BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Toward Generalizable and Intelligent Visual Reasoning Models - Joy
  Hsu\, Stanford University
DTSTART:20260323T140000Z
DTEND:20260323T150000Z
UID:TALK245911@talks.cam.ac.uk
CONTACT:Elliott Wu
DESCRIPTION:Visual reasoning models have made remarkable progress in recen
 t years\, yet they are still not widely deployed in critical real-world se
 ttings—where data is scarce\, tasks are multi-step\, and outputs must be
  inspectable and verifiable. To address this gap\, I propose building mult
 imodal reasoning models with structural priors that can robustly perceive\
 , interpret\, and interact with the physical world under human specified-i
 nstructions. In this talk\, I will cover a spectrum of modeling paradigms 
 and environments: (1) neuro-symbolic models\, where hybrid explicit-implic
 it representations provide efficiency and generalization by design in stru
 ctured settings\; (2) foundation model-distilled frameworks\, which extern
 alize prior knowledge to structure vision-language models’ reasoning pro
 cess in open-ended domains\; (3) structure-induction frameworks\, which us
 e interpretable representational bottlenecks to uncover patterns in comple
 x\, unlabeled visual data. I will conclude by outlining a path toward visu
 al-language models that can generalize across diverse sensing modalities a
 nd conduct intelligent decision-making in the real world.\n\nBio: Joy Hsu 
 is a PhD candidate in Computer Science at Stanford University\, advised by
  Prof. Jiajun Wu. Her research focuses on making visual reasoning models r
 eliable in real-world settings under sensing\, data\, and compute constrai
 nts. She develops multimodal reasoning models with structural priors that 
 enable systems to perceive\, interpret\, and interact intelligently with t
 he physical world across diverse\, data-scarce domains. She is a recipient
  of the Knight-Hennessy Fellowship and the NSF Fellowship\, and was awarde
 d third place in the Amazon Robotics PhD competition and named a Rising St
 ar in AI in 2025.\n\nZoom: https://cam-ac-uk.zoom.us/j/85290977324?pwd=C3K
 ItZS8d2XaVyUsb88HKuV5wWFrYV.1
LOCATION:LT6\, Department of Engineering
END:VEVENT
END:VCALENDAR