研究人员开发了新的方法来改进无词汇手语翻译，解决了将视觉手语视频与口语文本对齐的挑战。一种方法，手语翻译的选择性对比学习（SCL-SLT），使用对选择策略来识别和强调训练过程中信息量更大的负面示例，减少来自语义相似对的噪声。另一种方法，SignDPO，采用跨空间、时间、语言维度的多层次直接偏好优化来增强基于骨骼的手语翻译，其性能优于现有的无词汇技术。 AI
排序理由 该集群包含两篇详细介绍手语翻译新方法的学术论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2604.22374v1 Announce Type: new Abstract: Sign language translation (SLT) converts continuous sign videos into spoken-language text, yet it remains challenging due to the intrinsic modality mismatch between visual signs and written text, particularly in gloss-free settings.…
Sign language translation (SLT) converts continuous sign videos into spoken-language text, yet it remains challenging due to the intrinsic modality mismatch between visual signs and written text, particularly in gloss-free settings. Recent SLT systems increasingly adopt CLIP-like…
We present SignDPO, a novel multi-level Direct Preference Optimisation (DPO) framework designed to enhance the alignment of skeleton-based Sign Language Translation. While current skeleton-based models have made significant progress using Maximum Likelihood Estimation, they are p…