← 返回名录条目信息来自其公开主页与公开发布内容。按本站规范，页面不展示任何联系方式。 需要更正或删除？通过收录与更正通道提交，24 小时内处理。
H
Haowen Xu
🎓 PhD student in CS @ ZJU | Vision-Language Models & Multimodal Learning | Building efficient AI that sees and reasons | Open-source enthusiast
- 公司
- @ZJU-CCNT
- 位置
- Hangzhou, China
- Stars
- 2
- 粉丝 / 仓库
- 4
代表作品 / 项目
- vlm-eval⭐ 1Unified evaluation harness for Vision-Language Models — VQAv2, TextVQA, COCO, MMBench, MMBench-CN
- visual-rag⭐ 1Lightweight multimodal RAG pipeline for image-and-text understanding — CLIP encoding + FAISS retrieval + VLM generation
- amai_ocl⭐ 0
- verl-omni⭐ 0Multimodal RL training framework for diffusion & omni models
- AdaRubrics⭐ 0AdaRubric: Adaptive Dynamic Rubric Evaluator for Agent Trajectories