# Chinese Competitive Debating Dataset and Benchmark

arXiv: https://arxiv.org/abs/2609.21637 · HTML: https://mlsanity.com/paper/2609.21637/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 12 references that are in the map.
- First posted on arXiv 18 Sep 2026.

## Builds on (most cited first, 6 of 6)

- [Debatrix: Multi-dimensional Debate Judge with Iterative Chronological Analysis Based on LLM](https://mlsanity.com/paper/2403.08010/) (2024)
- [ORCHID: A Chinese Debate Corpus for Target-Independent Stance Detection and Argumentative Dialogue Summarization](https://mlsanity.com/paper/2410.13667/) (2024)
- [DebateBench: A Challenging Long Context Reasoning Benchmark For Large Language Models](https://mlsanity.com/paper/2502.06279/) (2025)
- [Debatable Intelligence: Benchmarking LLM Judges via Debate Speech Evaluation](https://mlsanity.com/paper/2506.05062/) (2025)
- [Conch: Competitive Debate Analysis via Visualizing Clash Points and Hierarchical Strategies](https://mlsanity.com/paper/2507.14482/) (2025)
- [DEFINED: A Data-Efficient Computational Framework for Fine-Grained Creativity Assessment in Debate Scenarios](https://mlsanity.com/paper/2606.07226/) (2026)

## Abstract (arXiv)

Debate adjudication requires tracking how arguments develop through interaction, yet existing datasets rarely combine fine-grained debate transcripts with professional judgments collected during real competitions under a shared rubric. We introduce a dataset and benchmark for evaluating large language models' understanding of competitive Chinese-language debate at the match, stage, and speaker levels. We organized 182 matches and recruited 120 professional judges, with each match independently adjudicated by three judges using a predefined rubric. After excluding matches with incomplete records, the dataset contains 148 matches, 2,698 stages, and 20,542 exchange units, with manually verified transcripts and segmentation. It preserves original stage scores, match votes, best-debater ballots, and adjudication rationales. We define three tasks: winner-tendency prediction, stage-score prediction, and best-debater prediction. Zero-shot evaluation of multiple large language models yields a highest winner-prediction accuracy of 66.2%, a highest Pearson correlation of 0.250 between model stage scores and mean human ratings, and a highest best-debater prediction accuracy of 56.8%. The dataset and benchmark provide a testbed for studying large language models' understanding of interactive argumentation and their agreement with professional judges.