Signformer 即所需：迈向手语的边缘人工智能
Signformer is all you need: Towards Edge AI for Sign Language
AI总结：
本文提出从零构建的 Signformer 模型，通过手语本质分析和卷积-注意力创新，在仅 57 万参数下实现边缘可部署的高效手语翻译，并取得排行榜第二。
摘要：
手语翻译，尤其是在无 gloss 范式中，正面临着由于日益资源密集的方法而导致的不可行性与不可持续性的困境。当代最先进技术（SOTA）在很大程度上依赖于预训练的复杂骨干网络，例如大语言模型（LLM）、嵌入源或大规模数据集，从而在现实世界场景的可持续使用中造成了相当大的参数和计算效率低下问题。尽管这些方法取得了成功，但沿着这一研究方向推进，会削弱该领域的总体使命，即为沟通听障人群与普通人群创造实质性价值。遵循大语言模型（LLM）和自然语言处理（NLP）研究的主流趋势，我们追求在架构上进行深刻的本质性变革，以在不借助预训练模型、不进行先验知识迁移、不使用任何被视为非从零开始的 NLP 策略的情况下，实现自底向上的改进。我们提出 Signformer，一个从零开始构建的 Feather-Giant，将该领域推向边缘人工智能（Edge AI），以具备 LLM 能力且可边缘部署的紧凑性重新定义性能与效率的极限。在本文中，我们对手语进行了本质分析，以指导我们的算法设计，并提供了一种具有卷积和注意力创新性的可扩展 transformer 流水线。我们在排行榜上取得了新的第二名，与截至 2024 年的最优方法相比，参数量减少了 467-1807 倍，并在 57 万参数的更轻量配置下超越了几乎所有其他方法。
英文摘要：
Sign language translation, especially in gloss-free paradigm, is confronting a dilemma of impracticality and unsustainability due to growing resource-intensive methodologies. Contemporary state-of-the-arts (SOTAs) have significantly hinged on pretrained sophiscated backbones such as Large Language Models (LLMs), embedding sources, or extensive datasets, inducing considerable parametric and computational inefficiency for sustainable use in real-world scenario. Despite their success, following this research direction undermines the overarching mission of this domain to create substantial value to bridge hard-hearing and common populations. Committing to the prevailing trend of LLM and Natural Language Processing (NLP) studies, we pursue a profound essential change in architecture to achieve ground-up improvements without external aid from pretrained models, prior knowledge transfer, or any NLP strategies considered not-from-scratch. Introducing Signformer, a from-scratch Feather-Giant transforming the area towards Edge AI that redefines extremities of performance and efficiency with LLM-competence and edgy-deployable compactness. In this paper, we present nature analysis of sign languages to inform our algorithmic design and deliver a scalable transformer pipeline with convolution and attention novelty. We achieve new 2nd place on leaderboard with a parametric reduction of 467-1807x against the finests as of 2024 and outcompete almost every other methods in a lighter configuration of 0.57 million parameters.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
