AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Griffon v2: Advancing Multimodal Perception with High-Resolution Scaling and Visual-Language Co-Referring

Yufei Zhan, Shurong Zheng, Yousong Zhu, Hongyin Zhao, Fan Yang, Ming Tang, Jinqiao Wang

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Instruction-tuned multimodal assistants · 28 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace Flamingo-style resampler with strided-conv down-sampling projector (ablation)Griffon v2 · MSCOCO detectionmAP: 9.6 -> 28.4Improved
use direct high-resolution encoding instead of image partitioning (Monkey) (ablation)Griffon v2 · MSCOCO detection (same train data)mAP: 17.6 -> 24.3Improved
use direct high-resolution instead of LLaVA-NeXt partition design (ablation)Griffon v2 · ChartQAaccuracy: 62.2 -> 69.8Improved
use EVA2-CLIP visual encoder instead of original CLIP-L/14 (ablation)Griffon v2 · MSCOCO detectionmAP: 29.8 -> 31.9Improved
use EVA2-CLIP with interpolated positions instead of SAM-CLIP-16 (ablation)Griffon v2 · MSCOCO detectionmAP: 28.4 -> 31.9Improved
raise input resolution from 448 to 700 using down-sampling projector (ablation)Griffon v2 · MSCOCO detectionmAPImproved
increase resolution further with down-sampling projector (ablation)Griffon v2 · MSCOCO detectionmAPImproved
scale to high resolution with down-sampling projector vs Griffon-13B at 448Griffon v2 · MSCOCO val2017 detectionAP_S: 5.9 -> 19.4Improved
use Griffon v2 instead of Qwen2.5-VL-7BODinW · ODinW (13 sets)avg mAP: 37.0 -> 42.0Improved
use Griffon v2 instead of ODinW specialist G-DINO-LODinW · ODinW (13 sets)avg mAP: 52.4 -> 42.0Hurt
use Griffon v2 instead of Lumen (task-specific decoder)MSCOCO detection · MSCOCO val2017 detectionmAP: 35.3 -> 38.5Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy