Dancing in Chains: Strategic Persuasion in Academic Rebuttal via Theory of Mind

Dimensions overall score 8.8

GitHub Code Pulse

No public code linked for this paper yet.

• RebuttalAgent is the first framework to ground academic rebuttal in Theory of Mind (ToM).method 95%
• RebuttalAgent uses a ToM-Strategy-Response (TSR) pipeline that models reviewer mental state, formulates persuasion strategy, and generates strategy-grounded response.method 95%
• The training process consists of two stages: supervised fine-tuning followed by reinforcement learning with a self-reward mechanism.method 95%
• RebuttalAgent significantly outperforms the base model by an average of 18.3% on automated metrics.result 90%
• RebuttalAgent outperforms advanced proprietary models across both automated and human evaluations.result 90%
• The Rebuttal-RM evaluator achieves scoring consistency with human preferences surpassing powerful judge GPT-4.1.result 90%
• The RebuttalBench dataset was constructed via a novel critique-and-refine approach and contains over 100K samples.technical 90%
• The generated rebuttal content is for reference only and not intended to replace the author's own critical analysis and response.limitation 95%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card