ProAR: Learning Prospective Reasoning with Autoregressive Video Models

ProAR introduces goal-frame prediction and future self-alignment to enable goal-directed reasoning in autoregressive video models, surpassing baselines with 25% training steps.

Linghui Shen, Tinghui Zhu, Sheng Zhang, Muhao Chen

Published Oct 2, 2026 · 0 citations · ▲ 26 on Hugging Face · Code ★ 2

Pixel-space Autoregressive Image Synthesis via Spectrum Serialization and Flow-based Refinement

Guiwei Zhang, Tianyu Zhang, Yalong Bai, Ying Ba and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

BAL: Bidirectional Autoregression in Latent Space for Learning Human Movement Representations

Genki Kinoshita, Ko Nishino

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Collapse Hunter: Tackling the Dimensional Degeneration in Generative Ranking

Haoran Xin, Junwei Pan, Yongqi Zhou, Tianqu Zhuang and 7 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Demystifying Classifier-Free Guidance for Auto-Regressive Image Generation

Zhiling Zhou, Jiachun Pan, Fengzhuo Zhang, Dirk Bergemann and 1 more

Atlanta Poster Session 4, Thu, Dec 10, 4:30 PM–7:30 PM, Hall C1 · Published 2026

Context-Aware Autoregressive Image Generation for Emerging Reasoning Properties

Jixuan Ying, Haoyu Liu, Timing Yang, Tingyu Zhu and 6 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

XTraj: A Coarse-to-Fine Autoregressive Framework for Transferable Trajectory Generation

wang chao, ZHIQUAN LAI, Xinwei Fang, Yuanshao Zhu and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Localizing Concepts in Visual Autoregressive Models

Nanxiang Jiang, Yang Liu, Yuanhao Wang, Min Xu

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026

Rethinking Memorization–Generalization Trade-Off in Generative Models

Jiseok Chae, Kyuwon Kim, Donghwan Kim

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

RefineTok: Scale-Wise Tokenization for Progressive Visual Refinement

Yitian Zhang, Long Mai, Yizhou Wang, Yun Fu

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026

Unlocking Any-Order Generation in Pretrained Autoregressive Image Models

Rishav Pramanik, Marco Pedersoli, Zhaozheng Yin

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

BitDance: Scaling Autoregressive Generative Models with Binary Tokens

BitDance is an autoregressive image generator that predicts binary visual tokens via a diffusion head and next-patch decoding, achieving state-of-the-art FID with far fewer parameters and much faster inference.

Yuang Ai, Jiaming Han, Shaobin Zhuang, Weijia Mao and 7 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 46 on Hugging Face · Code ★ 485

ExtraVAR: Stage-Aware RoPE Remapping for Resolution Extrapolation in Visual Autoregressive Models

ExtraVAR proposes stage-aware RoPE remapping and entropy-driven attention calibration to eliminate repetition and detail degradation when extrapolating VAR models to higher resolutions without retraining.

Feihong Yan, Shaoyu Liu, Haixuan Wang, Shuai Lu and 4 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Guiding Visual Autoregressive Models through Spectrum Weakening

Spectrum-weakening guidance constructs a weak visual autoregressive model via spectral selection to improve unconditional quality and conditional alignment without retraining or architectural changes.

Chaoyang Wang, Tianmeng Yang, Yunhai Tong

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder

IDEAL aligns discrete visual tokens with shallow and deep vision features to preserve both semantics and fine details, achieving 0.61 rFID and 1.89 gFID.

Yitong Chen, Zijie Diao, Junke Wang, Lingyu Kong and 4 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 3 on Hugging Face · Code ★ 6

DrawingsDreamer: A Unified Multi-View Engineering Drawings Generation Model

DrawingsDreamer is a unified LLM-driven sequence model that generates multi-view engineering SVG drawings with high geometric fidelity and cross-view alignment via hierarchical tokenization and progressive training.

Shurui Liu, Weide Chen, Changwang Yi, Ancong Wu

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Channel-wise Vector Quantization

Channel-wise Vector Quantization replaces patch tokens with channel tokens to achieve full codebook usage and improves reconstruction and text-to-image generation via sequential channel prediction.

Wei Song, Tianhang Wang, Yitong Chen, Zuxuan Wu and 4 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 14 on Hugging Face · Code ★ 11

StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

StableVQ decouples encoder-decoder and codebook training via Dynamic STE, Region VQ Loss, and independent schedules to stabilize VQ tokenizers and boost utilization and reconstruction.

Bao Tang, Jiahao Guo, Haoxiang Cao, Wenyu Liu and 3 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

NSARM: Next-Scale Autoregressive Modeling for Robust Real-World Image Super-Resolution

NSARM uses two-stage next-scale autoregressive modeling for real-world super-resolution, achieving robust, fast, high-quality results across varying degradations.

Xiangtao Kong, Rongyuan Wu, Shuaizheng Liu, Lingchen Sun and 1 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

CausalCine: Real-Time Autoregressive Generation for Multi-Shot Video Narratives

CausalCine enables real-time interactive multi-shot video generation via causal modeling, content-aware memory routing, and few-step distillation, surpassing autoregressive baselines with streaming interactivity.

Yihao Meng, Zichen Liu, Hao Ouyang, Qiuyu Wang and 7 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 30 on Hugging Face

Balancing Image Compression and Generation with Bootstrapped Tokenization

SelfBootTok decomposes image tokens into global and local groups via self-bootstrapped learning to shift detail burden to the tokenizer, cutting generator computation by ~40% while achieving state-of-the-art 1.56 gFID with 64 tokens.

Haozhe Chi, Jinghan Li, Hao Jiang, Wu Sheng and 3 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 1 on Hugging Face

ActWorld: From Explorable to Interactive World Model via Action-Aware Memory

ActWorld extends interactive world models to object interaction via a 100K dataset and hierarchical action-aware memory, improving fidelity over navigation-only baselines.

Zhexiao Xiong, Yizhi Song, Hao Kang, Qing Yan and 9 more

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026 · ▲ 8 on Hugging Face

ClothTransformer: Unified Latent-Space Transformers for Scalable Cloth Simulation

ClothTransformer reformulates cloth simulation as autoregressive latent-space sequence modeling to unify body-driven garments, robotic manipulation, and collisions with 4-9x lower error.

Yu Zhang, YIDI SHAO, Wenqi Ouyang, Yushi LAN and 4 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 1 on Hugging Face

Vermeer: Autoregressive generative modeling of microscopy predicts protein localization

Vermeer is an autoregressive generative model that predicts protein localization microscopy from sequences and cell landmarks, enabling zero-shot transfer across imaging conditions.

Sandeep Kambhampati, Eric Zimmermann, Emre Hayir, Kevin K Yang and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

AnalogToBi: Device-Level Analog Circuit Topology Generation via Bipartite Graph and Grammar Guided Decoding

AnalogToBi generates device-level analog circuit topologies via bipartite graphs and grammar-guided decoding, achieving high validity and novelty without human-in-the-loop training.

Seungmin Kim, Mingun Kim, Yuna Lee, Yulhwa Kim

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026