Good Papers

Showing papers from Adobe Research, US Show all papers

91%Must read
?Must readVote to see the score
NeurIPS 2026Australian NationalNew YorkAdobe ResearchAdobeAdobe Research, USDiffusion models

Improved Baselines with Representation Autoencoders

Using summed last-k encoder layers and combining RAE with REPA, RAEv2 achieves state-of-the-art gFID of 1.06 in 80 epochs with 10x faster convergence and free guidance.

Jaskirat Singh, Boyang Zheng, Zongze Wu, Richard Zhang and 2 more

Paris Poster Session 3, Thu, Dec 10, 12:30 PM–2:30 PM, Paris Poster Hall · Published 2026

– ReadersNo votes yet
17/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 17 of 20 reviewers recommend it
lenient 3/5
medium 10/10
strict 4/5
86%Must read
?Must readVote to see the score
NeurIPS 2026CMUAdobe ResearchAdobeCarnegie MellonAdobe Research, USVision-language models

The Many Senses of Visual Similarity: A Text-Prompted Image Perceptual Metric

A dataset of multi-aspect human visual similarity judgments benchmarks vision-language models and yields the TPIPS metric, which aligns with human perception and enables text-guided image retrieval and generative evaluation.

Sheng-Yu Wang, Yotam Nitzan, Aaron Hertzmann, Jun-Yan Zhu and 3 more

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

– ReadersNo votes yet
14/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 14 of 20 reviewers recommend it
lenient 5/5
medium 7/10
strict 2/5
86%Must read
?Must readVote to see the score
NeurIPS 2026Massachusetts Institute of TechnAdobe ResearchAdobeAdobe Research, USMITDiffusion models

End-to-End Training for Unified Tokenization and Latent Denoising

UNITE unifies tokenization and latent diffusion via a shared generative encoder, enabling single-stage joint training from scratch without adversarial losses or pretrained encoders to reach near state-of-the-art FID scores.

Shivam Duggal, Xingjian Bai, Zongze Wu, Richard Zhang and 4 more

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

– ReadersNo votes yet
14/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 14 of 20 reviewers recommend it
lenient 4/5
medium 8/10
strict 2/5