Javen Qinfeng Shi

Beyond DAGs: A Latent Partial Causal Model for Multimodal Learning

Concept Component Analysis: A Principled Approach for Concept Extraction in LLMs

I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data?

On the Geometry of Uncertainty: Equivariant Uncertainty Estimation for Molecular Vector Properties

Towards Identifiable Latent Additive Noise Models

Identifying Unperturbed Cellular Programs Enables Accurate Single-Cell Perturbation Prediction

On the Value of Cross-Modal Misalignment in Multimodal Representation Learning

Analytic DAG Constraints for Differentiable DAG Learning

Solving the Asymmetric Traveling Salesman Problem via Trace-Guided Cost Augmentation

Coreset Selection via Reducible Loss in Continual Learning

Beyond DAGs: A Latent Partial Causal Model for Multimodal Learning

Turning Challenges into Opportunities: How Distribution Shifts Enhance Identifiability in Causal Representation Learning

The Quest for Winning Tickets in Low-Rank Adapters

Rethinking end-to-end prediction of adsorption energies from a causal perspective