Lune

Profiles2020-2026

Yang Liu

University of California, Santa Cruz

61Papers
37Last author
80Unique co-authors
4First author
1Solo

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

Recent papers61

  1. DRAGON: Guard LLM Unlearning in Context via Negative Detection and ReasoningAuthor 7 of 7ICLR 2026 · 10 citations
  2. Label Smoothing Improves Machine UnlearningAuthor 9 of 9ICLR 2026 · 8 citations
  3. Stabilizing Self-Consuming Diffusion Models with Latent Space FilteringAuthor 3 of 4AAAI 2026 · 2 citations
  4. Human and AI Perceptual Differences in Image Classification ErrorsAuthor 3 of 4AAAI 2025 · 11 citations
  5. Evaluating LLM-contaminated Crowdsourcing Data Without Ground TruthAuthor 4 of 4NeurIPS 2025 · 3 citations
  6. ACC-Collab: An Actor-Critic Approach to Multi-Agent LLM CollaborationAuthor 4 of 4ICLR 2025 · 1 citation
  7. Adversarial Machine UnlearningAuthor 4 of 4ICLR 2025 · 1 citation
  8. DiffTell: A High-Quality Dataset for Describing Image Manipulation ChangesAuthor 7 of 7ICCV 2025 · 1 citation
  9. Improving Data Efficiency via Curating LLM-Driven Rating SystemsAuthor 7 of 9ICLR 2025
  10. Noise-Resilient Point-wise Anomaly Detection in Time Series Using Weak Segment LabelsAuthor 9 of 9KDD 2025
  11. Token Cleaning: Fine-Grained Data Selection for LLM Supervised Fine-TuningAuthor 7 of 7ICML 2025
  12. Toward Optimal LLM Alignments Using Two-Player GamesAuthor 13 of 13EMNLP Findings 2025
  13. Large Language Model Unlearning via Embedding-Corrupted PromptsAuthor 4 of 4NeurIPS 2024 · 138 citations
  14. Multi-LLM Debate: Framework, Principals, and InterventionsAuthor 2 of 2NeurIPS 2024 · 131 citations
  15. FedFixer: Mitigating Heterogeneous Label Noise in Federated LearningAuthor 7 of 7AAAI 2024 · 30 citations
  16. Unmasking and Improving Data Credibility: A Study with Datasets for Training Harmless Language ModelsAuthor 4 of 5ICLR 2024 · 30 citations
  17. RobustTSF: Towards Theory and Design of Robust Time Series Forecasting with AnomaliesAuthor 3 of 4ICLR 2024 · 19 citations
  18. Performative Federated Learning: A Solution to Model-Dependent and Heterogeneous Distribution ShiftsAuthor 6 of 7AAAI 2024 · 15 citations
  19. Fairness without Harm: An Influence-Guided Active Sampling ApproachAuthor 8 of 9NeurIPS 2024 · 13 citations
  20. Performative Prediction with Bandit Feedback: Learning through ReparameterizationAuthor 4 of 4ICML 2024 · 13 citations
  21. Mitigating Reward Overoptimization via Lightweight Uncertainty EstimationAuthor 5 of 5NeurIPS 2024 · 11 citations
  22. User-Creator Feature Polarization in Recommender Systems with Dual InfluenceAuthor 7 of 7NeurIPS 2024 · 6 citations
  23. Achievable Fairness on Your Data With Utility GuaranteesAuthor 3 of 4NeurIPS 2024 · 3 citations
  24. Procedural Fairness Through Decoupling Objectionable Data Generating ComponentsAuthor 3 of 5ICLR 2024 · 3 citations
  25. Fair Participation via Sequential PoliciesAuthor 4 of 4AAAI 2024 · 1 citation
  26. Identifiability of Label Noise Transition MatrixAuthor 1 of 3ICML 2023 · 58 citations
  27. Long-Term Fairness with Unknown DynamicsAuthor 4 of 4NeurIPS 2023 · 33 citations
  28. T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image GenerationAuthor 4 of 5ACL Findings 2023 · 26 citations
  29. To Aggregate or Not? Learning with Separate Noisy LabelsAuthor 7 of 7KDD 2023 · 21 citations
  30. Model Transferability with Responsive Decision SubjectsAuthor 4 of 4ICML 2023 · 11 citations

Showing the 30 most recent indexed papers.

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