AI Paper Atlas

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

Enhancing Alignment for Unified Multimodal Models via Semantically-Grounded Supervision

Jiyeong Kim, Yerim So, Hyesong Choi, Uiwon Hwang, Dongbo Min

ECCV 2026 · Vision-language and multimodal › Cross-modal learning · approach: Contrastive vision-language alignment · 0 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
replace SFT with SeGroS fine-tuningHarmon-1.5B (BLIP3o-60k) · GenEval overalloverall score: 85.0 -> 88.66Improved
replace Reca with SeGroS fine-tuningOpenUni-3.6B · GenEval overalloverall score: 74.1 -> 75.37Improved
replace SFT with SeGroS fine-tuningOpenUni-3.6B · GenEval Positionposition accuracy: 25.0 -> 46.8Improved
replace Reca with SeGroS fine-tuningShow-o-512 · GenEval overalloverall score: 71.63 -> 70.69Hurt
add visual hints from top 30% grounded patches (ablation)Harmon-0.5B · GenEval overall (Harmon-0.5B, MidjourneyV6 ablation)GenEval overall: 72.1 -> 81.5Improved
use full image (100%) instead of top 30% patches as hint (ablation)SeGroS Harmon-0.5B · GenEval overall (Harmon-0.5B, MidjourneyV6 ablation)GenEval overall: 81.5 -> 79.2Hurt
add intra- and inter-modal affinity text token filtering (ablation)SeGroS Harmon-0.5B · GenEval overall (Harmon-0.5B, MidjourneyV6 ablation)GenEval overall: 79.2 -> 81.5Improved
add intra-modal affinity only text filtering (ablation)SeGroS Harmon-0.5B · GenEval overall (Harmon-0.5B, MidjourneyV6 ablation)GenEval overall: 79.2 -> 79.8Improved
use bottom patches as unmasked context instead of top patches (ablation)SeGroS Harmon-0.5B · GenEval overall (Harmon-0.5B, MidjourneyV6 ablation)GenEval overall: 79.9 -> 81.5Improved
use top rather than bottom patches as visual hints (ablation)SeGroS Harmon-0.5B · GenEval overall (Harmon-0.5B, MidjourneyV6 ablation)GenEval overall: 80.4 -> 81.5Improved
restrict reconstruction loss to top-30% grounded masked targets (ablation)SeGroS Harmon-0.5B · GenEval overall (Harmon-0.5B, MidjourneyV6 ablation)GenEval / DPGBench: 81.5 -> 82.1Mixed
add noise scale 0.5 to the visual grounding map (ablation)SeGroS Harmon-1.5B (BLIP3o-60k) · GenEval overalloverall score: 87.9 -> 88.7Improved

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