Generative modeling of intrinsically disordered protein regions by reinforcing sparse autoencoder features
cs.LG
Submitted: 2026-10-01
Updated: 2026-10-01
Code: https://github.com/rotskoff-group/idiom
Terminology
Sources
- Efficient Training of Language Models to Fill in the Middle
- Sparse Autoencoders Find Highly Interpretable Features in Language Models
- FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
- La-Proteina: Atomistic Protein Generation via Partially Latent Flow Matching
- GLU Variants Improve Transformer
- Guiding Generative Protein Language Models with Reinforcement Learning
- RoFormer: Enhanced Transformer with Rotary Position Embedding
- Interpreting and Steering Protein Language Models through Sparse Autoencoders
- A Comparative Analysis of Sparse Autoencoder and Activation Difference in Language Model Steering
- Interpretable Reward Model via Sparse Autoencoder
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks