2026 presentations
7 August 2026
Presentation 1
Paediatric-HGNN: A Hybrid Heterogeneous Graph Neural Network for Detecting Disfluency in Children's Speech via Multiscale Acoustic Fusion
Speaker
Rashini Liyanarachchi
PhD student听
Presentation 2
TriageSim: A Conversational Emergency Triage Simulation Framework from Structured Electronic Health Records
Speaker
Dipankar Srirag
PhD student听
听
31 July 2026
Presentation & Discussion:听
Learning how to learn in AI era: finding your unknowns
24 July 2026
Discussion:听
(Part 2)
Unsupervised Learning: With Jacob Effron
17 July 2026
Discussion:听
Unsupervised Learning: With Jacob Effron
10 July 2026
Presentation:听
Teaching and Learning in the Age of Generative AI
Presenter
Dr Oscar P茅rez Concha
91色情片
26 June 2026
Presentation:听
Judging the LLM-as-judge in clinical tasks
Presenter
Larry Bi
听
5 June 2026
Presentation:听
The Sisyphean Cycle of LLM Evaluation: Saturate, Rebuild, Repeat
Presenter
Larry Bi
听
22 May & 29 May 2026
Discussion on:听
Stanford Online
1 May 2026
Paper Presentation & Discussion:听
Article author
Mohammad Asadi, et al., Stanford University
24 April 2026
Presentation:听
Harmonising the Clinical Melody: Tuning Large Language Models for Hospital Course Summarisation in Clinical Coding
Presenter
Larry Bi
10 April 2026
Presentation:听
Presenter
Stanford Online
27 March 2026
Presentation:听
Presenter
Stanford Online
13 March 2026
Presentation:听
The Anthropic Precedent: Artificial Intelligence, Clinical Autonomy, and the Governance Vacuum
Presenter
Larry Bi
6 March 2026
Presentation:听
The Crab That Fought Back: Autonomous Agents, Open Source, and Why Healthcare AI Should Be Paying Attention.
Presenter
Sandy Sa
27 February 2026
Presentation:听
AI in 2026: Artificial Intelligence or Automated Irresponsibility
Presenter
Larry Bi
听
20 February 2026
Presentation:听
Hertion AI: A Unified Multi Agent System for Health Data Research and Preprocessing
Short Description: Hertion AI is an idea that emerged from our experience as research student that a significant amount of research time is spent on data cleaning, cohort construction, and preparing datasets before analysis can even begin. The system uses RAG and a multi agent AI architecture to support metadata understanding, cohort suggestions, variable derivation, data cleaning, and literature informed research planning. The goal is to streamline clinical research workflows and reduce manual data wrangling time while maintaining strict data governance.
Presenter
Gege Ardiansyah (Gege)
Team members
Fatimah Azzakiyah (Fatimah)听
Mifetika Lukitasari (Tika)
Ravicha Suksawasadi (Palm)
