Tang’s AI Lab

See also: Team.

Publications

2026

• Reconciling Set-Valued Policy & Dead-End Discovery in Healthcare Reinforcement Learning: An Empirical Analysis

  Victor Li, Sixing Wu, Shengpu Tang
  

  Pacific Symposium on Biocomputing (PSB) 2027

• Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents
  [link]
  
  

  Kaixuan Liu, Guojun Xiong, Weinan Zhang, Shengpu Tang

  NeurIPS 2026.

• Provably Efficient Model-free Representation Learning for Low-Rank Constrained Markov Decision Processes

  Kaixuan Liu, Guojun Xiong, Shengpu Tang, Wanyun Si, Jian Li
  

  NeurIPS 2026

• GlucoFM-Bench: Benchmarking Time-Series Foundation Models for Blood Glucose Forecasting
  [link]
  
  

  Baiying Lu, Zhaohui Liang, Ryan Pontius, Shengpu Tang, Temiloluwa Prioleau
  

  NeurIPS 2026 Evaluation and Dataset Track

• Beyond Rigid Guidelines: Learning the Nuance of Real-World Breast Cancer Imaging Surveillance

  Hanqi Chen, Anthony Girard, Arsh Shah, Rou-Zhen Chen, Winnie Lau, Ravi B Parikh, Paramita Chatterjee, Shengpu Tang

  IEEE EMBS BHI 2026

• When to Trust the Model, When to Draw Blood: a clinical decision utility framework for evaluating laboratory-value prediction models
  [link]

  Xintong Li, Minxiao Wang, Runze Yan, Stephanie R. Brown, Mark V. Mai, Sivasubramanium V. Bhavani, Xiao Hu, Shengpu Tang

  In submission

  Also abstract at ICCAI 2026 (24th International Conference on Complex Acute Illness).

• Off Policy Evaluation Under Temporal Mismatch
  [link]
  
  

  Ritam Majumdar, Shengpu Tang, Sonali Parbhoo
  

  RLC 2026 Finding the Frame Workshop

• Clarifying Uncertainty Quantification in Off-Policy Evaluation: Beyond Effective Sample Sizes, Towards Confidence Intervals
  [link]
  
  

  Aditya Dutta, Kaixuan Liu, Shengpu Tang
  

  ICML 2026 “DEMO” Workshop on Decision-Making from Offline Datasets to Online Adaptation: Black-Box Optimization to Reinforcement Learning

• CANDOR: Counterfactual ANnotated DOubly Robust Off-Policy Evaluation
  [link]
  
  

  Aishwarya Mandyam, Shengpu Tang, Jiayu Yao, Jenna Wiens, Barbara E Engelhardt
  

  CHIL 2026

• Causal Discovery for Efficient Offline RL with Factored Action Spaces
  [link]
  
  

  Cecilia Ehrlichman, Michael Dykstra, Shengpu Tang, Maggie Makar
  

  CLeaR 2026; also at UAI 2026 Workshop on Causality for Decision Making.

• Slow-Fast Policy Optimization: Reposition-Before-Update for LLM Reasoning
  [link]
  
  

  Ziyan Wang, Zheng Wang, Jie Fu, Xingwei Qu, Qi Cheng, Shengpu Tang, Minjia Zhang, Xiaoming Huo
  

  ICLR 2026

2025

• Exploring Time-Step Size in Reinforcement Learning for Sepsis Treatment
  [link]
  
  

  Yingchuan Sun, Shengpu Tang
  

  ML4H 2025 - Proceedings Track. Dec 2025.
  Also presented at workshops: ARLET @ NeurIPS 2025, TS4H @ NeurIPS 2025, RL4RS @ RLC 2025

• An Adaptive Machine Learning Triage Framework for Predicting Alzheimer’s Disease Progression
  [link]
  
  

  Richard Hou, Shengpu Tang^*, Wei Jin^*
  

  ML4H 2025 - Findings Track. Dec 2025

• Off by a Beat: Temporal Misalignment in Offline RL for Healthcare
  [link]
  
  

  Shengpu Tang, Jiayu Yao, Jenna Wiens, Sonali Parbhoo
  

  RLC 2025 Finding the Frame Workshop; RLC 2025 RL4RS Workshop. Aug 2025.

• Reconciling Set-Valued Policy & Dead-End Discovery in RL: An Empirical Analysis
  [link]
  
  

  Sixing Wu, Shengpu Tang
  

  RLC 2025 CoCoMARL Workshop; RLC 2025 Finding the Frame Workshop. Aug 2025.

• Between Life and Death: Examining Sparse Reward Designs in Healthcare RL
  [link]
  
  

  Yuxuan Shi^*, Matthew Lafrance^*, Shengpu Tang
  

  RLC 2025 RL4RS Workshop. Aug 2025.