AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

CELLE-2: Translating Proteins to Pictures and Back with a Bidirectional Text-to-Image Transformer

Emaad Khwaja, Yun S. Song, Aaron Agarunov, Bo Huang

NeurIPS 2023 · Image generation › Autoregressive and latent generation · approach: Diffusion for protein and antibody design · 0 citations (OpenAlex)

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
increase transformer depth (5 to 68 layers, with smaller hidden size) (ablation)CELL-E 2 on HPA · HPA validationNucleus Proportion MAPE: 0.0818 ± 0.0794 -> 0.0257 ± 0.0250Improved
increase transformer depth (5 to 68 layers, with smaller hidden size) (ablation)CELL-E 2 on OpenCell · OpenCell validationNucleus Proportion MAPE: 0.0584 ± 0.0511 -> 0.0161 ± 0.0148Improved
hidden size 640 instead of 480 (ablation)CELL-E 2 · OpenCell validationImage MAE: 0.4953 ± 0.0064 -> 0.4995 ± 0.0006No clear effect
larger ESM-2 embedding size (480 to 2560) (ablation)CELL-E 2 on HPA · HPA masked sequence in-fillingSequence MAE: 0.8628 ± 0.0951 -> 0.5759 ± 0.2322Improved
larger ESM-2 embedding size (480 to 2560) (ablation)CELL-E 2 on HPA · HPA masked sequence in-fillingCosine similarity: 0.9504 ± 0.0237 -> 0.9722 ± 0.0210Improved
larger ESM-2 embedding size (480 to 2560) (ablation)CELL-E 2 on OpenCell · OpenCell masked sequence in-fillingSequence MAE: 0.7507 ± 0.1709 -> 0.4950 ± 0.2456Improved
finetune HPA model on OpenCell with finetuned VQGANs (ablation)CELL-E 2 hidden 480 · OpenCell validationImage MAE: 0.4154 ± 0.0594 -> 0.3449 ± 0.1305Improved
finetune HPA model on OpenCell with finetuned VQGANs (ablation)CELL-E 2 hidden 480 · OpenCell validationInception score: 2.1739 ± 0.1255 -> 3.6083 ± 0.2013Improved
finetune HPA model on OpenCell with finetuned VQGANs (ablation)CELL-E 2 hidden 480 · OpenCell validationFID: 3.9509 -> 19.2683Hurt
swap HPA VQGAN for OpenCell VQGAN during finetuning (ablation)HPA Finetuned · OpenCell validationImage MAE: 0.3776 ± 0.0834 -> 0.4996 ± 0.0006Hurt
finetune on OpenCell (ablation)CELL-E 2 · OpenCell validationsequence reconstructionNo clear effect
replace CELL-E with CELL-E 2 (one-step non-autoregressive generation)CELL-E 2 hidden 480 · OpenCellgeneration time: 65x fasterImproved

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