English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Agents built with large language models (LLMs) have recently achieved greatadvancements. However, most of the efforts focus on single-agent or cooperativesettings, leaving more general multi-agent environments underexplored. Wepropose a new framework powered by reinforcement learning (RL) to developstrategic language agents, i.e., LLM-based agents with strategic thinkingability, for a popular language game, Werewolf. Werewolf is a social deductiongame with hidden roles that involves both cooperation and competition andemphasizes deceptive communication and diverse gameplay. Our agent tackles thisgame by first using LLMs to reason about potential deceptions and generate aset of strategically diverse actions. Then an RL policy, which selects anaction from the candidates, is learned by population-based training to enhancethe agents' decision-making ability. By combining LLMs with the RL policy, ouragent produces a variety of emergent strategies, achieves the highest win rateagainst other LLM-based agents, and stays robust against adversarial humanplayers in the Werewolf game.

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