AI Paper Atlas

Research map / Video / Video generation

Fastcar: Cache Attentive Replay for Fast Auto-Regressive Video Generation on the Edge

Xuan Shen, Weize Ma, Yufa Zhou, Enhao Tang, Yanyue Xie, Zhengang Li, Yifan Gong, Quanyi Wang, Henghui Ding, Yiwei Wang, Pu Zhao, Jun Lin and 1 more

ICLR 2026 · Video › Video generation · approach: Autoregressive long video generation · 7 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 consistent instead of layer-wise inconsistent thresholds at 10% replay (ablation)FastCar · VBench prompts (10% replay)LPIPS: 32.94 -> 27.31Improved
use consistent instead of layer-wise inconsistent thresholds at 10% replay (ablation)FastCar · VBench prompts (10% replay)VBench total: 71.8% -> 73.4%Improved
lower threshold tau below -2.5 (higher replay ratio) (ablation)FastCar · VBench promptsVBench total: 71.4% -> 71.5%No clear effect
increase replay ratio from 10% to 80%FastCar · VBench promptsLatency (s): 629.1 -> 390.7Improved
increase replay ratio from 10% to 80%FastCar · VBench promptsVBench total: 73.4% -> 71.5%Hurt
replace sparse attention (local 128) with FastCar at 80% replayFastCar vs StreamingLLM · VBench promptsVBench total: 60.7% -> 71.5%Improved
shrink sparse attention local size from 256 to 16StreamingLLM · VBench promptsLatency (s): 670.5 -> 662.7No clear effect
shrink sparse attention local size from 256 to 128StreamingLLM · VBench promptsVBench total: 72.1% -> 60.7%Hurt
add FastCar (87% replay) to sparse attention local 16StreamingLLM local 16 · VBench promptsVBench total: 64.5% -> 71.6%Improved
add FastCar (87% replay) to sparse attention local 16StreamingLLM local 16 · VBench promptsLatency (s): 662.7 -> 324.31Improved
raise replay in FastCar+SA (local 256) from tau=-1 to tau=-4FastCar + sparse attention local 256 · VBench promptsVBench total: 67.7% -> 71.8%Improved

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