationcited by 2
  38. 2025Learning "Partner-Aware" Collaborators in Multi-Party Collaborationcited by 2

50 most cited of 63 All 63 (JSON)