BAAI's encoder-free VLM family EVE: NeurIPS spotlight, ICCV highlight, open-sourced
liuziwei7 · x · 2026-11-14
A repost by liuziwei7 highlights BAAI's EVE series — encoder-free vision-language models that remove the ViT/CLIP vision encoder so the LLM processes visual patches directly. EVEv1 was a NeurIPS 2024 spotlight, EVEv2 an ICCV 2025 highlight; code and weights are open-sourced (376 stars, MIT license). The thread also links the NEO and SenseNova-U series from related efforts, celebrating community progress on this direction.
More from Multimodal
- Harvard-MIT UGG model unifies video generation, depth and camera pose in one diffusion model — burny_tech · 2026-11-14
- CapCut Web Ultra Offers a Week of Unlimited Seedream and GPT Image 2.5 Generation — JaynitMakwana · 2026-11-14
- Solv3R demo impresses: 3DGS reconstruction runs in about a minute — jonstephens85 · 2026-11-14
- Dreamina allegedly paid $82k in 'labor' for a 4-minute AI video — AIandDesign · 2026-11-14
- CapCut Adds Seedance 2.5 for 4K Video Generation in AI Lab and Video Studio — CurieuxExplorer · 2026-11-14
- Fable 5.2 builds full dragon scene in Blender from scratch in under 20 minutes — PokePR0 · 2026-11-14