AH
Asher James Hancock
cs.ROcs.AIcs.LGcs.SYeess.SYstat.MLcs.CVmath.OC
On Valency
published · living versionsW_z6b3gjgg·v1 · currentpublished
Should the Action Expert Train the VLM? Action-Token Supervision and Gradient Routing in Mixture-of-Transformer VLAs
with Boyi Li, Yuxiao Chen, Shuhan Tan, Wenhao Ding +4
1 version
Preprints & journals
7 papers in the corpus · 2023–2026LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfer2602.10556v2 · Lihan Zha, Asher J. Hancock, Mingtong Zhang et al.2026 · 0 citationsarXiv
Robotic Video World Models: A Survey of Applications, Research Challenges, Future Directions2601.07823v2 · Zhiting Mei, Tenny Yin, Ola Shorinwa et al.2026 · 0 citationsarXiv
Actions as Language: Fine-Tuning VLMs into VLAs Without Catastrophic Forgetting2509.22195v1 · Asher J. Hancock, Xindi Wu, Lihan Zha et al.2025 · 0 citationsarXiv
Is Your Imitation Learning Policy Better than Mine? Policy Comparison with Near-Optimal Stopping2503.10966v4 · David Snyder, Asher James Hancock, Apurva Badithela et al.2025 · 1 citationarXiv
Run-time Observation Interventions Make Vision-Language-Action Models More Visually Robust2410.01971v1 · Asher J. Hancock, Allen Z. Ren, Anirudha Majumdar2024 · 3 citationsarXiv
Blending Data-Driven Priors in Dynamic Games2402.14174v3 · Justin Lidard, Haimin Hu, Asher Hancock et al.2024 · 5 citationsarXiv
PAC-Bayes Generalization Certificates for Learned Inductive Conformal Prediction2312.04658v1 · Apoorva Sharma, Sushant Veer, Asher Hancock et al.2023 · 3 citationsarXiv
Career total: 7 works. 7 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.