Research

A curated list of my publications and invited talks. Author order and venues mirror the CV.

2025

  1. arXiv
    Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains
    A. Gunjal, A. Wang, Elaine Lau, V. Nath, Y. He, B. Liu, S. Hendryx
    arXiv preprint, 2025.
    arXiv PDF OpenReview Project
  2. ICLR
    Refusal-Trained LLMs Are Easily Jailbroken As Browser Agents
    P. Kumar, Elaine Lau, S. Vijayakumar, T. Trinh, … Z. Wang
    International Conference on Learning Representations (ICLR), 2025.
    arXiv OpenReview Code
  3. Under Review
    Adaptive Guidance Accelerates Reinforcement Learning of Reasoning Models
    V. Nath, Elaine Lau, A. Gunjal, M. Sharma, N. Baharte, S. Hendryx
    Under review, 2025.

2024

  1. NeurIPS
    QGFN: Controllable Greediness with Action Values
    Elaine Lau, S. Z. Lu, L. Pan, D. Precup, E. Bengio
    Neural Information Processing Systems (NeurIPS), 2024.
    arXiv NeurIPS Code

2023

  1. NeurIPS WS
    DGFN: Double Generative Flow Networks
    Elaine Lau, N. M. Vemgal, D. Precup, E. Bengio
    NeurIPS 2023 Workshop.
    arXiv
  2. ICML WS
    An Empirical Study of the Effectiveness of Using a Replay Buffer on Mode Discovery in GFlowNets
    N. Vemgal, Elaine Lau, D. Precup
    ICML 2023 Workshop.
    arXiv
  3. AAAI
    Deep Conservative Reinforcement Learning for Personalization of Mechanical Ventilation Treatment
    F. Kondrup*, T. Jiralerspong*, Elaine Lau*, N. de Lara, J. Shkrob, M. D. Tran, D. Precup, S. Basu
    AAAI, 2023. *Equal contributions.
    PDF

2022

  1. ICLR
    Policy Gradients Incorporating the Future
    D. Venuto, Elaine Lau, D. Precup, O. Nachum
    International Conference on Learning Representations (ICLR), 2022.
    arXiv Talk
  2. RLDM
    DeepVent: Deep Conservative Reinforcement Learning for Personalization of Mechanical Ventilation Treatment
    F. Kondrup, T. Jiralerspong, Elaine Lau, N. de Lara, J. Shkrob, M. D. Tran, D. Precup, S. Basu
    Reinforcement Learning and Decision Making (RLDM), 2022.
    Coverage PDF
  3. EMNLP (Ind.)
    Bringing the State-of-the-Art to Customers: A Neural Agent Assistant Framework for Customer Service Support
    S. Obadinma, F. K. Khattak, S. Wang, T. Sidhom, Elaine Lau, S. Robertson, … F. Rudzicz, E. Dolatabadi
    EMNLP 2022 Industry Track.
    arXiv

2021

  1. Long-PEGASUS
    Abstractive Summarization using Longformer-PEGASUS
    A. Jeeson-Daniel, C. Lin, Elaine Lau
    Project report, 2021.
    PDF

Patent

  1. US
    Heuristic-Systematic Decision-Making Model for User Feedback Based on User Behavior and System Telemetry
    N. Chorakhalikar, Elaine Lau, A. Arunachalam, B. Todur
    US Patent App. US20240037346A1.
    Patent

Invited Talks

  1. AAAI WS
    Refusal-Trained LLMs Are Easily Jailbroken As Browser Agents
    P. Kumar, Elaine Lau, S. Vijayakumar, T. Trinh, … Z. Wang
    AAAI @ Web Agents, 2025 — Invited Talk.