Skip to content
@tmlr-group

TMLR Group

Trustworthy Machine Learning and Reasoning Group

Trustworthy Machine Learning and Reasoning (TMLR) Group, an online-offline-mixed machine learning research group, locates in different cities, including Hong Kong, Melbourne, Shanghai, Nottingham and Sydney. We share the vision for the future ML technology: building trustworthy learning and reasoning algorithms, theories and systems.

Pinned Loading

  1. AR-Bench AR-Bench Public

    [ICML 2025] "From Passive to Active Reasoning: Can Large Language Models Ask the Right Questions under Incomplete Information?"

    Python 47 2

  2. DAD DAD Public

    Forked from JiachengZ01/DAD

    [ICML 2025] "One Stone, Two Birds: Enhancing Adversarial Defense Through the Lens of Distributional Discrepancy"

    Python 2

  3. G-effect G-effect Public

    Forked from QizhouWang/G-effect

    [ICLR 2025] "Rethinking LLM Unlearning Objectives: A Gradient Perspective and Go Beyond"

    Python 13 2

  4. AttrVR AttrVR Public

    Forked from caichengyi/AttrVR

    [ICLR 2025] "Attribute-based Visual Reprogramming for Vision-Language Models" Official Website: https://github.com/tmlr-group/AttrVR

    Python 3

  5. NoisyRationales NoisyRationales Public

    [NeurIPS 2024] "Can Language Models Perform Robust Reasoning in Chain-of-thought Prompting with Noisy Rationales?"

    Python 37 3

  6. BayesianLM BayesianLM Public

    Forked from caichengyi/BayesianLM

    [NeurIPS 2024 Oral] "Bayesian-Guided Label Mapping for Visual Reprogramming"

    Python 12 1

Repositories

Showing 10 of 102 repositories
  • ECON Public

    [ICML 2025] "From Debate to Equilibrium: Belief-Driven Multi-Agent LLM Reasoning via Bayesian Nash Equilibrium"

    tmlr-group/ECON’s past year of commit activity
    Python 29 MIT 3 1 1 Updated Nov 24, 2025
  • FedGPS Public Forked from visitworld123/FedGPS

    [NeurIPS 2025] "FedGPS: Statistical Rectification Against Data Heterogeneity in Federated Learning"

    tmlr-group/FedGPS’s past year of commit activity
    Python 0 MIT 2 0 0 Updated Nov 15, 2025
  • AlphaApollo Public

    [arXiv:2510.06261] "AlphaApollo: Orchestrating Foundation Models and Professional Tools into a Self-Evolving System for Deep Agentic Reasoning"

    tmlr-group/AlphaApollo’s past year of commit activity
    HTML 14 1 1 0 Updated Oct 27, 2025
  • AlignMI Public

    [NeurIPS 2025] "Generative Model Inversion Through the Lens of the Manifold Hypothesis"

    tmlr-group/AlignMI’s past year of commit activity
    Python 2 0 0 0 Updated Oct 22, 2025
  • ConV Public Forked from junz-debug/ConV

    [NeurIPS 2025 Spotlight] "Detecting Generated Images by Fitting Natural Image Distributions"

    tmlr-group/ConV’s past year of commit activity
    1 MIT 1 0 0 Updated Oct 22, 2025
  • WePe Public Forked from junz-debug/WePe

    [NeurIPS 2025] "Epistemic Uncertainty for Generated Image Detection"

    tmlr-group/WePe’s past year of commit activity
    0 MIT 1 0 0 Updated Oct 22, 2025
  • Easy2Hard Public Forked from Daftstone/Easy2Hard

    [NeurIPS 2025] "Advancing Machine-Generated Text Detection from an Easy to Hard Supervision Perspective"

    tmlr-group/Easy2Hard’s past year of commit activity
    Python 0 MIT 1 0 0 Updated Oct 15, 2025
  • AR-Bench Public

    [ICML 2025] "From Passive to Active Reasoning: Can Large Language Models Ask the Right Questions under Incomplete Information?"

    tmlr-group/AR-Bench’s past year of commit activity
    Python 47 2 1 0 Updated Oct 8, 2025
  • Co-rewarding Public Forked from resistzzz/Co-rewarding

    [arXiv:2508.00410] "Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models"

    tmlr-group/Co-rewarding’s past year of commit activity
    Python 44 4 0 0 Updated Oct 7, 2025
  • NSG-VD Public Forked from ZSHsh98/NSG-VD

    [NeurIPS 2025 Spotlight] "Physics-Driven Spatiotemporal Modeling for AI-Generated Video Detection"

    tmlr-group/NSG-VD’s past year of commit activity
    Python 0 Apache-2.0 2 0 0 Updated Oct 6, 2025

People

This organization has no public members. You must be a member to see who’s a part of this organization.