SSL-SSAW: Self-Supervised Learning with Sigmoid Self-Attention Weighting for Question-Based Sign Language Translation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909793058291712 |
|---|---|
| author | Liu, Zekang Feng, Wei Shang, Fanhua Hu, Lianyu Feng, Jichao Gao, Liqing |
| author_facet | Liu, Zekang Feng, Wei Shang, Fanhua Hu, Lianyu Feng, Jichao Gao, Liqing |
| contents | Sign Language Translation (SLT) bridges the communication gap between deaf people and hearing people, where dialogue provides crucial contextual cues to aid in translation. Building on this foundational concept, this paper proposes Question-based Sign Language Translation (QB-SLT), a novel task that explores the efficient integration of dialogue. Unlike gloss (sign language transcription) annotations, dialogue naturally occurs in communication and is easier to annotate. The key challenge lies in aligning multimodality features while leveraging the context of the question to improve translation. To address this issue, we propose a cross-modality Self-supervised Learning with Sigmoid Self-attention Weighting (SSL-SSAW) fusion method for sign language translation. Specifically, we employ contrastive learning to align multimodality features in QB-SLT, then introduce a Sigmoid Self-attention Weighting (SSAW) module for adaptive feature extraction from question and sign language sequences. Additionally, we leverage available question text through self-supervised learning to enhance representation and translation capabilities. We evaluated our approach on newly constructed CSL-Daily-QA and PHOENIX-2014T-QA datasets, where SSL-SSAW achieved SOTA performance. Notably, easily accessible question assistance can achieve or even surpass the performance of gloss assistance. Furthermore, visualization results demonstrate the effectiveness of incorporating dialogue in improving translation quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_14036 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SSL-SSAW: Self-Supervised Learning with Sigmoid Self-Attention Weighting for Question-Based Sign Language Translation Liu, Zekang Feng, Wei Shang, Fanhua Hu, Lianyu Feng, Jichao Gao, Liqing Computation and Language Artificial Intelligence Sign Language Translation (SLT) bridges the communication gap between deaf people and hearing people, where dialogue provides crucial contextual cues to aid in translation. Building on this foundational concept, this paper proposes Question-based Sign Language Translation (QB-SLT), a novel task that explores the efficient integration of dialogue. Unlike gloss (sign language transcription) annotations, dialogue naturally occurs in communication and is easier to annotate. The key challenge lies in aligning multimodality features while leveraging the context of the question to improve translation. To address this issue, we propose a cross-modality Self-supervised Learning with Sigmoid Self-attention Weighting (SSL-SSAW) fusion method for sign language translation. Specifically, we employ contrastive learning to align multimodality features in QB-SLT, then introduce a Sigmoid Self-attention Weighting (SSAW) module for adaptive feature extraction from question and sign language sequences. Additionally, we leverage available question text through self-supervised learning to enhance representation and translation capabilities. We evaluated our approach on newly constructed CSL-Daily-QA and PHOENIX-2014T-QA datasets, where SSL-SSAW achieved SOTA performance. Notably, easily accessible question assistance can achieve or even surpass the performance of gloss assistance. Furthermore, visualization results demonstrate the effectiveness of incorporating dialogue in improving translation quality. |
| title | SSL-SSAW: Self-Supervised Learning with Sigmoid Self-Attention Weighting for Question-Based Sign Language Translation |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2509.14036 |