AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

WalkVLM: Aid Visually Impaired People Walking by Vision Language Model

Zhiqiang Yuan, Ting Zhang, Yeshuang Zhu, Jiapei Zhang, Ying Deng, Zexi Jia, Peixiang Luo, Xiaoyue Duan, Jie Zhou, Jinchao Zhang

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: VLM capability analysis · 20 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 CoT-based hierarchical planning (ablation)WalkVLM · WAD reminder taskTF-IDF: 0.094 -> 0.166Improved
add CoT-based hierarchical planning (ablation)WalkVLM · WAD reminder taskROUGE-L: 0.066 -> 0.173Improved
add CoT-based hierarchical planning (ablation)WalkVLM · WAD reminder taskROUGE-1: 0.073 -> 0.191Improved
add approximate position prior (POLM) (ablation)WalkVLM · WAD reminder taskTF-IDF: 0.151 -> 0.166Improved
add approximate position prior (POLM) (ablation)WalkVLM · WAD reminder taskROUGE-2: 0.062 -> 0.062No clear effect
add exact POLM prior (location and category) (ablation)WalkVLM · WAD reminder taskROUGE-L: 0.164 -> 0.173Improved
add exact POLM prior (location and category) (ablation)WalkVLM · WAD reminder taskTF-IDF: 0.152 -> 0.166Improved
fine-tune MiniCPM-V2.6 on WADMiniCPM-V2.6 (8B) · WAD reminder taskGPT Score: 0.025 -> 1.024Improved
fine-tune Qwen2-VL on WADQwen2-VL (7B) · WAD QA taskGPT Score: 0.504 -> 3.246Improved
fine-tune DeepSeek-VL on WADDeepSeek-VL (7B) · WAD reminder taskROUGE-L: 0.068 -> 0.141Improved
add temporal-aware adaptive prediction and WalkVLM design (vs GPT-4o)WalkVLM · WAD reminder task (temporal redundancy)TRF: 0.430 -> 0.505Improved
WalkVLM vs fine-tuned Qwen2-VL on QA TF-IDFWalkVLM · WAD QA taskTF-IDF: 0.246 -> 0.189Hurt

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