AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Regression over Classification: Assessing Image Aesthetics via Multimodal Large Language Models

Xingyuan Ma, Shuai He, Anlong Ming, Haobin Zhong, Huadong Ma

AAAI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs · 3 citations (OpenAlex)

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.

Adapts MLLMs for image aesthetics assessment by replacing classification-based score decoding with a regression-style approach, addressing score insensitivity from tokenization.

Problem

Applying multimodal large language models to image aesthetics assessment (IAA) suffers from score insensitivity due to tokenization and from a classification-based decoding mechanism that ignores the numeric ordering of scores.

Approach

Reframes MLLM-based aesthetic scoring as regression rather than classification over score tokens, so predicted scores respect numeric proximity; the abstract only partially states the mechanism.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add decoupled score tokens and regression (ablation)mPLUG-Owl2 baseline · AVASRCC: 0.788 -> 0.825Improved
add decoupled score tokens and regression (ablation)mPLUG-Owl2 baseline · AVAPLCC: 0.786 -> 0.824Improved
add decoupled score tokens and regression (ablation)mPLUG-Owl2 baseline · ICAA17KSRCC: 0.688 -> 0.873Improved
add decoupled score tokens and regression (ablation)mPLUG-Owl2 baseline · ICAA17KPLCC: 0.707 -> 0.880Improved
add loss decoupling via position token (ablation)ROC4MLLM regression · AVASRCC: 0.825 -> 0.833Improved
add loss decoupling via position token (ablation)ROC4MLLM regression · ICAA17KSRCC: 0.873 -> 0.894Improved
add loss decoupling via position token (ablation)ROC4MLLM regression · ICAA17KPLCC: 0.880 -> 0.903Improved
add ROC4MLLM to mPLUG-Owl2mPLUG-Owl2 · AVA (score only)SRCC: 0.761 -> 0.831Improved
add ROC4MLLM to Qwen2-VLQwen2-VL · AVA (score only)SRCC: 0.784 -> 0.831Improved
add ROC4MLLM to LLaVA 1.5LLaVA 1.5 · AVA (score only)SRCC: 0.775 -> 0.793Improved
add comment generation to trainingmPLUG-Owl2 · AVASRCC: 0.761 -> 0.758 (-0.3%)Hurt
add comment generation to trainingQ-Align · AVASRCC: 0.821 -> 0.819 (-0.2%)Hurt

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