Pith. sign in

REVIEW 13 cited by

Language Agents with Reinforcement Learning for Strategic Play in the Werewolf Game

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 2310.18940 v4 pith:35TESAL5 submitted 2023-10-29 cs.AI cs.LGcs.MA

classification cs.AIcs.LGcs.MA
keywords agentslanguagegamestrategicactionsbiasdecision-makingintrinsic
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Agents built with large language models (LLMs) have shown great potential across a wide range of domains. However, in complex decision-making tasks, pure LLM-based agents tend to exhibit intrinsic bias in their choice of actions, which is inherited from the model's training data and results in suboptimal performance. To develop strategic language agents, i.e., agents that generate flexible language actions and possess strong decision-making abilities, we propose a novel framework that powers LLM-based agents with reinforcement learning (RL). We consider Werewolf, a popular social deduction game, as a challenging testbed that emphasizes versatile communication and strategic gameplay. To mitigate the intrinsic bias in language actions, our agents use an LLM to perform deductive reasoning and generate a diverse set of action candidates. Then an RL policy trained to optimize the decision-making ability chooses an action from the candidates to play in the game. Extensive experiments show that our agents overcome the intrinsic bias and outperform existing LLM-based agents in the Werewolf game. We also conduct human-agent experiments and find that our agents achieve human-level performance and demonstrate strong strategic play.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 13 Pith papers

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

  1. Training with (Swap) Regret Loss in a Single-Layer Self-Attention Model: A Case Study on the Probability Simplex

    cs.LG 2026-07 conditional novelty 7.0 of 10

    Training single-layer attention with squared regret loss has stationary points that implement smoothed fictitious play (external regret) and, via a new swap-regret loss, the Blum–Mansour no-swap-regret algorithm.

  2. Auditing Belief-Conditioned LLM Agents in Hidden-Information Social Deduction Games

    cs.MA 2026-07 conditional novelty 6.0 of 10

    An external belief audit framework for LLM Werewolf agents associates active belief with higher good-side win rates while exposing low action-belief consistency and rejecting forced consumption.

  3. Social Networks of LLM Agents

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Attention width and source social power determine whether LLM agent networks herd or achieve wisdom-of-crowds, with a pricing equalizer restoring optimal collective weights.

  4. GasAgent: A Multi-Agent Framework for Automated Gas Optimization in Smart Contracts

    cs.AI 2025-07 conditional novelty 6.0 of 10

    A four-agent LLM pipeline retrieves known Solidity gas-waste patterns, proposes new ones, and automatically verifies and applies the fixes, saving about 10% deployment gas on 82% of real contracts.

  5. Strategy Adaptation in Large Language Model Werewolf Agents

    cs.CL 2025-07 conditional novelty 6.0 of 10

    Dynamically switching between Support and Attack strategies based on role estimates raises win rates for Werewolf LLM agents, with mixed effects for Villagers.

  6. AutomataGPT: Forecasting and Ruleset Inference for Two-Dimensional Cellular Automata

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A transformer pretrained on 100 cellular automaton rules forecasts unseen rules at 98.5% one-step accuracy and infers new rules with up to 96% functional accuracy.

  7. DipLLM: Fine-Tuning LLM for Strategic Decision-making in Diplomacy

    cs.AI 2025-06 conditional novelty 6.0 of 10

    A fine-tuned LLM with a unit-by-unit action decomposition outperforms prior Diplomacy agents while using far less training data.

  8. Decompose, Plan in Parallel, and Merge: A Novel Paradigm for Large Language Models based Planning with Multiple Constraints

    cs.CL 2025-06 conditional novelty 6.0 of 10

    DPPM, a decompose-plan-in-parallel-and-merge framework with verify-and-refine feedback, improves final pass rates on travel-planning benchmarks over Direct, CoT, and LLM-Modulo.

  9. Herd Behavior: Investigating Peer Influence in LLM-based Multi-Agent Systems

    cs.MA 2025-05 conditional novelty 6.0 of 10

    LLM agents flip their answers more when their own confidence is low and their peer seems confident, and the format and order of peer information can amplify or dampen this herd behavior.

  10. CoMet: Metaphor-Driven Covert Communication for Multi-Agent Language Games

    cs.CL 2025-05 conditional novelty 6.0 of 10

    CoMet couples a hypothesis-testing metaphor reasoner with a self-improving metaphor generator, and the resulting LLM agents win more often in metaphor-heavy language games.

  11. Agent-to-Agent Theory of Mind: Testing Interlocutor Awareness among Large Language Models

    cs.CL 2025-06 conditional novelty 5.0 of 10

    LLMs show measurable interlocutor awareness: they identify same-family models well and adapt behavior when told who they are talking to, which helps cooperation but raises alignment and safety risks.

  12. Divide-Fuse-Conquer: Eliciting "Aha Moments" in Multi-Scenario Games

    cs.LG 2025-05 reject novelty 5.0 of 10

    A group, fuse, and retrain recipe for multi-game reinforcement learning lets a 32B model reach near-Claude3.5 performance on several TextArena games, though the headline score is internally inconsistent.

  13. Verbal Werewolf: Engage Users with Verbalized Agentic Werewolf Game Framework

    cs.CL 2025-05 reject novelty 4.0 of 10

    A system that voices LLM-driven Werewolf agents in near real time using parallel gameplay and TTS pipelines, with only anecdotal evaluation.

Pith tools