mlsanityData to 7 Oct 2026

Inferring Rewards from Language in Context

arXiv 2204.02515 ↗First posted 5 Apr 2022cs.CL cs.AIcited by 50

Not on the map. Why?

Abstract (arXiv)

In classic instruction following, language like "I'd like the JetBlue flight" maps to actions (e.g., selecting that flight). However, language also conveys information about a user's underlying reward function (e.g., a general preference for JetBlue), which can allow a model to carry out desirable actions in new contexts. We present a model that infers rewards from language pragmatically: reasoning about how speakers choose utterances not only to elicit desired actions, but also to reveal information about their preferences. On a new interactive flight-booking task with natural language, our model more accurately infers rewards and predicts optimal actions in unseen environments, in comparison to past work that first maps language to actions (instruction following) and then maps actions to rewards (inverse reinforcement learning).

Key facts. Cited by 50 papers in the map (data through 7 Oct 2026). Builds on 23 of its 47 references that are in the map. First posted on arXiv 5 Apr 2022.

Builds on

23 of 47 references.

References · 23

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2019HuggingFace's Transformers: State-of-the-art Natural Language Processingcited by 2,201
  4. 2017Vision-and-Language Navigation: Interpreting visually-grounded navigation instructions in real environmentscited by 1,013
  5. 2019ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Taskscited by 773
  6. 2018Pyro: Deep Universal Probabilistic Programmingcited by 379
  7. 2018Speaker-Follower Models for Vision-and-Language Navigationcited by 337
  8. 2017Inverse Reward Designcited by 246
  9. 2020Reward-rational (implicit) choice: A unifying formalism for reward learningcited by 119
  10. 2017Mapping Instructions and Visual Observations to Actions with Reinforcement Learningcited by 80
  11. 2018Learning to Understand Goal Specifications by Modelling Rewardcited by 80
  12. 2019From Language to Goals: Inverse Reinforcement Learning for Vision-Based Instruction Followingcited by 67
11 more
  1. 2017Colors in Context: A Pragmatic Neural Model for Grounded Language Understandingcited by 57
  2. 2017Unified Pragmatic Models for Generating and Following Instructionscited by 55
  3. 2015Listen, Attend, and Walk: Neural Mapping of Navigational Instructions to Action Sequencescited by 50
  4. 2016Learning Language Games through Interactioncited by 43
  5. 2017From Language to Programs: Bridging Reinforcement Learning and Maximum Marginal Likelihoodcited by 41
  6. 2018Learning to Map Context-Dependent Sentences to Executable Formal Queriescited by 23
  7. 2015Alignment-based compositional semantics for instruction followingcited by 21
  8. 2017Accurately and Efficiently Interpreting Human-Robot Instructions of Varying Granularitiescited by 19
  9. 2017Translating Neuralesecited by 16
  10. 2018Situated Mapping of Sequential Instructions to Actions with Single-step Reward Observationcited by 12
  11. 2021Extending rational models of communication from beliefs to actionscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 6; 2023: 16; 2024: 15; 2025: 4; 2026: 9.

50 papers.

Sort:

By area:

