AI Paper Atlas

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

What Images Cannot Say: Language-Guided Olfactory Representation Learning

Eleftherios Tsonis, Xi WANG, Vicky Kalogeiton

ECCV 2026 · Vision-language and multimodal › Cross-modal learning · approach: Contrastive vision-language alignment · 2 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 context and inferred-smell descriptors to object labels in text supervision (ablation)SCENT · New York Smells S2TR@5: 8.0 -> 11.9Improved
add context and inferred-smell descriptors to object labels (ablation)SCENT · New York Smells S2IR@5: 22.1 -> 23.0Improved
align smell jointly with image and text instead of image only (frozen CLIP) (ablation)SCENT · New York SmellsR@5: 12.1 -> 14.3Improved
use object-only text instead of image alignment (ablation)SCENT · New York SmellsR@5: 12.1 -> 8.8Hurt
use native three-stream design instead of Image Bridge (ablation)SCENT · New York Smells S2TR@5: 9.1 -> 11.9Improved
use native three-stream design instead of Image Bridge (ablation)SCENT · New York Smells S2ITR@5: 21.4 -> 23.3Improved
use VLM annotations (Qwen3-VL-30B) instead of text-only LLM annotations (ablation)SCENT · New York Smells S2TR@20: 23.7 -> 29.2Improved
use stronger annotating VLM (Qwen3-VL-30B over Qwen2.5-VL-72B) (ablation)SCENT · New York Smells S2TR@5: 8.1 -> 11.9Improved
use smaller annotating VLM (Qwen3-VL-8B) (ablation)SCENT · New York Smells S2TR@5: 11.9 -> 6.5Hurt
use decoded synthesis from latent decomposition instead of raw recombination (ablation)SCENT · zero-shot recombination retrievalR@20: 9.8 -> 13.7Improved
replace NYS with SCENT on open scenesSCENT · New York Smells open scenes S2TR@5: 9.2 -> 13.0Improved
native text retrieval versus image bridgeSCENT · New York Smells T2SR@5: 9.9 -> 7.2Hurt

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