ECCV 2026新作：输出感知安全护栏，精准解决多模态大模型“宁可错杀一千”的过度拒绝问题。
arXiv:2607.09697v1 Announce Type: new Abstract: Existing safety mechanisms for multimodal large language models (MLLMs) face a fundamental trade-off b…
物理运动感知与置信度引导LLM结合的生成推理新方法，被ECCV 2026接收
arXiv:2505.16456v3 Announce Type: replace Abstract: Recent advances in 3D content generation have amplified demand for dynamic models that are both vi…
提出Caption Bottleneck模型，通过瓶颈模块压缩视觉特征并生成描述，在ECCV 2026上展现多模态理解新范式。
arXiv:2607.00578v1 Announce Type: new Abstract: Concept Bottleneck Models (CBMs) provide interpretability by routing predictions through a layer of hu…
群核科技联合英伟达、Adobe等提出SPEAR高保真仿真平台，解决物理AI数据与仿真瓶颈，三篇论文入选ECCV 2026。
共同推进物理AI前沿技术探索。
ECCV 2026新作，MotionAtlas攻克运动中心视频的精细区域描述，为视频理解提供更精准的细节维度。
arXiv:2606.29531v1 Announce Type: new Abstract: We propose MotionAtlas, a system for detailed captioning of motion-centric videos, comprising (1) a de…
通过强化多模态参考游戏实现MLLM的个性化定制，论文已被ECCV 2026接收。
arXiv:2606.28845v1 Announce Type: new Abstract: Personalizing Multimodal Large Language Models (MLLMs) aims to recognize users' unique concepts from v…
最新ECCV 2026录用论文，探索如何对预训练视觉Transformer的表示进行可控操作，为视觉AI带来新的灵活性。
arXiv:2604.02327v2 Announce Type: replace Abstract: Pretrained Vision Transformers (ViTs) such as DINOv2 and MAE provide generic image features that c…
清华开源Spatial-TTT，让模型在视频中持续学习更新空间记忆，空间智能超越Gemini，入选ECCV 2026！
120分钟长视频一边看一边记
ECCV 2026接收，ELVA提出排名驱动的通用多模态检索新范式，让跨模态搜索更精准高效。
arXiv:2606.20280v1 Announce Type: cross Abstract: Leveraging Multimodal Large Language Models (MLLMs) via contrastive learning has become a mainstream…
远程感知MLLM在否定句理解上存在显著短板，新研究首次系统评估并提出增强策略，精准提升多模态模型语义层次。
arXiv:2606.20177v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) have demonstrated remarkable success in various Remote Sens…