Pith. sign in

REVIEW 11 cited by

Randomized Autoregressive Visual Generation

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2411.00776 v1 pith:Z2MP5YMO submitted 2024-11-01 cs.CV

Qihang Yu , Ju He , Xueqing Deng , Xiaohui Shen , Liang-Chieh Chen This is my paper · ORCID
classification cs.CV
keywords autoregressivemodelinggenerationimagetrainingcompatibilityfactorizationfull
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

This paper presents Randomized AutoRegressive modeling (RAR) for visual generation, which sets a new state-of-the-art performance on the image generation task while maintaining full compatibility with language modeling frameworks. The proposed RAR is simple: during a standard autoregressive training process with a next-token prediction objective, the input sequence-typically ordered in raster form-is randomly permuted into different factorization orders with a probability r, where r starts at 1 and linearly decays to 0 over the course of training. This annealing training strategy enables the model to learn to maximize the expected likelihood over all factorization orders and thus effectively improve the model's capability of modeling bidirectional contexts. Importantly, RAR preserves the integrity of the autoregressive modeling framework, ensuring full compatibility with language modeling while significantly improving performance in image generation. On the ImageNet-256 benchmark, RAR achieves an FID score of 1.48, not only surpassing prior state-of-the-art autoregressive image generators but also outperforming leading diffusion-based and masked transformer-based methods. Code and models will be made available at https://github.com/bytedance/1d-tokenizer

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. D-AR: Diffusion via Autoregressive Models

    cs.CV 2025-05 conditional novelty 7.0 of 10

    D-AR recasts pixel-space diffusion as vanilla autoregressive next-token prediction using a diffusion-ordered discrete tokenizer, reaching 2.09 FID on ImageNet 256x256 with a 775M Llama backbone.

  2. SSDD: Single-Step Diffusion Decoder for Efficient Image Tokenization

    cs.CV 2025-10 conditional novelty 6.0 of 10

    A GAN-free, single-step distilled diffusion decoder that reconstructs images from latent codes with better perceptual FID than KL-VAE at higher throughput.

  3. Localizing and Mitigating Memorization in Image Autoregressive Models

    cs.LG 2025-08 conditional novelty 6.0 of 10

    Memorization in image autoregressive models sits in early blocks at coarse scales for VAR models and in middle/late blocks for RAR models; halving the flagged neurons' weights cuts extractable images by 65 to 84 percent.

  4. Implementing Adaptations for Vision AutoRegressive Model

    cs.CV 2025-07 conditional novelty 6.0 of 10

    Fine-tuned Vision AutoRegressive models mostly beat a strong diffusion baseline on downstream image generation, but DP fine-tuning yields poor FID scores.

  5. FreeLoRA: Enabling Training-Free LoRA Fusion for Autoregressive Multi-Subject Personalization

    cs.CV 2025-07 conditional novelty 6.0 of 10

    A method for multi-subject image personalization that fuses independently trained LoRA modules at inference time on visual autoregressive models.

  6. Rethinking Discrete Tokens: Treating Them as Conditions for Continuous Autoregressive Image Synthesis

    cs.CV 2025-07 conditional novelty 6.0 of 10

    DisCon treats discrete image tokens as conditioning signals rather than targets, letting a continuous autoregressive model refine details and reach gFID 1.38 on ImageNet-256.

  7. SpectralAR: Spectral Autoregressive Visual Generation

    cs.CV 2025-06 conditional novelty 6.0 of 10

    An autoregressive image generator that tokenizes images in the DCT frequency domain into nested 1D spectral sequences and generates them coarse-to-fine, reaching 3.02 gFID with 64 tokens on ImageNet-1K.

  8. Autoregressive Images Watermarking through Lexical Biasing: An Approach Resistant to Regeneration Attack

    cs.CR 2025-06 conditional novelty 6.0 of 10

    LBW embeds watermarks into autoregressive image token maps by biasing token sampling toward a secret green list and detects them with a z-test on green-token counts.

  9. Grouping First, Attending Smartly: Training-Free Acceleration for Diffusion Transformers

    cs.CV 2025-05 conditional novelty 6.0 of 10

    GRAT accelerates pretrained diffusion transformers by grouping tokens and restricting each group's attention to neighboring blocks or criss-cross rows and columns, achieving large speedups with near-full-attention quality.

  10. Spectral Origins of the Self-Correction Blind Spot in Autoregressive Generation

    cs.LG 2026-07 conditional novelty 5.5 of 10

    Self-correction blind spots in residual-stream autoregressive models arise iff the product of attention Jacobians has spectral radius ≥1, with a sharp marker threshold and RL coupling condition derived from that radius.

  11. Marrying Autoregressive Transformer and Diffusion with Multi-Reference Autoregression

    cs.CV 2025-06 conditional novelty 5.0 of 10

    A hybrid image generator that uses an autoregressive transformer as an encoder and a diffusion model as a decoder reaches FID 1.42 on ImageNet 256x256, with an iterative multi-reference mode that improves quality over...

Pith tools