AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

SpecV: Specification Verification for Robust Unified Multimodal Evaluation

Weihao Yu, Rongyao Fang, Yuxuan Cai, Linjiang Huang, Yuhuan Yang, Xianwei Zhuang, Junyang Lin, Yixuan Yuan, Shuai Bai

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs · 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 holistic judge scoring with specification verification (SVP)VLM judges · Cross-family judge ranking stabilityKendall tau (avg): 0.856 -> 0.902Improved
replace holistic scoring with SVPVLM judges · Intra-family judge ranking stabilityKendall tau (avg): 0.805 -> 0.867Improved
replace holistic scoring with SVPVLM judges · Cross-family stability, EditingKendall tau: 0.758 -> 0.903Improved
replace holistic scoring with SVPVLM judges · Cross-family stability, Many-to-One/InterleaveKendall tau: 0.807 -> 0.830; intra Many-to-One 0.749 -> 0.758Mixed
replace holistic scoring with SVPVLM judges · Agreement with human ratings (avg)Kendall tau: 0.459 -> 0.543Improved
use multi-model SER instead of Only Claude (ablation)SER · Human correlation (avg)Kendall tau: 0.471 -> 0.493Improved
add cross-model quality filtering to merged specs (ablation)SER · Human correlation (avg)Kendall tau: 0.444 -> 0.493Improved
merge specs from three models without filtering vs best single model (ablation)SER · Human correlation (avg)Kendall tau: 0.471 -> 0.444Hurt
use SER instead of single-model specs on Understanding/T2I (ablation)SER · Human correlation, T2IKendall tau: 0.561 (Only Claude) -> 0.516Hurt
use SER on Many-to-One (ablation)SER · Human correlation, Many-to-OneKendall tau: 0.312 (Only Claude) -> 0.442Improved

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