AI Paper Atlas

Research map / Video / Video-language understanding

CAST: Modeling Visual State Transitions for Consistent Video Retrieval

Yanqing Liu, Yingcheng Liu, Fanghong Dong, Budianto Budianto, Cihang Xie, Yan jiao

ICML 2026 · Video › Video-language understanding · approach: Text-video retrieval · 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 direct target prediction with additive residual update (ablation)Early Fusion · CVR YouCook2 valAccuracy: 35.99 -> 38.95Improved
replace direct target prediction with additive residual update (ablation)Early Fusion · CVR YouCook2 valState Accuracy: 38.92 -> 43.51Improved
use dual-path design with cross-attention instead of early concatenation (ablation)CAST with residual · CVR YouCook2 valAccuracy: +4.05% Acc.Improved
add CAST prediction score C to text matching (semantic ensemble) (ablation)Inference scoring · CVR YouCook2 valAccuracy: 25.03 -> 45.46Improved
add visual continuity score B to semantic ensemble (ablation)Full Ensemble · CVR YouCook2 valIdent. Acc.: 70.38 -> 78.48Improved
add visual continuity score B to semantic ensemble (ablation)Full Ensemble · CVR YouCook2 valAccuracy: 45.46 -> 44.77Hurt
use CAST prediction alone vs text matching alone (ablation)CAST score C · CVR YouCook2 valAccuracy: 25.03 -> 42.60Improved
increase context length from L=0 to L=1 (ablation)CAST · CVR YouCook2, COIN, CrossTaskAccuracy and Ident. Acc.Improved
increase context length beyond L=1 (ablation)CAST · CVR YouCook2, COIN, CrossTaskAccuracyNo clear effect
add CAST adapter to InternVideo2-1BInternVideo2-1B · CVR YouCook2Accuracy: 36.75 -> 71.68Improved
add CAST adapter to InternVideo2-1BInternVideo2-1B · CVR CrossTaskAccuracy: 20.61 -> 64.36Improved
add CAST adapter to Qwen3-VL-Embedding-2BQwen3-VL-Embedding-2B · CVR YouCook2Accuracy: 33.45 -> 56.64Improved

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