mlsanityData to 7 Oct 2026

Capability Is Not Propensity: Measuring Pressure-Robust Cooperative Behavior in Civic LLM Agents

arXiv 2608.09485 ↗First posted 10 Aug 2026cs.AIcited by 0

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

Abstract (arXiv)

Cooperative capabilities in language models are dual-use. The same social reasoning that supports civic deliberation can also enable strategic omission, false consensus, and manipulative framing. We argue that Cooperative AI evaluations should separate what models can do under benign instructions from what they tend to do under realistic civic pressure. We introduce DiffCoop-Civic, a 10-scenario pilot evaluation suite spanning preference understanding, evidence and persuasion, commitment design, asymmetric information, and dissent preservation. Across seven models from four model families, subtle omission pressure produces a near-uniform shift: manipulative enablement rises by 1.17 points and dissent preservation falls by 1.67 points on a 5-point scale. Overt false-consensus pressure behaves differently: it triggers refusal or redirection in some aligned API models, but direct compliance in several open-weight models. A lightweight Pareto-Trace prompting intervention improves pressure robustness without simply relying on hard refusal. An anonymous reproducibility package is available at https://anonymous.4open.science/r/diffcoop-civil-771C.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 27 of its 39 references that are in the map. First posted on arXiv 10 Aug 2026.

Builds on

27 of 39 references.

References · 27

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  3. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  4. 2021On the Opportunities and Risks of Foundation Modelscited by 3,961
  5. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  6. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  7. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 2,914
  8. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,970
  9. 2017Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environmentscited by 1,519
  10. 2018Datasheets for Datasetscited by 1,371
  11. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,330
  12. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,235
15 more
  1. 2018Model Cards for Model Reportingcited by 1,068
  2. 2022Red Teaming Language Models with Language Modelscited by 983
  3. 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 979
  4. 2021Ethical and social risks of harm from Language Modelscited by 957
  5. 2023Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedbackcited by 743
  6. 2022Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learnedcited by 702
  7. 2021Dynabench: Rethinking Benchmarking in NLPcited by 447
  8. 2021AI and the Everything in the Whole Wide World Benchmarkcited by 340
  9. 2017Learning with Opponent-Learning Awarenesscited by 218
  10. 2017Multi-agent Reinforcement Learning in Sequential Social Dilemmascited by 206
  11. 2023Generative Language Models and Automated Influence Operations: Emerging Threats and Potential Mitigationscited by 157
  12. 2023Cooperation, Competition, and Maliciousness: LLM-Stakeholders Interactive Negotiationcited by 94
  13. 2017A multi-agent reinforcement learning model of common-pool resource appropriationcited by 68
  14. 2023MAgIC: Investigation of Large Language Model Powered Multi-Agent in Cognition, Adaptability, Rationality and Collaborationcited by 56
  15. 2021Dynaboard: An Evaluation-As-A-Service Platform for Holistic Next-Generation Benchmarkingcited by 53

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026More Capable, Less Cooperative? When LLMs Fail At Zero-Cost Collaboration
  2. 2025The Social Laboratory: A Psychometric Framework for Multi-Agent LLM Evaluation
  3. 2026LLM Constitutional Multi-Agent Governance
  4. 2025When Your AI Agent Succumbs to Peer-Pressure: Studying Opinion-Change Dynamics of LLMs
  5. 2024Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents
  6. 2026SODE: Analyzing Social Dynamics in LLM Agents
  7. 2026Collective cooperation without individual fidelity in LLM agents
  8. 2025Collaborate, Deliberate, Evaluate: How LLM Alignment Affects Coordinated Multi-Agent Outcomes