AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

UniME-V2: MLLM-as-a-Judge for Universal Multimodal Embedding Learning

Tiancheng Gu, Kaicheng Yang, Kaichen Zhang, Xiang An, Ziyong Feng, Yueyi Zhang, Weidong Cai, Jiankang Deng, Lidong Bing

AAAI 2026 · Vision-language and multimodal › Cross-modal learning · approach: Probabilistic embeddings for cross-modal retrieval · 28 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 MLLM-as-a-Judge hard negative mining (ablation)UniME-V2 (Qwen2-VL-2B) · MMEBPrecision@1: 60.1 -> 61.6Improved
add soft semantic matching score alignment (ablation)UniME-V2 (Qwen2-VL-2B) · MMEBPrecision@1: 61.6 -> 63.6Improved
add soft semantic matching score alignment (ablation)UniME-V2 (Qwen2-VL-2B) · SugarCrepe compositionalRecall@1: 63.7 -> 64.1No clear effect
use Qwen2.5-VL-7B instead of InternVL3-8B as judge (ablation)UniME-V2 (Qwen2-VL-2B) · MMEBPrecision@1: 58.5 -> 63.6Improved
increase hard negatives from 4 to 8 (ablation)UniME-V2 (Qwen2-VL-2B) · MMEBPrecision@1: 61.3 -> 63.6Improved
increase hard negatives from 8 to 10 (ablation)UniME-V2 (Qwen2-VL-2B) · MMEBPrecision@1: 63.6 -> 63.0Hurt
UniME-V2 instead of UniME on text-to-image retrievalUniME-V2 · Flickr30K text-to-imageRecall@1: 94.4 -> 93.7 (LLaVA-OV-7B)No clear effect
UniME-V2 instead of UniME on compositional retrievalUniME-V2 (LLaVA-OV-7B) · SugarCrepe ReplaceRecall@1: 80.5 -> 88.6Improved
add UniME-V2-Reranker instead of LamRAUniME-V2 7B retrieval · SugarCrepe compositionalRecall@1: 87.4 -> 94.8Improved
add UniME-V2-RerankerUniME-V2 (Qwen2-VL-7B) · MMEBPrecision@1: 68.0 -> 69.6Improved
UniME-V2 vs UniME on overall MMEB with LLaVA-OV-7BUniME-V2 · MMEBPrecision@1: 70.7 -> 71.2Improved

Later papers that compared with this method

3 later accepted papers report UniME-V2 as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.

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