利用多模态大语言模型中的思维链推理进行人脸活体检测
Harnessing Chain-of-Thought Reasoning in Multimodal Large Language Models for Face Anti-Spoofing
将思维链视觉问答引入人脸活体检测，用 MLLM 的视觉-语言联合推理替代单一视觉模态判别，并给出可复用的 CoT VQA 数据集构建与渐进式训练策略，可直接迁移到其他需要细粒度真伪判别的视觉任务。
原文依据 we introduce FaceCoT (Face Chain-of-Thought), the first large-scale Visual Question Answering (VQA) dataset tailored for FAS
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论人脸活体检测（FAS）在防御打印攻击、屏幕重放和3D面具等呈现攻击时通常仅依赖单一视觉模态，导致其在跨设备、跨环境和跨攻击类型上的泛化能力有限。与此同时，多模态大语言模型（MLLM）近期在图文理解与语义推理方面取得突破，表明将视觉与语言协同推理引入FAS可显著提升鲁棒性与可解释性。然而，缺乏高质量的视觉-语言多模态数据集一直是关键瓶颈。为此，我们提出FaceCoT（Face Chain-of-Thought），这是首个专为FAS定制的大规模视觉问答（VQA）数据集。FaceCoT涵盖14种欺骗攻击类型，并通过高质量的思维链VQA标注丰富模型学习。同时，我们开发了一个经强化学习精炼的描述生成模型，用于扩展数据集并提升标注质量。此外，我们引入思维链增强渐进学习（CEPL）策略，以更好地利用思维链数据并提升模型在FAS任务上的性能。大量实验表明，使用FaceCoT与CEPL训练的模型在多个基准数据集上优于最优方法。
英文原文摘要 1304 字符
Face Anti-Spoofing (FAS) typically depends on a single visual modality when defending against presentation attacks such as print attacks, screen replays, and 3D masks, resulting in limited generalization across devices, environments, and attack types. Meanwhile, Multimodal Large Language Models (MLLMs) have recently achieved breakthroughs in image-text understanding and semantic reasoning, suggesting that integrating visual and linguistic co-inference into FAS can substantially improve both robustness and interpretability. However, the lack of a high-quality vision-language multimodal dataset has been a critical bottleneck. To address this, we introduce FaceCoT (Face Chain-of-Thought), the first large-scale Visual Question Answering (VQA) dataset tailored for FAS. FaceCoT covers 14 spoofing attack types and enriches model learning with high-quality CoT VQA annotations. Meanwhile, we develop a caption model refined via reinforcement learning to expand the dataset and enhance annotation quality. Furthermore, we introduce a CoT-Enhanced Progressive Learning (CEPL) strategy to better leverage the CoT data and boost model performance on FAS tasks. Extensive experiments demonstrate that models trained with FaceCoT and CEPL outperform state-of-the-art methods on multiple benchmark datasets.