AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

FASTer: Toward Powerful and Efficient Autoregressive Vision–Language–Action Models with Learnable Action Tokenizer and Block-wise Decoding

Yicheng Liu, Shiduo Zhang, Zibin Dong, Baijun Ye, Tianyuan Yuan, Xiaopeng Yu, Linqi Yin, Chenhao Lu, Junhao Shi, Luca Jiang-Tao Yu, Liangtao Zheng, Jingjing Gong and 3 more

ICLR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models

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 block-wise autoregressive decoding (ablation)FASTer · LIBERO (average)success rate (%): 95.4 -> 97.9Improved
add block-wise autoregressive decoding (ablation)FASTer · Simpler-Bridge (average)success rate (%): 81.0 -> 87.9Improved
replace token-wise AR with block-wise decoding (ablation)FASTer · LIBEROlatency (ms): 323 -> 140Improved
replace token-wise AR with block-wise decoding (ablation)FASTer · LIBEROsuccess rate (%): 95.5 -> 96.7Improved
add pretrained action expert (ablation)FASTer · LIBEROsuccess rate (%): 95.5 -> 97.9Improved
add action expert without pretraining (ablation)FASTer · LIBEROsuccess rate (%): 95.5 -> 94.8Hurt
add action expert trained from scratch (ablation)FASTer · simpler-widowsuccess rate (%): 75.6 -> 23.6Hurt
add pretrained action expert (ablation)FASTer · simpler-widowsuccess rate (%): 75.6 -> 87.9Improved
use TAAE tokenizer instead of CNN (ablation)FASTerVQ · LIBEROsuccess rate (%): 96.2 -> 97.9Improved
increase codebook size from 4096 to 8192 (ablation)FASTerVQ · LIBEROsuccess rate (%): 97.9 -> 96.3Hurt
increase residual codebooks from 1 to 3 (ablation)FASTerVQ · LIBEROsuccess rate (%): 93.4 -> 97.9Improved
increase residual codebooks from 3 to 8 (ablation)FASTerVQ · LIBEROsuccess rate (%): 97.9 -> 96.6Hurt

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