AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Unified Vision–Language Modeling via Concept Space Alignment

Yifu QIU, Paul-Ambroise Duquenne, Holger Schwenk

ICLR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Recaptioning and text-supervised pretraining · 1 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
align Perception Encoder to SONAR (vSONAR) vs original PEPerception Encoder · PE-Video retrievalRecall@1: 63.91 -> 73.03Improved
align Perception Encoder to SONAR (vSONAR) vs original PEPerception Encoder · DREAM-1k retrievalRecall@1: 72.10 -> 63.30Hurt
use frozen linear projector instead of full PE fine-tuning (ablation)vSONAR · PE-Video captioningBLEU: 37.1 -> 38.0Improved
add asynchronous learning rates for projector and encoder (ablation)vSONAR · PE-Video captioningBLEU: 37.1 -> 39.7Improved
add temporal attention layer (ablation)vSONAR · PE-Video captioningBLEU: 39.8 -> 39.8No clear effect
add stage-1 image captioning (12M) data (ablation)vSONAR · PE-Video captioningBLEU: 39.8 -> 40.1Improved
add synthetic video captioning stage (ablation)vSONAR · PE-Video captioningBLEU: 39.6 -> 40.1Improved
add contrastive loss to MSE alignment (ablation)vSONAR · PE-Videoretrieval vs captioning: R@1 49.0 -> 52.4; BLEU 38.9 -> 38.6Mixed
fine-tune LCM on vision-language instructions (v-LCM)LCM · M3IT IVQAROUGE-L: 48.9 -> 63.9Improved
fine-tune LCM on vision-language instructions (v-LCM)LCM · M3IT VCRtoken F1: 0.385 -> 0.671Improved
use OmniSONAR instead of SONAR 1 as alignment targetvSONAR · video captioningBLEUImproved
feed vSONAR embeddings instead of SONAR text re-encoded captions to LCMLCM · VideoXumROUGE-LImproved

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