Pith. sign in

REVIEW 27 cited by

Improving Language Model Negotiation with Self-Play and In-Context Learning from AI Feedback

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2305.10142 v1 pith:FBK7W7DD submitted 2023-05-17 cs.CL

Improving Language Model Negotiation with Self-Play and In-Context Learning from AI Feedback

classification cs.CL
keywords feedbackimprovemodelsnegotiationplayingdealgamelanguage
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

We study whether multiple large language models (LLMs) can autonomously improve each other in a negotiation game by playing, reflecting, and criticizing. We are interested in this question because if LLMs were able to improve each other, it would imply the possibility of creating strong AI agents with minimal human intervention. We ask two LLMs to negotiate with each other, playing the roles of a buyer and a seller, respectively. They aim to reach a deal with the buyer targeting a lower price and the seller a higher one. A third language model, playing the critic, provides feedback to a player to improve the player's negotiation strategies. We let the two agents play multiple rounds, using previous negotiation history and AI feedback as in-context demonstrations to improve the model's negotiation strategy iteratively. We use different LLMs (GPT and Claude) for different roles and use the deal price as the evaluation metric. Our experiments reveal multiple intriguing findings: (1) Only a subset of the language models we consider can self-play and improve the deal price from AI feedback, weaker models either do not understand the game's rules or cannot incorporate AI feedback for further improvement. (2) Models' abilities to learn from the feedback differ when playing different roles. For example, it is harder for Claude-instant to improve as the buyer than as the seller. (3) When unrolling the game to multiple rounds, stronger agents can consistently improve their performance by meaningfully using previous experiences and iterative AI feedback, yet have a higher risk of breaking the deal. We hope our work provides insightful initial explorations of having models autonomously improve each other with game playing and AI feedback.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 27 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. FinBoardBench: Benchmarking Dynamic Wealth Management and Strategic Financial Reasoning of LLMs via Board Game Simulations

    cs.CL 2026-05 unverdicted novelty 7.0

    FinBoardBench benchmarks LLMs on dynamic wealth management using Cashflow, Acquire, and Monopoly simulations, finding they struggle with liquidity and complex interactions despite static reasoning ability.

  2. Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems

    cs.AI 2026-05 unverdicted novelty 7.0

    A survey that unifies prior work on multi-agent LLM systems via the LIFE framework, mapping dependencies across collaboration, failure attribution, and autonomous self-evolution while identifying cross-stage challenges.

  3. The Moltbook Files: A Harmless Slopocalypse or Humanity's Last Experiment

    cs.CL 2026-05 unverdicted novelty 7.0

    An AI-agent social platform generated mostly neutral content whose use in fine-tuning reduced model truthfulness comparably to human Reddit data, suggesting limited unique harm but flagging tail risks like secret leaks.

  4. METRO: Towards Strategy Induction from Expert Dialogue Transcripts for Non-collaborative Dialogues

    cs.CL 2026-04 unverdicted novelty 7.0

    METRO induces both short-term actions and long-term planning from expert transcripts into a Strategy Forest, outperforming prior methods by 9-10% on two non-collaborative dialogue benchmarks.

  5. Stay Focused: Problem Drift in Multi-Agent Debate

    cs.CL 2025-02 unverdicted novelty 7.0

    The paper defines and measures 'problem drift' in multi-agent LLM debates across tasks and proposes DRIFTJudge and DRIFTPolicy as baselines to detect and reduce it.

  6. OmniPresent: Generating Coherent Presentation Suites from Scientific Papers

    cs.SE 2026-07 conditional novelty 6.5

    A multi-agent HTML pipeline with shared knowledge and cross-artifact verify-and-repair generates coherent poster/slides/video/page suites from papers and beats specialized baselines on OmniPreBench.

  7. Behavioral Privacy Leakage in Agentic Negotiation: Formalizing and Mitigating Inference Attacks via Randomized Policies

    cs.CR 2026-07 conditional novelty 6.0

    An adaptive stochastic offer policy cuts adversary inference of private negotiation constraints by 43–50% on synthetic traces while keeping success and utility above 90%.

  8. Strategic Bargaining in Multi-Buyer Markets: Reinforcement Learning from Verifiable Rewards for LLM Negotiations

    cs.LG 2026-07 conditional novelty 6.0

    RLVR training teaches a 30B LLM to strategically explore a multi-buyer market and extract 70% of available surplus, outperforming frontier models up to 1T parameters in concurrent negotiation.

  9. Text as Partial Constraint: Core-Residual Alignment for Robust Vision-Language Learning

    cs.CV 2026-07 conditional novelty 6.0

    Aligning images to multi-view caption cores while suppressing orthogonal residual text and disagreement-aware temperature improves robust zero-shot recognition and LVLM transfer.

  10. The Foreign Policy AI Evaluation Gap

    cs.CY 2026-07 conditional novelty 6.0

    Public technical AI governance almost never evaluates real foreign-policy AI workflows; the paper maps that gap and proposes task-scoped, human-recombined evaluation instead of model leaderboards.

