AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

LISA: Locality-Informed Speculative Decoding for Accelerating Autoregressive Image Generation

Ying Li, Siyong Jian, Zhaode Wang, Zhiwen Chen, Chengfei Lyu, Huan Wang

ECCV 2026 · Image generation › Autoregressive and latent generation · approach: Accelerating autoregressive image generation · 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
add geometry-aware soft verification (GASV) (ablation)LISA drafter · LlamaGen-XL-T2Iaccept length: 1.24 -> 2.92Improved
add GASV (ablation)LISA drafter · LlamaGen-XL-C2Iaccept length: 1.07 -> 2.45Improved
add GASV (ablation)EAGLE-3 drafter · LlamaGen-XL-T2Iaccept length: 1.00 -> 1.32Improved
add GASV (ablation)EAGLE-3 drafter · LlamaGen-XL-C2Iaccept length: 1.00 -> 1.25Improved
use locality-informed distilled drafter instead of EAGLE-3 (standard verification) (ablation)LISA · LlamaGen-XL-T2Iaccept length: 1.00 -> 1.24Improved
use locality-informed distilled drafter instead of EAGLE-3 (standard verification) (ablation)LISA · LlamaGen-XL-C2Iaccept length: 1.00 -> 1.07Improved
increase drift threshold kappa from 0.6 to 1.0 (ablation)LISA · LlamaGen-XL-T2Iaccept length: 2.16 -> 2.92Improved
increase cluster-posterior threshold lambda from 0.4 to 0.8 (ablation)LISA · LlamaGen-XL-T2Iaccept length: 2.92 -> 2.49Hurt
loosen verification (kappa=1.1, lambda=0.4 vs kappa=1.0, lambda=0.6) (ablation)LISA · LlamaGen-XL-C2Ispeedup and FID: 1.39x -> 1.53x speedup; FID 10.50 -> 12.60Mixed
apply locality-informed distillation to Medusa-style drafter (ablation)Medusa-style drafter · LlamaGen-XL-T2Ispeedup: 1.04x -> 1.37xImproved
add GASV to LISA-Medusa (ablation)LISA-Medusa · LlamaGen-XL-T2Ispeedup: 1.37x -> 1.82xImproved
use LISA instead of EAGLE-3Lumina-mGPT-7B · MS-COCOspeedup: 1.86x -> 4.12xImproved

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