Lune

Profiles2021-2026

Min Lin

Sea AI Lab

28Papers
16Last author
40Unique co-authors
1First author
1Solo

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

Recent papers28

  1. Reinforcing General Reasoning Without VerifiersAuthor 8 of 9ICLR 2026 · 75 citations
  2. Nonparametric Data Attribution for Diffusion ModelsAuthor 5 of 6ICML 2026
  3. Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned ConceptsAuthor 7 of 7ICCV 2025 · 4 citations
  4. A Closer Look at Machine Unlearning for Large Language ModelsAuthor 6 of 6ICLR 2025
  5. Bootstrapping Language Models with DPO Implicit RewardsAuthor 8 of 8ICLR 2025
  6. Cheating Automatic LLM Benchmarks: Null Models Achieve High Win RatesAuthor 6 of 6ICLR 2025
  7. Continual Reinforcement Learning by Planning with Online World ModelsAuthor 5 of 5ICML 2025
  8. Improved Techniques for Optimization-Based Jailbreaking on Large Language ModelsAuthor 8 of 8ICLR 2025
  9. PipeOffload: Improving Scalability of Pipeline Parallelism with Memory OptimizationAuthor 4 of 5ICML 2025
  10. Scaling up Masked Diffusion Models on TextAuthor 7 of 8ICLR 2025
  11. When Attention Sink Emerges in Language Models: An Empirical ViewAuthor 8 of 8ICLR 2025
  12. Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMsAuthor 6 of 6NeurIPS 2024 · 177 citations
  13. Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially FastAuthor 8 of 8ICML 2024 · 128 citations
  14. Finetuning Text-to-Image Diffusion Models for FairnessAuthor 4 of 6ICLR 2024 · 97 citations
  15. Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their DefensesAuthor 6 of 6NeurIPS 2024 · 96 citations
  16. Intriguing Properties of Data Attribution on Diffusion ModelsAuthor 5 of 5ICLR 2024 · 41 citations
  17. Stochastic Taylor Derivative Estimator: Efficient amortization for arbitrary differential operatorsAuthor 3 of 4NeurIPS 2024 · 32 citations
  18. Zero Bubble (Almost) Pipeline ParallelismAuthor 4 of 4ICLR 2024 · 31 citations
  19. Pipeline Parallelism with Controllable MemoryAuthor 4 of 4NeurIPS 2024 · 25 citations
  20. Graph Diffusion Policy OptimizationAuthor 5 of 6NeurIPS 2024 · 23 citations
  21. Automatic Functional Differentiation in JAXAuthor 1 of 1ICLR 2024 · 5 citations
  22. Beyond Memorization: The Challenge of Random Memory Access in Language ModelsAuthor 6 of 6ACL 2024
  23. On Evaluating Adversarial Robustness of Large Vision-Language ModelsAuthor 7 of 7NeurIPS 2023 · 404 citations
  24. Bag of Tricks for Training Data Extraction from Language ModelsAuthor 7 of 8ICML 2023 · 87 citations
  25. Nonparametric Generative Modeling with Conditional Sliced-Wasserstein FlowsAuthor 5 of 5ICML 2023 · 15 citations
  26. On Calibrating Diffusion Probabilistic ModelsAuthor 4 of 6NeurIPS 2023 · 7 citations
  27. Causal Attention for Interpretable and Generalizable Graph ClassificationAuthor 3 of 6KDD 2022 · 166 citations
  28. How Should Pre-Trained Language Models Be Fine-Tuned Towards Adversarial Robustness?Author 4 of 6NeurIPS 2021 · 80 citations
Dusk over the sea between two cliffs drawn in fine vertical lines