AI Paper Atlas

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

How can embedding models bind concepts?

Arnas Uselis, Darina Koishigarina, Seong Joon Oh

ICML 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded visual concept learning · 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 concept components to scene embedding (removing them eliminates concept decoding) (ablation)CLIP-B/32 text probe · PUG:SPAREconcept decoding accuracy: 1.00 → 0.06Improved
add concept components to scene embedding (removing them eliminates concept decoding) (ablation)CLIP-B/32 image probe · PUG:SPAREconcept decoding accuracy: 0.94 → 0.05Improved
keep object components (removing concepts leaves object decoding largely intact) (ablation)CLIP-B/32 text probe · PUG:SPAREobject decoding accuracy: 1.00 → 0.99No clear effect
keep concept components (removing concepts barely changes image object decoding) (ablation)CLIP-B/32 image probe · PUG:SPAREobject decoding accuracy: 0.96 → 0.85Mixed
add object components to scene embedding (removing them collapses object decoding) (ablation)CLIP-B/32 text probe · PUG:SPAREobject decoding accuracy: 1.00 → 0.04Improved
add object components to scene embedding (removing them collapses object decoding) (ablation)CLIP-B/32 image probe · PUG:SPAREobject decoding accuracy: 0.96 → 0.01Improved
use position-aware (AVG+POS) object embeddings instead of AVG (ablation)CLIP image encoder · PUG:SPARER^2 of scene reconstruction: 0.75 / 0.84Improved
use position-aware (AVG+POS) object embeddings instead of AVG (ablation)DINO · PUG:SPARE object replacement editretrieval accuracy: 0.48 / 0.76Improved
use single-object scene embeddings for object edits instead of averaged (ablation)CLIP · CLEVR object replacement editretrieval accuracy: 1.00 → 0.97Hurt
add object-level components on top of bag-of-concepts (scenes from objects vs concepts) (ablation)CLIP-B/32 image · PUG:SPARE (AVG)R^2: 0.30 → 0.75Improved
add multiplicative interactions to additive bag-of-concepts probe (ablation)probe on transformer embeddings · held-out objectsobject recognition accuracyImproved
increase training coverage of objects from 30% to 40%transformer dual-encoder, |O|=125,000 · held-out objectsobject recognition accuracy: 30% to 40%Improved

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