AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

OrderChain: Towards General Instruct-Tuning for Stimulating the Ordinal Understanding Ability of MLLM

Jinhong Wang, Shuo Tong, Jian Liu, Dongqi Tang, Weiqiang Wang, Wentong Li, Hongxia Xu, Danny Z. Chen, Jintai Chen, Jian Wu

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: VLM capability analysis · 3 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 Domain Knowledge Prompts (ablation)LLaVA baseline · Adienceaccuracy: 47.5 -> 58.0Improved
add Category Feature Prompts alone (ablation)LLaVA baseline · Adienceaccuracy: 47.5 -> 32.5Hurt
add Category Feature Prompts on top of Domain Knowledge Prompts, no RO-CoT (ablation)LLaVA + DKP · Adienceaccuracy: 58.0 -> 38.7Hurt
add RO-CoT to Category Feature Prompts (ablation)LLaVA + CFP · Adienceaccuracy: 32.5 -> 84.6Improved
add Domain Knowledge Prompts to Category Feature Prompts + RO-CoT (full OrderChain) (ablation)LLaVA + CFP + RO-CoT · Adienceaccuracy: 84.6 -> 93.2Improved
add Domain Knowledge Prompts to Category Feature Prompts + RO-CoT (full OrderChain) (ablation)LLaVA + CFP + RO-CoT · AdienceMAE: 0.18 -> 0.12Improved
add OrderChain to LLaVA-1.5LLaVA-1.5 baseline · HCIaccuracy: 61.4 -> 73.0Improved
add OrderChain to LLaVA-1.5LLaVA-1.5 baseline · DRaccuracy: 30.0 -> 85.7Improved
add OrderChain to LLaVA-1.5LLaVA-1.5 baseline · Image Aesthetics (overall)accuracy: 59.56 -> 73.83Improved
compare OrderChain with NumCLIP on subjective aesthetics labelsLLaVA-1.5 + OrderChain · Image Aesthetics (overall)accuracy: 76.53 -> 73.83Hurt
use OrderChain-trained LLaVA for zero-shot transfer to APTOSLLaVA baseline · APTOS (zero-shot transfer)accuracy: 23.2 -> 70.0Improved
use OrderChain-trained LLaVA for zero-shot transfer to KonIQLLaVA baseline · KonIQ (zero-shot transfer)accuracy: 36.3 -> 51.5Improved

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