Ronald Clark

Associate Professor@University of Oxford·英国·OpenReview

研究方向

Computer graphics · rendering · path-tracing · Computer Vision · SLAM · 3D Reconstruction · Numerical optimisation · meta-learning · Robotics · control · inverse-kinematics · animation

4.5

Towards Understanding Multimodal Fine-Tuning: A Case Study into Spatial Features

ICLR 2026Rejected

4.0

IllumiCraft: Unified Geometry and Illumination Diffusion for Controllable Video Generation

NeurIPS 2025Poster

5.0

Large Language Monkeys: Scaling Inference Compute with Repeated Sampling

ICLR 2025Rejected

5.0

REVIP: Rethinking Visual Prompting for Multimodal Large Language Models with External Knowledge

ICLR 2025Withdrawn

2.5

Toward Robust Real-World Audio Deepfake Detection: Closing the Explainability Gap

合作者 (20)

Christian Schroeder de Witt

Ronald Clark

Towards Understanding Multimodal Fine-Tuning: A Case Study into Spatial Features

A Generative Likelihood Framework for High-Resolution Climate Model Evaluation

SpatialThinker: Reinforcing 3D Reasoning in Multimodal LLMs via Spatial Rewards

Cloud4D: Estimating Cloud Properties at a High Spatial and Temporal Resolution

MALT: Improving Reasoning with Multi-Agent LLM Training

IllumiCraft: Unified Geometry and Illumination Diffusion for Controllable Video Generation

Large Language Monkeys: Scaling Inference Compute with Repeated Sampling

REVIP: Rethinking Visual Prompting for Multimodal Large Language Models with External Knowledge

Toward Robust Real-World Audio Deepfake Detection: Closing the Explainability Gap