AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

VisReflect: Latent Visual Reflection for Fine-Grained Perception in Long Visual Context

Xiaoqian Shen, Mohamed Elhoseiny

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Attention-based visual token pruning · 0 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
add visual reflection with cosine loss (ablation)Qwen2.5-VL Finetune · V*accuracy: 80.1 -> 84.9Improved
add visual reflection with cosine loss (ablation)Qwen2.5-VL Finetune · HRBench-4Kaccuracy: 71.3 -> 73.8Improved
add visual reflection with cosine loss (ablation)Qwen2.5-VL Finetune · BLINKaccuracy: 55.2 -> 57.1Improved
fine-tune backbone without visual reflection (ablation)Qwen2.5-VL · BLINKaccuracy: 56.5 -> 55.2Hurt
use cosine similarity instead of MSE as alignment loss (ablation)VisReflect · V*accuracy: 83.2 -> 84.9Improved
use cosine similarity instead of MSE as alignment loss (ablation)VisReflect · HRBench-8Kaccuracy: 69.3 -> 70.1Improved
add visual supervision on reflection tokens instead of Coconut (ablation)VisReflect · V*accuracy: 79.4 -> 84.9Improved
increase training <VR> tokens from 16 to 128 (ablation)VisReflect · V*accuracy: 81.6 -> 84.9Improved
increase training <VR> tokens from 128 to 196 (ablation)VisReflect · V*accuracy: 84.9 -> 83.2Hurt
use one <VR> token at inference (ablation)VisReflect · V*accuracy: 78.53 -> 84.85Improved
increase inference <VR> tokens on images from 1 to 32 (ablation)VisReflect · V*accuracy: 84.85 -> 84.72No clear effect
increase inference <VR> tokens on video from 0 to 16 (ablation)VisReflect · MVBenchaccuracy: 68.44 -> 70.54Improved

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