Lune

Profiles2023-2026

Haohan Wang

University of Illinois Urbana -Champaign

18Papers
10Last author
38Unique co-authors
0First author
0Solo

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

Recent papers18

  1. Optimal Aggregation of LLM and PRM Signals for Efficient Test-Time ScalingAuthor 6 of 6ICLR 2026 · 5 citations
  2. Learning to Conceal Risk: Controllable Multi-turn Red Teaming for LLMs in the Financial DomainAuthor 4 of 5ACL 2026 · 3 citations
  3. Agent Primitives: Reuseable Latent Building Blocks for Multi-Agent SystemsAuthor 5 of 5ICML 2026
  4. Query-Efficient Domain Knowledge Stealing Against Large Language ModelsAuthor 5 of 8AAAI 2026
  5. Security–Fidelity Tradeoffs: No Universal Defense Against Prompt InjectionAuthor 5 of 5ICML 2026
  6. Towards Adversarially Robust Dataset Distillation by Curvature RegularizationAuthor 6 of 7AAAI 2025 · 25 citations
  7. Exploring the Vulnerability of the Content Moderation Guardrail in Large Language Models via Intent ManipulationAuthor 7 of 7EMNLP Findings 2025 · 5 citations
  8. Evaluating the Inductive Abilities of Large Language Models: Why Chain-of-Thought Reasoning Sometimes Hurts More Than HelpsAuthor 4 of 4NeurIPS 2025 · 1 citation
  9. Examining Alignment of Large Language Models through Representative Heuristics: the case of political stereotypesAuthor 3 of 4ICLR 2025
  10. Improving Noise Efficiency in Privacy-Preserving Dataset DistillationAuthor 4 of 5ICCV 2025
  11. Revolve: Optimizing AI Systems by Tracking Response Evolution in Textual OptimizationAuthor 8 of 8ICML 2025
  12. Robust Prompt Optimization for Defending Language Models Against Jailbreaking AttacksAuthor 3 of 3NeurIPS 2024 · 198 citations
  13. Jailbreaking Large Language Models Against Moderation Guardrails via Cipher CharactersAuthor 4 of 4NeurIPS 2024 · 55 citations
  14. Foundation Model-oriented Robustness: Robust Image Model Evaluation with Pretrained ModelsAuthor 6 of 6ICLR 2024 · 8 citations
  15. A Sentence Speaks a Thousand Images: Domain Generalization through Distilling CLIP with Language GuidanceAuthor 5 of 6ICCV 2023 · 56 citations
  16. Adaptive Test-Time Personalization for Federated LearningAuthor 3 of 4NeurIPS 2023 · 41 citations
  17. Distilling Out-of-Distribution Robustness from Vision-Language Foundation ModelsAuthor 4 of 4NeurIPS 2023 · 14 citations
  18. Toward Robust Diagnosis: A Contour Attention Preserving Adversarial Defense for COVID-19 DetectionAuthor 5 of 7AAAI 2023 · 8 citations
Dusk over the sea between two cliffs drawn in fine vertical lines