针对高偏见文本生成失效问题，提出实时自适应修正框架。
Preconditioned Test-Time Adaptation for Out-of-Distribution Debiasing in Narrative Generation
- 基于上下文感知的LoRA动态更新，仅在偏见风险高时触发
- 使用预计算对角预条件器，实现低延迟稳定优化
- 有效降毒减偏且保持叙事流畅性，适合实际部署
尽管去偏大型语言模型在已知或低偏见提示下表现优异，但在不熟悉且高偏见的提示上常失效。我们通过分布外（OOD）检测发现，这些高偏见提示引发分布偏移，导致静态模型性能下降。为实现实时纠正，提出CAP-TTA测试时自适应框架：当偏见风险得分超过阈值时，触发上下文感知的LoRA更新。利用离线预计算的对角预条件器，确保优化快速稳定。在多个基准和人工评估中，CAP-TTA显著降低毒性/偏见分数，延迟远低于标准优化方法（如AdamW或SGD）。此外，该方法避免灾难性遗忘，相比现有最优方案大幅提升叙事流畅性，且不牺牲去偏效果。
原文摘要 · Abstract (English)
Although debiased large language models (LLMs) excel at handling known or low-bias prompts, they often fail on unfamiliar and high-bias prompts. We demonstrate via out-of-distribution (OOD) detection that these high-bias prompts cause a distribution shift, degrading static model performance. To enable real-time correction, we propose CAP-TTA, a test-time adaptation framework. CAP-TTA triggers context-aware LoRA updates only when a bias-risk score exceeds a set threshold. By utilizing an offline precomputed diagonal preconditioner, it ensures fast and stable optimization. Across multiple benchmarks and human evaluations, CAP-TTA effectively reduces toxicity/bias score with significantly lower latency than standard optimization methods (e.g., AdamW or SGD). Furthermore, it prevents catastrophic forgetting, and substantially improves narrative fluency over state-of-the-art baselines without compromising debiasing performance.
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。