Author

Thomas Tian

Observed name: Tian, Thomas

Institutions: Nvidia (United States), University of California, Berkeley

Observed institutions: Nvidia (United States), University of California, Berkeley

OpenAlex ID: A5101252845

Counterfactual VLA: Self-Reflective Vision-Language-Action Model with Adaptive Reasoning

2025 · Peng, Zhenghao "Mark", Ding, Wenhao, et al. · arXiv (Cornell University) · 17

Human-oriented Representation Learning for Robotic Manipulation

2024 · Mingxiao Huo, Mingyu Ding, et al. · Robotics: Science and Systems · 11

TEXEDO : Test Time Scaling for Controller-aware Language-conditioned Humanoid Motion Generation

2026 · Cao, Jianuo, Chen, Yuxin, et al. · arXiv (Cornell University) · 7

VeriFine: Scaling Verification for Self-Improvement in Embodied Reasoning

2026 · Zhou, Zewei, Luo, Rachel, et al. · arXiv (Cornell University) · 3

TacitVLA: Reasoning Vision-Language-Action Models Without Reasoning Latency

2026 · Lu, Jack, You, Yurong, et al. · arXiv (Cornell University) · 2

CLIFT: Turning Gemini Robotics On-Device into Humanoid Specialists via Non-Invasive Closed-Loop Iterative Fine-Tuning

2026 · Chen, Yuxin, Srikanth, Hari, et al. · arXiv (Cornell University) · 2

Should the Action Expert Train the VLM? Action-Token Supervision and Gradient Routing in Mixture-of-Transformer VLAs

2026 · Hancock, Asher James, Li, Boyi, et al. · arXiv (Cornell University)

Works count: 7

Citation count: 42

H-index: 3

I10-index: 2