AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

MUSE: Resolving Manifold Misalignment in Visual Tokenization via Topological Orthogonality

Panqi Yang, Haodong Jing, Jiahao Chao, Tingyan Xiang, Li Lin, Yao Hu, Yang Luo, Yongqiang Ma

ICML 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Vision-language pretraining · 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
add semantic anchoring (L_ITC) alone to reconstruction baseline (ablation)reconstruction-only tokenizer · tokenizer ablationrFID: 0.79 -> 1.88Hurt
add semantic anchoring (L_ITC) alone (ablation)reconstruction-only tokenizer · tokenizer ablationmIoU: 18.5 -> 14.1Hurt
add semantic anchoring (L_ITC) alone (ablation)reconstruction-only tokenizer · tokenizer ablationzero-shot accuracy: 12.4 -> 70.5Improved
add topology alignment (L_Topo) alone (ablation)reconstruction-only tokenizer · tokenizer ablationmIoU: 18.5 -> 32.5Improved
add topology alignment (L_Topo) alone (ablation)reconstruction-only tokenizer · tokenizer ablationMMBench: 35.6 -> 10.8Hurt
add L_Topo on top of L_ITC (full MUSE) (ablation)MUSE with L_ITC only · tokenizer ablationmIoU: 14.1 -> 46.5Improved
add L_Topo on top of L_ITC (full MUSE) (ablation)MUSE with L_ITC only · tokenizer ablationrFID: 1.88 -> 0.62Improved
add L_Topo on top of L_ITC (full MUSE) (ablation)MUSE with L_ITC only · UMM downstreamMMBench: 65.2 -> 73.4Improved
use Synergistic Block decoupling instead of naive shared architecture (ablation)MUSE · gradient conflictMMBench: 65.2 -> 73.4Improved
use Synergistic Block instead of soft regularization (ablation)MUSE · gradient conflictmIoU: 22.1 -> 46.5Improved
use Synergistic Block instead of soft regularization (ablation)MUSE · gradient conflictgradient cosine similarity: -0.05 -> +0.04Improved
use full two-stream instead of Synergistic Block (ablation)MUSE · gradient conflictMMBench: 73.4 vs 73.6No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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