Lune

Profiles2020-2026

Tong Zhang

University of Illinois Urbana-Champaign

41Papers
32Last author
64Unique co-authors
0First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers41

  1. Lean4Physics: Comprehensive Reasoning Framework for College-level Physics in Lean4Author 8 of 9ICLR 2026 · 7 citations
  2. GAR: Generative Adversarial Reinforcement Learning for Formal Theorem ProvingAuthor 5 of 5ICLR 2026 · 5 citations
  3. Sharp Analysis for KL-Regularized Contextual Bandits and RLHFAuthor 4 of 4NeurIPS 2025 · 33 citations
  4. Building Math Agents with Multi-Turn Iterative Preference LearningAuthor 12 of 13ICLR 2025 · 1 citation
  5. Bridge-Coder: Transferring Model Capabilities from High-Resource to Low-Resource Programming LanguageAuthor 8 of 8ACL Findings 2025
  6. EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied AgentsAuthor 13 of 13ICML 2025
  7. Let's Reason Formally: Natural-Formal Hybrid Reasoning Enhances LLM's Math CapabilityAuthor 3 of 3EMNLP 2025
  8. Logarithmic Regret for Online KL-Regularized Reinforcement LearningAuthor 5 of 5ICML 2025
  9. Personalized Visual Instruction TuningAuthor 6 of 6ICLR 2025
  10. ScaleBiO: Scalable Bilevel Optimization for LLM Data ReweightingAuthor 9 of 9ACL 2025
  11. Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-constraintAuthor 8 of 8ICML 2024 · 346 citations
  12. Regularizing Hidden States Enables Learning Generalizable Reward Model for LLMsAuthor 5 of 5NeurIPS 2024 · 157 citations
  13. Online Iterative Reinforcement Learning from Human Feedback with General Preference ModelAuthor 6 of 6NeurIPS 2024 · 60 citations
  14. Towards Robust Model-Based Reinforcement Learning Against Adversarial CorruptionAuthor 4 of 4ICML 2024 · 10 citations
  15. Plum: Prompt Learning using MetaheuristicsAuthor 9 of 9ACL Findings 2024 · 9 citations
  16. TheoremLlama: Transforming General-Purpose LLMs into Lean4 ExpertsAuthor 7 of 7EMNLP 2024 · 9 citations
  17. DetGPT: Detect What You Need via ReasoningAuthor 11 of 11EMNLP 2023 · 57 citations
  18. A Theoretical Analysis of Optimistic Proximal Policy Optimization in Linear Markov Decision ProcessesAuthor 2 of 2NeurIPS 2023 · 47 citations
  19. Corruption-Robust Algorithms with Uncertainty Weighting for Nonlinear Contextual Bandits and Markov Decision ProcessesAuthor 4 of 4ICML 2023 · 40 citations
  20. What is Essential for Unseen Goal Generalization of Offline Goal-conditioned RL?Author 6 of 6ICML 2023 · 35 citations
  21. Covariate-Shift Generalization via Random Sample WeightingAuthor 4 of 7AAAI 2023 · 7 citations
  22. Double Randomized Underdamped Langevin with Dimension-Independent Convergence GuaranteeAuthor 3 of 3NeurIPS 2023 · 2 citations
  23. Sparse Invariant Risk MinimizationAuthor 4 of 4ICML 2022 · 85 citations
  24. Model Agnostic Sample Reweighting for Out-of-Distribution LearningAuthor 7 of 7ICML 2022 · 73 citations
  25. Pessimistic Minimax Value Iteration: Provably Efficient Equilibrium Learning from Offline DatasetsAuthor 5 of 7ICML 2022 · 46 citations
  26. Probabilistic Bilevel Coreset SelectionAuthor 5 of 5ICML 2022 · 39 citations
  27. A Theoretical Analysis on Independence-driven Importance Weighting for Covariate-shift GeneralizationAuthor 4 of 5ICML 2022 · 36 citations
  28. A Self-Play Posterior Sampling Algorithm for Zero-Sum Markov GamesAuthor 5 of 5ICML 2022 · 24 citations
  29. Benefits of Overparameterized Convolutional Residual Networks: Function Approximation under Smoothness ConstraintAuthor 5 of 6ICML 2022 · 16 citations
  30. Efficient Neural Network Training via Forward and Backward Propagation SparsificationAuthor 5 of 5NeurIPS 2021 · 57 citations

Showing the 30 most recent indexed papers.

Dusk over the sea between two cliffs drawn in fine vertical lines