Serena Yeung-Levy

AccidentBench: Benchmarking Multimodal Understanding and Reasoning in Vehicle Accidents and Beyond

Transductive Visual Programming: Evolving Tool Libraries from Experience for Spatial Reasoning

Closing the Modality Gap for Mixed Modality Search

SciVideoBench: Benchmarking Scientific Video Reasoning in Large Multimodal Models

Can Large Language Models Match the Conclusions of Systematic Reviews?

Fine-tuning VLMs Without Forgetting Is Easier Than You Think

Temporal Preference Optimization of Large Multimodal Models

VisualOverload: Probing Visual Understanding of VLMs in Really Dense Scenes

Separating signal from noise: a self-distillation approach for amortized heterogeneous cryo-EM reconstruction

TTRV: Test-Time Reinforcement Learning for Vision Language Models

Three Forms of Stochastic Injection for Improved Distribution-to-Distribution Generative Modeling

CryoHype: Transformer-based hypernetworks for extreme heterogeneity in cryo-EM

CellFlux: Simulating Cellular Morphology Changes via Flow Matching

Video Action Differencing

Video-STaR: Self-Training Enables Video Instruction Tuning with Any Supervision

Foundation Models Secretly Understand Neural Network Weights: Enhancing Hypernetwork Architectures with Foundation Models

Weakly Supervised Understanding of Skilled Human Activity in Videos