AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

VisWordBench: Bridging the Gap in Cross-modal Reasoning for Multimodal Large Language Models

Tianshu Zhang, Junzhe Chen, Yean Cheng, Demin Zhu, Haoze Zheng, Lijie Wen

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Unified any-modality language models · 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 inference-time steering from hypothesis-diversified dataGLM-4.1v-thinking · VisWordBench-enaccuracy (%): 3.43 -> 5.23Improved
add inference-time steering from hypothesis-diversified dataKimi-vl-thinking · VisWordBench-enaccuracy (%): 1.66 -> 2.77Improved
add inference-time steering from hypothesis-diversified dataGLM-4.5v · VisWordBench-enaccuracy (%): 14.95 -> 16.18Improved
steer with conventional distillation data instead of our diversified data (ablation)GLM-4.1v-thinking · VisWordBench-enaccuracy (%): 3.43 -> 4.43No clear effect
fine-tune on our diversified synthetic data instead of answer-only data (ablation)GLM-4.1v-thinking · VisWordBench-enaccuracy (%): 5.23 -> 5.72Improved
fine-tune on our diversified synthetic data instead of answer-only data (ablation)Kimi-vl-thinking · VisWordBench-enaccuracy (%): 2.52 -> 3.26Improved
LoRA on our diversified synthetic data instead of answer-only data (ablation)GLM-4.5v · VisWordBench-enaccuracy (%): 16.12 -> 18.46Improved
raise steering strength alpha too high (ablation)GLM-4.1v-thinking · VisWordBench-enaccuracy (%)Hurt
apply steering vector from VisWordBench zero-shotGLM-4.1v-thinking · VisuLogicaccuracy (%): 24.7 -> 25.9Improved
apply steering vector from VisWordBench zero-shotGLM-4.1v-thinking · VisuRiddlesaccuracy (%): 24.0 -> 26.7Improved
apply steering vector from VisWordBench zero-shotGLM-4.1v-thinking · CrosswordBenchWCR_all (%): 6.04 -> 7.58Improved
enable thinking modeQwen3.5-9B · VisWordBench-chaccuracy (%): 1.60 -> 2.20Improved

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