| Paper | Code | Accuracy | ModelName | ReleaseDate |
|---|---|---|---|---|
| MMVistaReason: Toward Open-Data and Post-Training Recipes for Multimodal Reasoning | ✓ Link | 68.5 | MMVistaReason-9B | 2026-10-22 |
| Twin-T & TwintVQA: A Reliable Structure-Detail Separating VLM and a Comprehensive Benchmark for Chart and Table Tasks | ✓ Link | 59.06 | Twin-T-7B | 2026-06-03 |
| AnE: Pushing the Reasoning Frontier of Multimodal LLMs via Anchor Evolution | ✓ Link | 51.3 | AnE-3rd | 2026-05-25 |
| LASER: A Corrective Lens for LVLMs via Visual Attention Preservation and Sink Suppression | ✓ Link | 51.2 | LASER | 2026-07-02 |
| LoopVL: Recurrent Visual Intelligence | ✓ Link | 45.19 | LoopVL-1B | 2026-09-29 |