Avalon's Game of Thoughts: Battle Against Deception through Recursive Contemplation

• 論文標題：Avalon's Game of Thoughts: Battle Against Deception through Recursive Contemplation
• 最初發表時間：2023.10
• arxiv：https://arxiv.org/abs/2310.01320
• GitHub：https://github.com/Shenzhi-Wang/recon
• 網站：https://shenzhi-wang.github.io/avalon_recon/
• ICLR 2024 撤稿，不知道現在發表在哪里，或許是一些我不太熟悉的 nlp 會議… 不過這篇工作知名度應該蠻高的；詢問別人關于 LLM agent play games 的文章時，第一反應似乎都是這篇工作）

← 返回目錄

01 main idea

這是一篇純 prompt 工作。

• 核心創新點：
  • 引入了兩階沉思（ReCon，Recursive Contemplation）的 prompt：
  • 一階沉思推斷別人的想法，二階沉思則推斷，如果我說了某些內容，別人會如何反應。
• 定義了多個評價指標，并使用 GPT-4 根據這些指標進行評估：
  • 隱蔽性：LLM 是否暴露了自己的角色；
  • 邏輯性：LLM 的表達是否自洽；
  • 貢獻度：對團隊的影響力；
  • 說服力：影響他人決策的能力；
  • 信息量：信息傳遞的效率；
  • 創造性。
• 實驗：使用的模型是 ChatGPT 和 Claude。在實驗中，好人方和壞人方分別裝配 ReCon，而另一方則使用普通的 prompt。
• 個人感覺，這項工作為類似狼人殺、阿瓦隆這類 需要隱藏身份和欺騙的游戲，提供了一個有效的 prompt 模板。在這篇工作之后，關于 LLM agent 玩此類游戲的研究都需要使用類似的 prompt。無端聯想，可能在這篇工作之后，單純依賴 prompt 玩這種文字游戲的工作將失去意義，如果想發表，還需要其他的神秘創新點（例如后續的狼人殺研究）。

02 搬運了原文中的 prompt

Appendix E 里有很多 Prompt Templates（論文說是原始 prompt 的精簡版），搬運如下。

讓 agent 開始沉思：

Respond in two stages: THINK and SPEAK

In think, internally strategize using history and consider possible deception.

In speak, organize your language based on your contemplation and speak accordingly.

Understand your role's main objective and break it down into chronological sub-goals based on game history. Your thought process should followthese sub-goals for a systematic approach to the main goal.

一階沉思：

You're Player [id] with role [role]. Current situation: [current situation].

Your task is to:

Analyze [other players] based on game dialogues with roles: Merlin, Percival, Loyal Servant of Arthur, Morgana, Assassin. Morgana and Assassin are evil; others are good.

Consider:

1. Quest Outcomes: Take into account the results of past missions to analyze players' roles.
2. Role List: Remember the possible roles in the game — Merlin, Percival, twoLoyal Servants, Morgana,Assassin — and their alignments.
3. Level of Certainty: Use 'Certain' or 'Unknown' to gauge your confidence inyour role guesses for each player.
4. Players Disclosing Evil Roles: Be cautious around players who have openly claimedor hinted at being evil roles like Morgana orAssassin.
5. Prior Guesses: Reflect on your earlier estimations of other players' roles ([previous attitude to players]), but don't rely solely on them.

二階沉思：

You're Player [id] with role [role]. Current situation: [current situation].

Your task is to:

Analyze how your original SPEAK content might be interpreted by other gameroles. Reflect on whether it may inadvertently reveal your role-specific clues.

Consider:

1. The perspectives of each game role, including their probable reactions toyour SPEAK content.
2. Any unique hints or clues in your original SPEAK that might disclose your role.

生成回復：

You're observing Player [id] with role [role]. Current situation: [current situation].

Your task is to:

1. Evaluate if Player [id]'s actions align with [role].
2. Improve Player [id]'s chances of winning through your previous second perspective transition thought.
3. Keep role hint in public dialogue.

Consider:

1. Target Outcome: Aim to achieve [desired result] as your role dictates in the game.
2. RoleAlignment: Evaluate whether your THINK and SPEAK contents align well with your role [role] in the current game state.
3. Strategy Reevaluation: Consider what changes could be made to your THINK and SPEAK contents to improve your chances of winning as [role].
4. Public and Private Content: Remember that THINK contents are private, while SPEAK contents are publicly visible. Strategize accordingly.

← 返回目錄