VEDJE shrinks per-video cache 4x to 12 KiB with near-lossless video retrieval
Insight-bgu · hf · 2026-10-14
VEDJE (Ben-Gurion University, Texas A&M) targets scalable video-text retrieval: it compresses features within sampled frames into a reusable cache and uses feature-change prediction as an auxiliary training signal, with no extra query-time cost. It reaches 59.8 text-to-video R@1 on MSR-VTT with a VideoCLIP-XL first stage, and shrinking the per-video cache 4x to 12 KiB preserves recall within 0.2 points.
More from Research
- RUC and Kling AI propose EventLens: event-structure RL to fix video LLMs' temporal understanding — Guowen Zhang · 2026-10-14
- JSIR dataset drops: 267 jazz recordings with beat, chord and structure annotations — umpedronosapato · 2026-10-14
- There is no such thing as an AI drug, argues researcher Patrick Kidger — PatrickKidger · 2026-10-14
- More thinking, worse scores: refusals tank closed-model evals at max reasoning effort — _lewtun · 2026-10-14
- Integer multiplication repo becomes an agent harness benchmark — and collaboration loses — iskander · 2026-10-14
- Recurrent self-conditioning: framing recent diffusion methods as depth decoupled from time — alec_helbling · 2026-10-14