mlsanityData to 14 Oct 2026

Sentient Agent as a Judge: Evaluating Higher-Order Social Cognition in Large Language Models

arXiv 2505.02847 ↗First posted 1 May 2025cs.CL cs.AI cs.CYcited by 21

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

Abstract (arXiv)

Assessing how well a large language model (LLM) understands human, rather than merely text, remains an open challenge. To bridge the gap, we introduce Sentient Agent as a Judge (SAGE), an automated evaluation framework that measures an LLM's higher-order social cognition. SAGE instantiates a Sentient Agent that simulates human-like emotional changes and inner thoughts during interaction, providing a more realistic evaluation of the tested model in multi-turn conversations. At every turn, the agent reasons about (i) how its emotion changes, (ii) how it feels, and (iii) how it should reply, yielding a numerical emotion trajectory and interpretable inner thoughts. Experiments on 100 supportive-dialogue scenarios show that the final Sentient emotion score correlates strongly with Barrett-Lennard Relationship Inventory (BLRI) ratings and utterance-level empathy metrics, validating psychological fidelity. We also build a public Sentient Leaderboard covering 18 commercial and open-source models that uncovers substantial gaps (up to 4x) between frontier systems (GPT-4o-Latest, Gemini2.5-Pro) and earlier baselines, gaps not reflected in conventional leaderboards (e.g., Arena). SAGE thus provides a principled, scalable and interpretable tool for tracking progress toward genuinely empathetic and socially adept language agents.

Key facts. Cited by 21 papers in the map (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 56 of its 75 references that are in the map. First posted on arXiv 1 May 2025.

Builds on

56 of 75 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024PersonaGym: Evaluating Persona Agents and LLMs25 Jul
  2. 2024SocialBench: Sociality Evaluation of Role-Playing Conversational Agents20 Mar
  3. 2024CharacterEval: A Chinese Benchmark for Role-Playing Conversational Agent Evaluation2 Jan
  4. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate30 May

Other references · 52

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  3. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
  4. 2023Let's Verify Step by Stepcited by 3,789
  5. 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preferencecited by 1,184
  6. 2023WizardLM: Empowering large pre-trained language models to follow complex instructionscited by 1,125
  7. 2023Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotationscited by 928
  8. 2023AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedbackcited by 748
  9. 2023ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debatecited by 742
  10. 2023RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedbackcited by 625
  11. 2024Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Modelscited by 557
  12. 2023Orca: Progressive Learning from Complex Explanation Traces of GPT-4cited by 379
38 more
  1. 2023SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agentscited by 367
  2. 2023JudgeLM: Fine-tuned Large Language Models are Scalable Judgescited by 367
  3. 2023PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimizationcited by 316
  4. 2023Large Language Models are Effective Text Rankers with Pairwise Ranking Promptingcited by 307
  5. 2021Towards Emotional Support Dialog Systemscited by 299
  6. 2024Agent-as-a-Judge: Evaluate Agents with Agentscited by 238
  7. 2022Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMscited by 220
  8. 2024V-STaR: Training Verifiers for Self-Taught Reasonerscited by 217
  9. 2024Meta-Rewarding Language Models: Self-Improving Alignment with LLM-as-a-Meta-Judgecited by 204
  10. 2023InCharacter: Evaluating Personality Fidelity in Role-Playing Agents through Psychological Interviewscited by 179
  11. 2023FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactionscited by 171
  12. 2023Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Modelscited by 163
  13. 2024Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluatorscited by 149
  14. 2024InFoBench: Evaluating Instruction Following Ability in Large Language Modelscited by 135
  15. 2023Branch-Solve-Merge Improves Large Language Model Evaluation and Generationcited by 106
  16. 2024On scalable oversight with weak LLMs judging strong LLMscited by 99
  17. 2024OffsetBias: Leveraging Debiased Data for Tuning Evaluatorscited by 94
  18. 2024SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agentscited by 93
  19. 2023EQ-Bench: An Emotional Intelligence Benchmark for Large Language Modelscited by 80
  20. 2024Roleplay-doh: Enabling Domain-Experts to Create LLM-simulated Patients via Eliciting and Adhering to Principlescited by 73
  21. 2023HI-TOM: A Benchmark for Evaluating Higher-Order Theory of Mind Reasoning in Large Language Modelscited by 63
  22. 2024Enhancing LLM Reasoning via Critique Models with Test-Time and Training-Time Supervisioncited by 60
  23. 2023Emotional Intelligence of Large Language Modelscited by 58
  24. 2023Rehearsal: Simulating Conflict to Teach Conflict Resolutioncited by 55
  25. 2024EmoBench: Evaluating the Emotional Intelligence of Large Language Modelscited by 53
  26. 2020Knowledge Bridging for Empathetic Dialogue Generationcited by 50
  27. 2025SPC: Evolving Self-Play Critic via Adversarial Games for LLM Reasoningcited by 42
  28. 2024AgentSense: Benchmarking Social Intelligence of Language Agents through Interactive Scenarioscited by 40
  29. 2023CharacterGLM: Customizing Chinese Conversational AI Characters with Large Language Modelscited by 34
  30. 2023Training Models to Generate, Recognize, and Reframe Unhelpful Thoughtscited by 31
  31. 2025From Personas to Talks: Revisiting the Impact of Personas on LLM-Synthesized Emotional Support Conversationscited by 23
  32. 2022CASE: Aligning Coarse-to-Fine Cognition and Affection for Empathetic Response Generationcited by 18
  33. 2023Facilitating Multi-turn Emotional Support Conversation with Positive Emotion Elicitation: A Reinforcement Learning Approachcited by 15
  34. 2024Social Skill Training with Large Language Modelscited by 15
  35. 2025Self-Consistency of the Internal Reward Models Improves Self-Rewarding Language Modelscited by 13
  36. 2025Dancing with Critiques: Enhancing LLM Reasoning with Stepwise Natural Language Self-Critiquecited by 13
  37. 2024Academically intelligent LLMs are not necessarily socially intelligentcited by 12
  38. 2025Crisp: Cognitive Restructuring of Negative Thoughts through Multi-turn Supportive Dialoguescited by 11