  1. 2023Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedbackcited by 743
  2. 2024BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructionscited by 527
  3. 2023Language to Rewards for Robotic Skill Synthesiscited by 307
  4. 2023Reward Design with Language Modelscited by 257
  5. 2023LanguageMPC: Large Language Models as Decision Makers for Autonomous Drivingcited by 188
  6. 2023Grounding Language Models to Images for Multimodal Inputs and Outputscited by 129
  7. 2023Training Language Models with Language Feedback at Scalecited by 125
  8. 2024Building Machines that Learn and Think with Peoplecited by 73
  9. 2023Inverse Preference Learning: Preference-based RL without a Reward Functioncited by 68
  10. 2023Auto MC-Reward: Automated Dense Reward Design with Large Language Models for Minecraftcited by 60
  11. 2024Is this the real life? Is this just fantasy? The Misleading Success of Simulating Social Interactions With LLMscited by 59
  12. 2022Training Language Models with Language Feedbackcited by 48
38 more
  1. 2024Beyond Preferences in AI Alignmentcited by 46
  2. 2025Bayesian Teaching Enables Probabilistic Reasoning in Large Language Modelscited by 36
  3. 2024Bayesian Preference Elicitation with Language Modelscited by 32
  4. 2023Distilling Internet-Scale Vision-Language Models into Embodied Agentscited by 30
  5. 2023Decision-Oriented Dialogue for Human-AI Collaborationcited by 29
  6. 2024Pragmatic Instruction Following and Goal Assistance via Cooperative Language-Guided Inverse Planningcited by 28
  7. 2022Law Informs Code: A Legal Informatics Approach to Aligning Artificial Intelligence with Humanscited by 23
  8. 2022How to talk so AI will learn: Instructions, descriptions, and autonomycited by 22
  9. 2026Learning Personalized Agents from Human Feedbackcited by 16
  10. 2022Pragmatics in Language Grounding: Phenomena, Tasks, and Modeling Approachescited by 12
  11. 2024Computational Experiments Meet Large Language Model Based Agents: A Survey and Perspectivecited by 11
  12. 2024Preference-Conditioned Language-Guided Abstractioncited by 11
  13. 2024Problem Solving Through Human-AI Preference-Based Cooperationcited by 11
  14. 2026Value of Information: A Framework for Human-Agent Communicationcited by 10
  15. 2024Professional Agents -- Evolving Large Language Models into Autonomous Experts with Human-Level Competenciescited by 9
  16. 2023Large Language Models as Fiduciaries: A Case Study Toward Robustly Communicating With Artificial Intelligence Through Legal Standardscited by 8
  17. 2024Pragmatic Feature Preferences: Learning Reward-Relevant Preferences from Human Inputcited by 8
  18. 2024Video2Reward: Generating Reward Function from Videos for Legged Robot Behavior Learningcited by 8
  19. 2022Define, Evaluate, and Improve Task-Oriented Cognitive Capabilities for Instruction Generation Modelscited by 7
  20. 2024Mapping out the Space of Human Feedback for Reinforcement Learning: A Conceptual Frameworkcited by 5
  21. 2023FoMo Rewards: Can we cast foundation models as reward functions?cited by 4
  22. 2024Representational Alignment Supports Effective Machine Teachingcited by 3
  23. 2024Amortizing Pragmatic Program Synthesis with Rankingscited by 3
  24. 2023A Reminder of its Brittleness: Language Reward Shaping May Hinder Learning for Instruction Following Agentscited by 2
  25. 2025Learning Instruction-Following Policies through Open-Ended Instruction Relabeling with Large Language Modelscited by 2
  26. 2025Collaborative Rational Speech Act: Pragmatic Reasoning for Multi-Turn Dialogcited by 2
  27. 2026AdaptFuse: Training-Free Sequential Preference Learning via Externalized Bayesian Inferencecited by 2
  28. 2026Agentic Coding Needs Proactivity, Not Just Autonomycited by 2
  29. 2026Ask Now, Use Later: Benchmarking the Proactivity Gap in Long-Lived LLM Agentscited by 2
  30. 2022Overcoming Referential Ambiguity in Language-Guided Goal-Conditioned Reinforcement Learningcited by 1
  31. 2023Improve the efficiency of deep reinforcement learning through semantic exploration guided by natural languagecited by 1
  32. 2023Make a Donut: Hierarchical EMD-Space Planning for Zero-Shot Deformable Manipulation with Toolscited by 1
  33. 2026Learning to Learn from Language Feedback with Social Meta-Learningcited by 1
  34. 2023Amortizing Pragmatic Program Synthesis with Rankingscited by 0
  35. 2025NatSGLD: A Dataset with Speech, Gesture, Logic, and Demonstration for Robot Learning in Natural Human-Robot Interactioncited by 0
  36. 2026From Task Solving to Robust Real-World Adaptation in LLM Agentscited by 0
  37. 2026LEMUR: Learning to Align with Multi-Objective Reinforcement Learning from Preference Feedbackcited by 0
  38. 2026Bayesian Fine-tuning Yields Language Models that are as Bayesian as their Beliefs Allowcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2019Fine-Tuning Language Models from Human Preferences
  2. 2025Sailing by the Stars: A Survey on Reward Models and Learning Strategies for Learning from Rewards
  3. 2025Masked IRL: LLM-Guided Reward Disambiguation from Demonstrations and Language
  4. 2025Reward Model Interpretability via Optimal and Pessimal Tokens
  5. 2025RewardAnything: Generalizable Principle-Following Reward Models
  6. 2025Better Language Model-Based Judging Reward Modeling through Scaling Comprehension Boundaries
  7. 2025RewardBench 2: Advancing Reward Model Evaluation
  8. 2025QuickLAP: Quick Language-Action Preference Learning for Semi-Autonomous Agents