About
EVisRAG (VisRAG 2.0) is an evidence-guided Vision Retrieval-augmented Generation framework that equips VLMs for multi-image questions by first linguistically observing retrieved images to collect per-image evidence, then reasoning over those cues to answer. EVisRAG trains with Reward-Scoped GRPO, applying fine-grained token-level rewards to jointly optimize visual perception and reasoning.
Open Source Health
- Stars
- 982
- Forks
- 79
- License
- Apache-2.0
- Last commit
- 2 months ago
Resources & Links
Related Categories
Vendor
OpenBMB
OpenBMB (Open Lab for Big Model Base) aims to build foundation models and systems towards AGI
Quick Links
Open Source
More by OpenBMB
Related Products
Private Gpt
Complete API layer for private AI applications on local models: RAG, skills, tools, MCP, text-to-sql, and more. Works with any OpenAI-compatible inference server.
Top category match
PixelRAG
The end of web parsing. The beginning of scalable pixel-native search. link
Top category match
TrustRAG
TrustRAG：The RAG Framework within Reliable input,Trusted output
Top category match
RAG-Anything
"RAG-Anything: All-in-One RAG Framework"
Top category match
Easy Vecdb
📚 从零开始的向量数据库原理与实践教程，在线阅读地址：https://easy-vecdb.datawhale.cc/
Top category match