AI Paper Atlas

Research map / Video / Video-language understanding

Dynamic Reflections: Probing Video Representations with Text Alignment

Maks Ovsjanikov, Viorica Patraucean, Leonidas Guibas, Tyler Zhu, Tengda Han

ICLR 2026 · Video › Video-language understanding · approach: Video-language 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
use multiple captions (1 to 10) at test time (ablation)vision-text alignment · VaTeXmutual kNN alignment: +60% from 1 to 10 captionsImproved
use more video frames at test time (ablation)VideoMAEv2 · VaTeXmutual kNN alignmentImproved
use more frames for image models (DINOv2) (ablation)DINOv2 · PVDmutual kNN alignmentNo clear effect
average image features over 8 frames instead of single frame (ablation)image models · VaTeXmutual kNN alignment: 0.206 -> 0.223 (best)Improved
use Gemma-2 text encoder instead of older text encoders (ablation)text encoder · VaTeXmutual kNN alignment: 0.18 -> 0.206Improved
add synthesized captions (Gemini 2.5 Pro) at test time (ablation)vision models · PVDmutual kNN alignmentImproved
increase synthesized captions beyond about 6 (ablation)vision models · PVDmutual kNN alignmentNo clear effect
self-supervised video encoder (VideoMAEv2) instead of image encoder (DINOv2)vision encoder · VaTeXtext alignmentImproved
higher text alignment vs downstream performance, SSL video modelsSSL video models · SSv2, Kinetics, depth, pose, trackingcorrelation with downstreamMixed
use negative reordered captions instead of positivetext encoders · VideoComp temporal reorderalignmentHurt

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