AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Social Debiasing for Fair Multi-modal LLMs

Harry Cheng, Yangyang Guo, Qingpei Guo, Ming Yang, Tian Gan, Weili Guan, Liqiang Nie

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Benchmarks and analysis of MLLMs · 4 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 dataset resampling on top of fine-tuning (ablation)LLaVA-7B+FT · CMSC testMaxSkew@C: 0.8058 -> 0.7741Improved
add SFLoss on top of fine-tuning (ablation)LLaVA-7B+FT · CMSC testMaxSkew@C: 0.8058 -> 0.5838Improved
add SFLoss on top of resampling (full CSD) (ablation)LLaVA-7B+Resample · CMSC testMaxSkew@C: 0.7741 -> 0.5633Improved
add resampling on top of fine-tuning (ablation)Bunny-8B+FT · CMSC testMaxSkew@C: 0.9026 -> 0.4008Improved
add SFLoss on top of fine-tuning (ablation)Qwen-VL-7B+FT · CMSC testMaxSkew@C: 1.0227 -> 0.8041Improved
add SFLoss to resampling (full CSD) (ablation)LLaVA-13B+Resample · CMSC testMaxSkew@C: 0.4292 -> 0.4192No clear effect
replace naive fine-tuning with CSDLLaVA-13B+FT · CMSC testMinSkew@C: -1.5107 -> -0.8167Improved
replace naive fine-tuning with CSDQwen-VL-7B+FT · FairFaceMaxSkew@C: 0.6916 -> 0.2607Improved
fine-tune the full model instead of only the language component (ablation)LLaVA-7B · CMSC testMaxSkew@C: 0.9839 -> 0.8058Improved
fine-tune the visual component instead of language (ablation)LLaVA-7B · CMSC testMaxSkew@C: 0.9839 -> 1.0036No clear effect
use full-parameter CSD instead of LoRA (ablation)LLaVA-7B · CMSC testMaxSkew@C: 0.9945 -> 0.5633Improved
raise Bunny learning rate from 1e-7 to 5e-6 (ablation)Bunny-8B · TextVQAaccuracy: 65.20 -> 41.06Hurt

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