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 16.

21 papers.

Sort:

By area:

  1. 2025RLVER: Reinforcement Learning with Verifiable Emotion Rewards for Empathetic Agentscited by 25
  2. 2026Agent-as-a-Judgecited by 17
  3. 2025Too Good to be Bad: On the Failure of LLMs to Role-Play Villainscited by 11
  4. 2026Reward Modeling from Natural Language Human Feedbackcited by 11
  5. 2026Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learningcited by 9
  6. 2025Enhancing Large Language Model Reasoning with Reward Models: An Analytical Surveycited by 8
  7. 2025RecGPT-V2 Technical Reportcited by 8
  8. 2026Think Fast and Slow: Step-Level Cognitive Depth Adaptation for LLM Agentscited by 5
  9. 2026EMPA: Evaluating Persona-Aligned Empathy as a Processcited by 4
  10. 2026MICA: Multi-granularity Intertemporal Credit Assignment for Long-Horizon Emotional Support Dialoguecited by 3
  11. 2026EIBench: A Simulator-Based Benchmark and Turn-Credit RL for Emotion Managementcited by 3
  12. 2025PEER: Unified Process-Outcome Reinforcement Learning for Structured Empathetic Reasoningcited by 2
9 more
  1. 2026VLM-Guided Group Preference Alignment for Diffusion-based Human Mesh Recoverycited by 2
  2. 2026COGTRL: Training LLMs for Scientific Discovery Assistance using Cognitive Traces via Reinforcement Learningcited by 1
  3. 2026Can You Break RLVER? Probing Adversarial Robustness of RL-Trained Empathetic Agentscited by 0
  4. 2026User-Aware Active Knowledge Acquisition for Emotional Support Dialoguecited by 0
  5. 2026The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modescited by 0
  6. 2026Towards High-Level Semantic Intelligencecited by 0
  7. 2026R3-Bench: LLMs Struggle with Resource-Rational Reasoning under Shared Budgetscited by 0
  8. 2026TRACER: Trajectory-Aligned Learning for Multi-Turn User Simulationcited by 0
  9. 2026Evolving Support Priorities in Empathetic Reinforcement Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025ESC-Judge: A Framework for Comparing Emotional Support Conversational Agents
  2. 2025AI shares emotion with humans across languages and cultures
  3. 2025MetaMind: Modeling Human Social Thoughts with Metacognitive Multi-Agent Systems
  4. 2026Emotional intelligence in large language models is fragmented across perception, cognition, and interaction
  5. 2025Emergence of Hierarchical Emotion Organization in Large Language Models
  6. 2025PersuasiveToM: A Benchmark for Evaluating Machine Theory of Mind in Persuasive Dialogues
  7. 2026Sentipolis: Emotion-Aware Agents for Social Simulations
  8. 2024OpenToM: A Comprehensive Benchmark for Evaluating Theory-of-Mind Reasoning Capabilities of Large Language Models