AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

IVQA-LD: Inclusive Multimodal Understanding for Population with Limb Deficiency

Yan Ke, Xin Shen, Jiaying Ying, Xin Li, Xin Yu

ICML 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Multimodal LLM benchmarks

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Introduces IVQA-LD, a limb-deficiency-aware vision-language dataset and benchmark for inclusive multimodal reasoning for people with limb differences.

Problem

People with limb differences face barriers to inclusive AI services, largely because structured, high-quality multimodal resources centered on disability contexts are lacking.

Approach

A limb-deficiency-aware body-centric learning and evaluation paradigm combining a large-scale limb-aware vision-language dataset with an evaluation benchmark for multimodal reasoning; the second component is cut off in the available abstract.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add stage-1 residual-limb grounding warm-up (S1) (ablation)Qwen2.5-VL-7B SFT · IVQA-LD caption generationCIDEr: 65.37 -> 76.15Improved
add stage-1 residual-limb grounding warm-up (S1) (ablation)Qwen2.5-VL-7B SFT · IVQA-LD caption generationMETEOR: 26.04 -> 27.29Improved
add stage-1 residual-limb grounding warm-up (S1) (ablation)Qwen2.5-VL-7B SFT · IVQA-LD guide generationguide score: 4.54 -> 4.64Improved
add stage-2 grounding supervision (S2) on top of S1 (ablation)Qwen2.5-VL-7B · IVQA-LD caption generationCIDEr: 76.15 -> 79.20Improved
add stage-2 grounding supervision (S2) on top of S1 (ablation)Qwen2.5-VL-7B · IVQA-LD caption generationMETEOR: 27.29 -> 27.33No clear effect
add stage-2 grounding supervision (S2) on top of S1 (ablation)Qwen2.5-VL-7B · IVQA-LD guide generationguide score: 4.64 -> 4.64No clear effect
add BSI over standard SFTQwen2.5-VL-7B · IVQA-LD visual groundingAcc@0.5: 18.91 -> 26.34Improved
add BSI over standard SFTQwen2.5-VL-7B · IVQA-LD deficiency typeaccuracy: 37.60 -> 52.60Improved
add BSI over standard SFTQwen2.5-VL-7B · IVQA-LD scene reasoningaccuracy: 83.00 -> 80.13Hurt
add BSI over standard SFTQwen2.5-VL-7B · IVQA-LD prosthesis interactionaccuracy: 67.33 -> 88.29Improved
add BSI over standard SFTQwen2.5-VL-7B · IVQA-LD WPA classificationaccuracy: 63.80 -> 72.40Improved
fine-tune Qwen2.5-VL-7B on IVQA-LD vs zero-shotQwen2.5-VL-7B · IVQA-LD ICF classificationaccuracy: 0.30 -> 73.11Improved

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