Unlocking Any-Order Generation in Pretrained Autoregressive Image Models
READ THE ORIGINAL SOURCEOpen original sourcehuggingface.co · Opens in a new tabAutoregressive image generators trained in a fixed raster-scan order cannot inpaint, outpaint, or edit at inference time. The paper proposes adapting pretrained raster-scan models into any-order generators using future-aware positional embeddings and an ordering curriculum that gradually shifts training from raster order to random orders. Applied to four base models (LlamaGen-L/XL, RAR-L/XL), the authors report outperforming MaskGIT on ImageNet inpainting (FID 6.21 vs. 10.44), competitiveness with MAR-L, a 1.2-1.9 class-conditional FID increase, and 13-33% extra compute beyond pretraining.
One primary source; other links are context, not independent corroboration.
- Primary sourceHugging Face Daily Papers
- Discovery linkHugging Face
- Linked in sourceView arXiv page
- Linked in sourceView PDF
The paper page is linked. Peer review and findings have not been independently checked by AINewsIQ.
Open full article view →- Original paper+45
- Fetched source page+30
- No editor link review+0
- Original non-discussion link+5
This measures the stored source trail and editorial steps. It does not certify accuracy, peer review, or independent corroboration. Popularity adds no points.
View full scoring →Read the scoring rubric →What the source says
Autoregressive image generators trained in a fixed raster-scan order cannot inpaint, outpaint, or edit at inference time. The paper proposes adapting pretrained raster-scan models into any-order generators using future-aware positional embeddings and an ordering curriculum that gradually shifts training from raster order to random orders. Applied to four base models (LlamaGen-L/XL, RAR-L/XL), the authors report outperforming MaskGIT on ImageNet inpainting (FID 6.21 vs. 10.44), competitiveness with MAR-L, a 1.2-1.9 class-conditional FID increase, and 13-33% extra compute beyond pretraining.
A READER'S EVIDENCE CHECKHow to read this signalSix questions about claims, sources, risks, and open checks
This is a signal-level guide until claim-level evidence is available. “Not recorded” means the evidence has not been assessed here; it does not mean a result is false. How we check AI news →
- 01Who is making each claim?
The claimant and whether each claim was measured or asserted are not recorded in this public brief.
- 02What was actually tested?
Not recorded per claim. Check the source for sample size, tasks, model and benchmark versions, and reproducibility.
- 03Where are the sources?
Original link: huggingface.co. Brief basis: fetched source page. Primary page date and quote locations are not established in this panel.
- 04What could mislead?
Source claims not independently checked; results are reported on ImageNet and four base models.
- 05What is at risk if wrong?
Not assessed per claim. Consider safety, privacy, cost, and adoption decisions before relying on the result.
- 06What would change the view?
No claim-specific check is recorded. An inspectable primary method or independent replication could change the assessment.
Key findings from the stored brief
- Adapting pretrained raster-scan AR models into any-order generators reportedly costs 13-33% additional compute beyond pretraining.
- The method reports ImageNet inpainting FID of 6.21 versus 10.44 for MaskGIT.
- Class-conditional FID rises by 1.2-1.9 as a tradeoff for the added flexibility.
- The standard AR inference stack is left unchanged, per the paper.
Source claims not independently checked; results are reported on ImageNet and four base models.
This is a source-based abstract, not independent validation of the source's claims.
Evidence & sources
These links show where the signal came from and what the source page linked to. A linked page is context, not independent corroboration.
Related archive coverage
These relationships record a citation or topic overlap, not confirmation of the finding.
The paper page is linked. Peer review and findings have not been independently checked by AINewsIQ.
Timeline & updates
- Signal dated by source or discovery feed
This timestamp places the signal in the radar. It may be the discovery post date rather than the primary page's first publication date.
- Shared brief stored
AINewsIQ prepared one abstract from fetched source page.
A publication revision history is not stored for this signal. New engagement snapshots do not mean the source itself changed.
Why this may matter
Adding inpainting, outpainting, and editing to existing autoregressive image models without training from scratch could reduce the cost of new capabilities and keep compatibility with existing AR inference tooling.
Limits and open questions
Source claims not independently checked; results are reported on ImageNet and four base models.
Organization-specific financial, security, and operational effects require the organization's own context and review.
Discussion
Comments are reviewed before publication. Discuss the evidence, not the person.
No published comments yet.
Reader discussion opens with accounts. You can read the original source and published comments now.