Tag: Multimodal Data
All the articles with the tag "Multimodal Data".
-
Compression via Pre-trained Transformers: A Study on Byte-Level Multimodal Data
本文通过大规模实验证明,预训练小型Transformer模型在考虑参数大小的情况下,能在文本、图像和音频的分布外数据上实现与传统压缩算法竞争的压缩比,尤其在训练模态内表现优异,但跨模态迁移能力较弱。
-
Learning Composable Chains-of-Thought
本文提出Composable Chain-of-Thought方法,通过数据增强改进原子任务CoT格式,并结合多任务学习或模型合并实现零样本组合推理,使用拒绝采样微调进一步提升性能,在字符串操作和自然语言任务上优于标准CoT基准。
-
General-Reasoner: Advancing LLM Reasoning Across All Domains
本文提出General-Reasoner,通过零强化学习结合跨领域高质量数据集和基于生成模型的验证器,显著提升大型语言模型在多领域推理任务上的性能,同时保持数学推理的有效性。
-
How does Transformer Learn Implicit Reasoning?
本文通过在受控符号环境中从头训练Transformer模型,揭示了隐式多跳推理的三阶段发展轨迹,并利用跨查询语义补丁和余弦表示透镜工具,阐明了推理能力与隐藏空间聚类的关联,为模型可解释性提供了新见解。
-
Thermal Detection of People with Mobility Restrictions for Barrier Reduction at Traffic Lights Controlled Intersections
This paper introduces a thermal detector-based traffic light system using YOLO-Thermal, a modified YOLOv8 framework, to dynamically adjust signal timings for individuals with mobility restrictions, achieving superior detection accuracy (89.1% APval) and enhancing intersection accessibility while addressing privacy and adverse condition challenges.