AI Paper Atlas

Research map / Video / Video generation

What Remains in Sight? Autoregressive Video Decoding as Representation-Guided Context Rewriting

Lixuan He, Jihyeon Je, Yang You, Leonidas Guibas

NeurIPS 2026 · Video › Video generation · approach: Autoregressive long video generation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

The paper reframes autoregressive long-horizon video decoding as representation-guided rewriting of the visible context, deciding what past evidence remains in sight rather than relying on implicit recent windows or caches.

Problem

Autoregressive long-horizon video generation typically decides the visible past implicitly through recent windows, compressed KV caches, or external memory, rather than explicitly selecting what evidence to keep.

Approach

Treats each autoregressive decoding step as explicitly rewriting the visible context guided by learned representations, so the model chooses which past content stays in sight for future prediction.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ReCR context rewritingSelf-Forcing · VBench-Long 60sTotal Score: 78.19 -> 81.60Improved
add ReCR context rewritingCausal-Forcing · VBench-Long 60sTotal Score: 77.41 -> 82.08Improved
add ReCR context rewritingCausal-Forcing · VBench-Long 60sDynamic Degree: 56.82 -> 80.47Improved
add ReCR context rewritingLongLive · VBench-Long 30sTotal Score: 81.32 -> 81.60No clear effect
add ReCR to training-free Deep ForcingDeep Forcing · VBench-Long 60sTotal Score: 81.42 -> 82.27Improved
add ReCR context rewritingSelf-Forcing · MovieGen-128 30s driftaesthetic drift: 5.81 -> 3.99Improved
add ReCR to Deep ForcingDeep Forcing · MovieGen-128 30s driftaesthetic drift: 3.56 -> 1.76Improved
add ReCR context rewritingLongLive · MovieGen-128 30s driftTA tail: +0.37 -> +1.05Hurt
add ReCR context rewritingSelf-Forcing · VBench-Long 60sSubject Consistency: 96.67 -> 96.43Hurt
add ReCR context rewritingSelf-Forcing · 60s block latencyms: 728 -> 712Improved
add ReCR context rewritingInfinity-RoPE · 60s block latencyms: 667 -> 710Hurt
use ReCR selection instead of contiguous recent, full, or random intermediate context (ablation)ReCR · 60s rollouts under perturbationquality-score dropImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 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