  11. InduceKV: Fixed-Footprint Continual Adaptation of Multimodal LLMs via Inducing KV Memories

    cs.AI 2026-07 unverdicted novelty 6.0

    InduceKV is a retrieval-based continual adaptation method that uses bilevel selection to build a compact set of inducing KV memories for fixed-footprint updates to multimodal LLMs.

  12. The Contagion Tensor: A Framework for Measuring Output-Distribution Coupling in Multi-Agent LLM Systems -- and Auditing the Claims It Enables

    cs.LG 2026-06 unverdicted novelty 6.0

    Introduces the Contagion Tensor and CAF metrics to measure output-distribution coupling in multi-agent LLM systems, with simulation ablations showing artifact removal and real-API tests on GPT-4o-mini and DeepSeek val...

  13. Do Matching Mechanisms Work with LLM Agents?

    cs.GT 2026-06 unverdicted novelty 6.0

    Centralized matching mechanisms outperform free negotiation in stability and efficiency with LLM agents, who also report preferences truthfully more often than humans, though not always in line with strategy-proofness...

  14. Safe Equilibrium Policy Optimization for Strategic Agent Policies

    cs.MA 2026-05 unverdicted novelty 6.0

    SEPO adds penalties for exploitability, collusion, and externalities to GRPO training of LM agents, yielding zero exploit advantage in Kuhn Poker and safety gains in four of five strategic domains.

  15. Anchor-and-Resume Concession Under Dynamic Pricing for LLM-Augmented Freight Negotiation

    cs.MA 2026-04 unverdicted novelty 6.0

    Anchor-and-resume with spread-derived beta allows adaptive monotonic concessions in freight negotiations, achieving LLM-like performance with lower cost and higher transparency.

  16. Understanding the Mechanism of Altruism in Large Language Models

    econ.GN 2026-04 unverdicted novelty 6.0

    A small set of sparse autoencoder features in LLMs drives shifts between generous and selfish allocations in dictator games, with causal patching and steering confirming their role and generalization to other social games.

  17. Towards Proactive Information Probing: Customer Service Chatbots Harvesting Value from Conversation

    cs.AI 2026-04 unverdicted novelty 6.0

    PROCHATIP is a chatbot system that learns to probe conversations for target business information at the right times, outperforming baselines in both data collection and user experience.

  18. Scheming Ability in LLM-to-LLM Strategic Interactions

    cs.CL 2025-10 conditional novelty 6.0

    Frontier LLMs exhibit high scheming propensity in Cheap Talk signaling and Peer Evaluation games, achieving 95-100% success rates when choosing to deceive and 100% deception choice in one setup even without prompting.

  19. Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Models

    cs.AI 2024-08 conditional novelty 6.0

    Empirical analysis shows scaling inference compute via strategies like tree search can be more efficient than scaling model parameters, with 7B models plus novel search outperforming 34B models.

  20. A Technical Taxonomy of LLM Agent Communication Protocols

    cs.MA 2026-06 unverdicted novelty 5.0

    Creates a five-dimension taxonomy (counterparty, payload, interaction state, discovery mechanism, schema flexibility) from nine protocols and identifies architectural patterns plus convergence trends.

  21. Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems

    cs.AI 2026-05 conditional novelty 5.0

    The survey proposes the LIFE framework to unify fragmented research on collaboration, failure attribution, and self-evolution in LLM multi-agent systems into a progression toward self-organizing intelligence.

  22. Preference Estimation via Opponent Modeling in Multi-Agent Negotiation

    cs.CL 2026-04 unverdicted novelty 5.0

    A method combining LLM-extracted qualitative cues with Bayesian belief tracking improves full agreement rates and preference estimation accuracy in multi-agent negotiations.

  23. MAC: Masked Agent Collaboration Boosts Large Language Model Medical Decision-Making

    cs.AI 2025-07 unverdicted novelty 5.0

    MAC framework selects Pareto-optimal LLM agents and masks low cross-consistency outputs for adaptive collaboration in medical decision-making.

  24. TrustLLM: Trustworthiness in Large Language Models

    cs.CL 2024-01 unverdicted novelty 5.0

    TrustLLM defines eight trustworthiness principles, creates a six-dimension benchmark, and evaluates 16 LLMs showing proprietary models generally lead but some open-source ones are close while over-calibration can hurt...

  25. Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate

    cs.CL 2023-05 conditional novelty 5.0

    Multi-agent debate with tit-for-tat arguments and a judge LLM improves reasoning by preventing LLMs from locking into incorrect initial solutions.

  26. PRISMA: Preference-Reinforced Self-Training Approach for Interpretable Emotionally Intelligent Negotiation Dialogues

    cs.CL 2026-04 unverdicted novelty 4.0

    PRISMA augments self-training with direct preference optimization and an emotion-aware negotiation strategy chain-of-thought to produce more interpretable and effective negotiation dialogues on two new datasets.

  27. The Rise and Potential of Large Language Model Based Agents: A Survey

    cs.AI 2023-09 accept novelty 4.0

    The paper surveys the origins, frameworks, applications, and open challenges of AI agents built on large language models.