AI Paper Atlas

Research map / Video / Video-language understanding

LIVEditor-14B: Lightning Unified Video Editing via In-Context Sparse Attention

Shitong Shao, zikai zhou, Haopeng Li, Yingwei Song, Wenliang Zhong, Lichen Bai, Zeke Xie

ICML 2026 · Video › Video-language understanding · approach: Long video understanding · 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
apply ISA training-free instead of full attentionLIVEditor-14B · EditVerseBenchspeedup vs FA3: 1.00 -> 1.47Improved
train with ISA instead of full attention (ablation)LIVEditor-14B · EditVerseBenchVLM evaluation score (first column): 7.62 -> 7.89Improved
use ISA instead of other sparse attention (Radial, Sparge, STA, SWA, VSA) training-free (ablation)LIVEditor-14B (full-attn) · EditVerseBenchVLM evaluation score (first column): 7.78 vs 7.44 (Sparge Attention)Improved
add Stage II fine-tuning on 0.089M curated high-quality samples (ablation)LIVEditor (ISA) · EditVerseBenchVLM evaluation score (first column): 6.46 -> 7.89Improved
add Stage II fine-tuning on high-quality data (ablation)LIVEditor (ISA) · VIE-BenchAvg. score: 5.16 -> 6.81Improved
reduce Flat Ratio (more sparsity) (ablation)ISA (training-free) · EditVerseBenchall metricsHurt
reduce No Sparsity Ratio and Select Ratio (ablation)ISA (training-free) · EditVerseBenchall metricsNo clear effect
fine-tune with ISA (post-training) rather than training-free (ablation)ISA · attention approximation error across blocksmean absolute error vs full attentionImproved
use 1st- and 2nd-order Taylor expansion instead of 0-th order (ablation)ISA sparse attention · kernel implementationhardware efficiencyHurt
use query-key deviation norm instead of sharpness as error proxy for grouping (ablation)ISA grouping · Taylor error correlationproxy qualityHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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