Tommi Jaakkola

Composing Unbalanced Flows for Flexible Docking and Relaxation

ProtComposer: Compositional Protein Structure Generation with 3D Ellipsoids

Generator Matching: Generative modeling with arbitrary Markov processes

Fictitious Synthetic Data Can Improve LLM Factuality via Prerequisite Learning

Data Distillation for extrapolative protein design through exact preference optimization

An Information Criterion for Controlled Disentanglement of Multimodal Data

Fine-Tuning Discrete Diffusion Models via Reward Optimization with Applications to DNA and Protein Design

Think while You Generate: Discrete Diffusion with Planned Denoising

Symmetry-Driven Discovery of Dynamical Variables in Molecular Simulations

Predicting perturbation targets with causal differential networks

Hierarchical Protein Backbone Generation with Latent and Structure Diffusion

MOFDiff: Coarse-grained Diffusion for Metal-Organic Framework Design

Hamiltonian Score Matching and Generative Flows

Removing Biases from Molecular Representations via Information Maximization

A Recipe for Charge Density Prediction

Equivariant Scalar Fields for Molecular Docking with Fast Fourier Transforms

Conformal Language Modeling

Generative Modeling of Molecular Dynamics Trajectories

In-Context Symmetries: Self-Supervised Learning through Contextual World Models

Deep Confident Steps to New Pockets: Strategies for Docking Generalization

Particle Guidance: non-I.I.D. Diverse Sampling with Diffusion Models

Neural Network Reparametrization for Accelerated Optimization in Molecular Simulations

Improving protein optimization with smoothed fitness landscapes

Harmonic Prior Flow Matching for Multi-Ligand Docking and Binding Site Design

Risk-Controlling Model Selection via Guided Bayesian Optimization