Lune

Profiles2021-2026

Yiwen Guo

The University of Hong

14Papers
1Last author
15Unique co-authors
0First author
0Solo

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

Recent papers14

  1. Breaking Contextual Inertia: Reinforcement Learning with Single-Turn Anchors for Stable Multi-Turn InteractionAuthor 3 of 4ACL Findings 2026
  2. DualOptim+: Bridging Shared and Decoupled Optimizer States for Better Machine Unlearning in Large Language ModelsAuthor 3 of 4ICML 2026
  3. C3: A Bilingual Benchmark for Spoken Dialogue Models Exploring Challenges in Complex ConversationsAuthor 3 of 6EMNLP 2025 · 7 citations
  4. Cultivating Gaming Sense for Yourself: Making VLMs Gaming ExpertsAuthor 4 of 5ACL 2025 · 2 citations
  5. Reasoning-as-Logic-Units: Scaling Test-Time Reasoning in Large Language Models Through Logic Unit AlignmentAuthor 3 of 3ICML 2025
  6. Improved Generation of Adversarial Examples Against Safety-aligned LLMsAuthor 2 of 4NeurIPS 2024 · 23 citations
  7. Improving Adversarial Transferability via Intermediate-level Perturbation DecayAuthor 2 of 4NeurIPS 2023 · 43 citations
  8. MHCN: A Hyperbolic Neural Network Model for Multi-view Hierarchical ClusteringAuthor 3 of 6ICCV 2023 · 17 citations
  9. Towards Evaluating Transfer-based Attacks Systematically, Practically, and FairlyAuthor 2 of 4NeurIPS 2023 · 11 citations
  10. Black-Box Tuning of Vision-Language Models with Effective Gradient ApproximationAuthor 6 of 7EMNLP Findings 2023 · 5 citations
  11. Making Substitute Models More Bayesian Can Enhance Transferability of Adversarial ExamplesAuthor 2 of 4ICLR 2023 · 5 citations
  12. Squeeze Training for Adversarial RobustnessAuthor 2 of 4ICLR 2023 · 2 citations
  13. Texts as Images in Prompt Tuning for Multi-Label Image RecognitionAuthor 5 of 6CVPR 2023
  14. Robust and Fully-Dynamic Coreset for Continuous-and-Bounded Learning (With Outliers) ProblemsAuthor 2 of 3NeurIPS 2021 · 10 citations
Dusk over the sea between two cliffs drawn in fine vertical lines