Research map / Multi-agent systems and games / LLM agents
CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society
Guohao Li, Hasan Abed Al Kader Hammoud, Hani Itani, Dmitrii Khizbullin, Bernard Ghanem
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| role-playing two-agent cooperation instead of single-shot answer | gpt-3.5-turbo · AI Society | human preference win rate: 10.4% vs 76.3% | Improved |
| role-playing two-agent cooperation instead of single-shot answer | gpt-3.5-turbo · Code | GPT4 win rate: 24.0% vs 76.0% | Improved |
| fine-tune on CAMEL-generated datasets | LLaMA-7B · HumanEval | pass@1: 10.5 -> 14.0 | Improved |
| add Code dataset to AI Society fine-tuning (ablation) | LLaMA-7B · Code tasks (GPT4 judge) | pairwise wins: 9 vs 10 wins | Improved |
| add Math dataset to AI Society+Code fine-tuning (ablation) | LLaMA-7B · Math tasks (GPT4 judge) | pairwise wins: 3 vs 16 wins | Improved |
| add Science dataset to AI Society+Code+Math (ablation) | LLaMA-7B · Math tasks (GPT4 judge) | pairwise wins: 10 draws, 5 vs 5 | No clear effect |
| add Code dataset to AI Society fine-tuning (ablation) | LLaMA-7B · Math tasks (GPT4 judge) | pairwise wins: 8 vs 7 wins | Hurt |
Closest papers by meaning
- $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment (ICML 2026 · similarity 0.87)
- Cooper: Coordinating Specialized Agents towards a Complex Dialogue Goal (AAAI 2024 · similarity 0.87)
- Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action (NeurIPS 2026 · similarity 0.86)
- SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents (ICLR 2024 · similarity 0.86)
- Lemur: Harmonizing Natural Language and Code for Language Agents (ICLR 2024 · similarity 0.86)
- LLawCo: Learning Laws of Cooperation for Modeling Embodied Multi-Agent Behavior (ICML 2026 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
