7 papers
SurgAtlas: A Large-Scale Surgical Video-Language Dataset with 2,391 Hours of Open and Minimally Invasive Surgery
Filippos Bellos, Andre S. Gala-Garza, Miaowei Wang +8
We introduce SurgAtlas, the largest surgical video-language dataset to date, comprising 15,291 videos (2,391 hours) spanning 18 surgical specialties and over 5,000 procedure types,…
NTILC: Neural Tool Invocation via Learned Compression
Andrew Krikorian, Yayuan Li, Jason J. Corso
Agentic tool-calling language models depend on large registries of callable APIs, functions, and local actions. Placing full tool specifications directly in the prompt incurs a cos…
EchoVQA: Enabling Conversational Assistance for Point-of-Care Cardiac Ultrasound
Filippos Bellos, Yutong Li, Jessie N Dong +6
Point-of-care transthoracic echocardiography (TTE) enables cardiac assessment in virtually any clinical setting, yet its diagnostic utility remains constrained by the expertise req…
Substantial, Decomposable, and Invisible: Visual Context Misalignment in Instructional Videos for Physical Tasks
Yayuan Li, Chenglin Li, Jingying Wang +3
Instructional videos are the dominant medium for learning physical tasks, yet they rarely match the user's real-world visual context. Motor simulation and cognitive load theories p…
When to Think and When to Look: Uncertainty-Guided Lookback
Jing Bi, Filippos Bellos, Junjia Guo +8
Test-time thinking (that is, generating explicit intermediate reasoning chains) is known to boost performance in large language models and has recently shown strong gains for large…
Towards Consistent Long-Term Pose Generation
Yayuan Li, Filippos Bellos, Jason Corso
Current approaches to pose generation rely heavily on intermediate representations, either through two-stage pipelines with quantization or autoregressive models that accumulate er…