Research map / Image generation / Autoregressive and latent generation
Graph Denoising Diffusion for Inverse Protein Folding
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| use GraDe-IF vs PiFold | CATH short · CATH short | recovery rate %: 39.84 -> 45.27 | Helped |
| use GraDe-IF vs PiFold | CATH single-chain · CATH single-chain | recovery rate %: 38.53 -> 42.77 | Helped |
| use GraDe-IF vs PiFold | TS50 · TS50 | recovery %: 58.72 -> 56.32 | Hurt |
| use BLOSUM transition matrix instead of uniform (ablation) | GraDe-IF · CATH | recovery rate: +2% | Helped |
| add equivariance to denoising network (ablation) | GraDe-IF · CATH | recovery rate | Helped |
| add secondary structure conditioning (ablation) | GraDe-IF · CATH | recovery rate | Helped |
| skip 100 sampling steps with discrete DDIM instead of 1 (ablation) | GraDe-IF · CATH all | recovery rate %: 52.21 -> 47.66 | Hurt |
| use GraDe-IF at low recovery level vs ProteinMPNN | GraDe-IF · Diversity | diversity: 0.51 -> 0.61 | Helped |
Closest papers by meaning
- Bridge-IF: Learning Inverse Protein Folding with Markov Bridges (NeurIPS 2024 · similarity 0.91)
- Spectral Feedback for Test-Time Alignment of Protein Diffusion Models (NeurIPS 2026 · similarity 0.90)
- Fine-Tuning Discrete Diffusion Models via Reward Optimization with Applications to DNA and Protein Design (ICLR 2025 · similarity 0.89)
- ProtInvTree: Deliberate Protein Inverse Folding with Reward-guided Tree Search (NeurIPS 2025 · similarity 0.89)
- Protein Inverse Folding From Structure Feedback (NeurIPS 2025 · similarity 0.89)
- Steering Masked Discrete Diffusion Models via Discrete Denoising Posterior Prediction (ICLR 2025 · similarity 0.88)
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.