AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Full-Atom Peptide Design via Riemannian–Euclidean Bayesian Flow Networks

Hao Qian, Shikui Tu, Lei Xu

AAAI 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Generative protein structure design · 1 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
replace a single Gaussian with a Gaussian mixture for side-chain torsion Bayesian flow (ablation)PepBFN side-chain packing · Side-chain packingcorrect % (within 20 deg): 59.54 -> 75.24Improved
replace a single Gaussian with a Gaussian mixture (ablation)PepBFN side-chain packing · Side-chain packingchi1 MAE (deg): 24.50 -> 10.75Improved
replace a single Gaussian with a Gaussian mixture (ablation)PepBFN side-chain packing · Side-chain packingchi2 MAE (deg): 32.44 -> 12.26Improved
replace a single Gaussian with a Gaussian mixture (ablation)PepBFN side-chain packing · Side-chain packingchi3 MAE (deg): 56.22 -> 34.25Improved
replace a single Gaussian with a Gaussian mixture (ablation)PepBFN side-chain packing · Side-chain packingchi4 MAE (deg): 60.35 -> 53.21Improved
model peptides in fully continuous Bayesian parameter space instead of hybrid continuous-discrete space (ablation)PepFlow · Peptide binder designconvergence speed and sequence stabilityImproved
use PepBFN_sc instead of the best baseline DiffPackDiffPack · Side-chain packingcorrect % (within 20 deg): 62.58 -> 75.24Improved
use PepBFN_seq instead of PepFlow for reverse foldingPepFlow · Peptide reverse foldingAAR %: 53.98 -> 62.46Improved
use PepBFN_seq instead of PepFlow for reverse foldingPepFlow · Peptide reverse foldingHamming diversity: 26.40 -> 13.21Hurt
use PepBFN instead of PepFlow for co-designPepFlow · Peptide sequence-structure co-designbinding affinity %: 21.37 -> 22.26Improved
use PepBFN instead of PepFlow for co-designPepFlow · Peptide sequence-structure co-designRMSD (A): 2.07 -> 3.58Hurt
use PepBFN instead of RFDiffusion for co-designRFDiffusion · Peptide sequence-structure co-designstability %: 26.82 -> 17.56Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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