AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Beyond the Eye: Efficient Multimodal Reasoning Via Self-Regulated Implicit Visual Tools

Xiuwei Chen, Quanlin Chen, Wentao Hu, Zisheng Chen, Kun Xiang, Zehua Ma, Mingyang Zhang, Jianhua Han, Hanhui Li, Hang Xu, Xiaodan Liang

TPAMI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning · 1 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 implicit visual tools and BEE training to Qwen2.5-VL-7BQwen2.5-VL-7B · V*accuracy: 76.4 -> 83.8Improved
add implicit visual tools to Qwen3-VL-4BQwen3-VL-4B · V*accuracy: 80.6 -> 91.6Improved
add implicit visual tools to Qwen3-VL-8BQwen3-VL-8B · MME-Real-Liteaccuracy: 49.3 -> 59.4Improved
add implicit visual tools to Qwen3-VL-8BQwen3-VL-8B · POPEaccuracy: 87.6 -> 87.7No clear effect
add implicit visual tools to Qwen3-VL-4BQwen3-VL-4B · LogicVistaaccuracy: 53.2 -> 53.4No clear effect
add MC-Reward (self-regulated tool use) instead of static reward (ablation)BEE-7B · NTGNTG: 6.2 -> 66.0Improved
add MC-Reward instead of static reward (ablation)BEE-7B · system latencyavg. latency (s): 2.53 -> 1.37Improved
add MC-Reward instead of static reward (ablation)BEE-7B · system throughputthroughput (sample/s): 0.88 -> 1.61Improved
use shared single token instead of generated implicit tokens (Replace) (ablation)BEE-7B · V* overallaccuracy: 81.1 -> 83.8Improved
use target image instead of original image, vs unk tokens, vs text-only (ablation)BEE-7B · HRBenchaccuracy: largest drop on HRBench 8K when target image replaced by originalImproved
add adapter decoupling text and visual token optimization (ablation)BEE-7B · V* overallaccuracy: 83.8 -> 84.2Improved
add adapter decoupling text and visual token optimization (ablation)BEE-7B · HRBench-4K overallaccuracy: 75.4 -> 76.2Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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