用置信度分层验证，提升政治回避回答识别准确率
AsymVerify at SemEval-2026 Task 6: Asymmetric Confidence-Gated Verification for Political Evasion Detection
- 根据置信度选择性启用升级或降级验证，动态调整判断
- 在237条测试数据上达0.85宏平均F1，排名第二
- 适合需要高精度检测模糊回应的舆情分析场景
政治回避难以检测，因其回答表面合作却回避实质承诺。我们提出AsymVerify，一种用于SemEval-2026任务6的置信度门控验证系统，该任务为清晰回复、模糊、清晰不回复三分类。AsymVerify在评估集（D_eval, n=237）上取得0.85宏平均F1，位列41支队伍中的第2名。系统先对每对问答进行分类，再对低置信预测选择性执行降级验证（清晰回复/清晰不回复 → 模糊）或升级验证（模糊 → 清晰回复）。开发集分析显示错误集中于模糊边界两侧，由此设计不对称双验证机制；置信度门控使额外推理成本保持在1.48次调用/样本。在开发集（D_dev, n=308）上，使用GLM-4.7的AsymVerify相比单次分类提升+17.1宏平均F1；仅升级验证器即在所有测试的大语言模型后端上带来+6.8至+15.2的宏平均F1提升。代码已开源：https://github.com/kaons-research/AsymVerify-ACL。
原文摘要 · Abstract (English)
Political evasion is difficult to detect because evasive answers often appear cooperative while avoiding concrete commitment. We present AsymVerify, a confidence-gated verification system for SemEval-2026 Task 6, a three-way classification of Clear Reply, Ambivalent, and Clear Non-Reply responses. AsymVerify scored 0.85 Macro F1 on the evaluation split (D_eval, n=237), placing 2nd out of 41 teams on the official leaderboard. The system first classifies each question-answer pair, then selectively applies downgrade verification (CR/CNR -> AMB) or upgrade verification (AMB -> CR) to low-confidence predictions. Development analysis shows that errors concentrate at the Ambivalent boundary in both directions, motivating this asymmetric two-verifier design while confidence gating keeps additional inference cost low. On D_dev (n=308), AsymVerify with GLM-4.7 gains +17.1 Macro F1 over single-pass classification at 1.48 calls/example, and the upgrade verifier alone improves every tested LLM backend on D_dev by +6.8 to +15.2 Macro F1 over its single-pass baseline. Code is available at https://github.com/kaons-research/AsymVerify-ACL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。