mlsanityData to 14 Oct 2026

What Makes a Social Agent's Interactions Effective? Learning from Verifiable Outcomes in Paired Worlds

arXiv 2610.20195 ↗First posted 14 Oct 2026cs.CL cs.AI cs.SIcited by 0

On the map: LLM agents and tool use → Social simulation and multi-agent debate

Abstract (arXiv)

Effective social interaction requires more than fluent responses: agents must reach plans that fit the participants' needs and circumstances. Existing evaluations of social agents mainly rely on LLM judges that rate the conversation, without checking the resulting plan. We introduce HICO-Bench, a benchmark of 180 everyday coordination scenarios, such as resource allocation, scheduling, and authorization, in which two agents with private information converse and then independently record the plan they agreed on. A program checks Feasibility (the two records match and satisfy both participants' private conditions), and an LLM judge checks Integrity (the conversation supports the recorded plan); passing both is a Verified Success. Each scenario has two versions that share the same public context but differ in one private condition, so the correct plan differs between them. Across thirteen LLMs, 36% of conversations rated highly for goal completion fail Verified Success. To learn from outcomes that arrive only at the end of a conversation, we propose Paired-Credit Policy Optimization (PaCPO). After each turn, PaCPO asks the agent for a tentative plan, checks it in both versions of the scenario, and rewards the turns after which the tentative plans fit their respective versions. PaCPO raises Qwen3-8B's Verified Success from 20.4% to 28.3%.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 44 of its 55 references that are in the map. First posted on arXiv 14 Oct 2026.

Builds on

44 of 55 references.

References · 44

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  4. 2025Qwen3 Technical Reportcited by 8,820
  5. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  6. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  7. 2024GPT-4o System Cardcited by 4,588
  8. 2023Let's Verify Step by Stepcited by 3,789
  9. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
  10. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 2,026
  11. 2024τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domainscited by 1,402
  12. 2025Understanding R1-Zero-Like Training: A Critical Perspectivecited by 1,345
32 more
  1. 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 1,006
  2. 2018MultiWOZ -- A Large-Scale Multi-Domain Wizard-of-Oz Dataset for Task-Oriented Dialogue Modellingcited by 766
  3. 2025τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environmentcited by 627
  4. 2025Group-in-Group Policy Optimization for LLM Agent Trainingcited by 497
  5. 2023SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agentscited by 367
  6. 2017Deal or No Deal? End-to-End Learning for Negotiation Dialoguescited by 209
  7. 2020MultiWOZ 2.2 : A Dialogue Dataset with Additional Annotation Corrections and State Tracking Baselinescited by 204
  8. 2024VinePPO: Refining Credit Assignment in RL Training of LLMscited by 183
  9. 2018RUDDER: Return Decomposition for Delayed Rewardscited by 173
  10. 2023FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactionscited by 171
  11. 2025Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systemscited by 126
  12. 2024STaR-GATE: Teaching Language Models to Ask Clarifying Questionscited by 104
  13. 2024SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agentscited by 93
  14. 2021CaSiNo: A Corpus of Campsite Negotiation Dialogues for Automatic Negotiation Systemscited by 62
  15. 2024CLAMBER: A Benchmark of Identifying and Clarifying Ambiguous Information Needs in Large Language Modelscited by 54
  16. 2025Agentic Reinforcement Learning with Implicit Step Rewardscited by 47
  17. 2024OSCAR: Operating System Control via State-Aware Reasoning and Re-Planningcited by 38
  18. 2026Beyond Trajectory-Level Attribution: Graph-Based Credit Assignment for Agentic Reinforcement Learningcited by 31
  19. 2025System-1.5 Reasoning: Traversal in Language and Latent Spaces with Dynamic Shortcutscited by 27
  20. 2025Sotopia-RL: Reward Design for Social Intelligencecited by 24
  21. 2026When and What to Ask: AskBench and Rubric-Guided RLVR for LLM Clarificationcited by 16
  22. 2025Towards Interactive Intelligence for Digital Humanscited by 10
  23. 2025Multi-speaker Attention Alignment for Multimodal Social Interactioncited by 9
  24. 2025Can MLLMs Read the Room? A Multimodal Benchmark for Verifying Truthfulness in Multi-Party Social Interactionscited by 6
  25. 2025Can MLLMs Read the Room? A Multimodal Benchmark for Assessing Deception in Multi-Party Social Interactionscited by 6
  26. 2026Vinci2: Providing Proactive Assistance in Continuous Egocentric Videoscited by 5
  27. 2026SAVOIR: Learning Social Savoir-Faire via Shapley-based Reward Attributioncited by 4
  28. 2026SocialDirector: Training-Free Social Interaction Control for Multi-Person Video Generationcited by 4
  29. 2026SOTOPIA-TOM: Evaluating Privacy and Information Management in Multi-Agent Interaction with Theory of Mindcited by 3
  30. 2026GROVE: Growing and Reasoning over Temporally Stratified Memory from Streaming Video Experiencecited by 2
  31. 2026Social Gym and SPaRTan: Benchmarking and Improving LLM Social Reasoning via Multi-Agent Game Tournamentscited by 2
  32. 2026Perception or Prejudice: Can MLLMs Go Beyond First Impressions of Personality?cited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024AgentSense: Benchmarking Social Intelligence of Language Agents through Interactive Scenarios
  2. 2026When LLM Agents Fail to Read the Room: ReAdapt for Relational Social Reasoning
  3. 2026SocialRL: Refining LLMs' Social Intelligence through Multi-turn Reinforcement Learning and Reward Design
  4. 2026Verifiable Social Reasoning for LLM Assistants
  5. 2026M3-BENCH: Process-Aware Evaluation of LLM Agents' Social Behaviors in Mixed-Motive Games
  6. 2025The Stackelberg Speaker: Optimizing Persuasive Communication in Social Deduction Games
  7. 2025SPIN-Bench: How Well Do LLMs Plan Strategically and Reason Socially?
  8. 2025LLMs Can't Handle Peer Pressure: Crumbling under Multi-Agent Social Interactions