Lune

Profiles2024-2026

Xiaoyu Shen

Eastern Institute of Technology

29Papers
22Last author
23Unique co-authors
1First author
0Solo

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

Recent papers29

  1. HiDrop: Hierarchical Vision Token Reduction in MLLMs via Late Injection, Concave Pyramid Pruning, and Early ExitAuthor 6 of 6ICLR 2026 · 21 citations
  2. StreamingThinker: Large Language Models Can Think While ReadingAuthor 5 of 5ICLR 2026 · 17 citations
  3. Tools are under-documented: Simple Document Expansion Boosts Tool RetrievalAuthor 6 of 6ICLR 2026 · 16 citations
  4. Rethinking Reasoning in Document Ranking: Why Chain-of-Thought Falls ShortAuthor 6 of 6ICLR 2026 · 11 citations
  5. Think-as-You-See: Streaming Chain-of-Thought Reasoning for Large Vision-Language ModelsAuthor 7 of 7CVPR 2026 · 6 citations
  6. UTPTrack: Towards Simple and Unified Token Pruning for Visual TrackingAuthor 8 of 8CVPR 2026 · 6 citations
  7. What Do Visual Tokens Really Encode? Uncovering Sparsity and Redundancy in Multimodal Large Language ModelsAuthor 4 of 4CVPR 2026 · 6 citations
  8. Bridging the Stability-Expressivity Gap: Synthetic Data Scaling and Preference Alignment for Low-Resource Spoken Language ModelsAuthor 6 of 6ICML 2026 · 1 citation
  9. What Makes Effective Supervision in Latent Chain-of-Thought: An Information-Theoretic AnalysisAuthor 6 of 6ICML 2026 · 1 citation
  10. Beyond Global Similarity: Multi-Conditional Retrieval for Fine-Grained Cross-Modal UnderstandingAuthor 6 of 6CVPR 2026
  11. Bring Future Vision: Dynamic Computation Allocation Guided by Lightweight Feature ForecasterAuthor 6 of 6ICML 2026
  12. From Static Inference to Dynamic Interaction: A Survey of Streaming Large Language ModelsAuthor 7 of 7ACL Findings 2026
  13. ProactiveLLM: Learning Active Interaction for Streaming Large Language ModelsAuthor 6 of 6ICML 2026
  14. Revealing Scaling Paradox in Large-scale Time Series Models: Implications for More Efficient and Accurate ForecastingAuthor 5 of 5ICML 2026
  15. VisiPruner: Decoding Discontinuous Cross-Modal Dynamics for Efficient Multimodal LLMsAuthor 8 of 8EMNLP 2025 · 11 citations
  16. Unveiling the Key Factors for Distilling Chain-of-Thought ReasoningAuthor 11 of 11ACL Findings 2025 · 7 citations
  17. HAF-RM: A Hybrid Alignment Framework for Reward Model TrainingAuthor 2 of 8ACL 2025 · 4 citations
  18. Auto-SLURP: A Benchmark Dataset for Evaluating Multi-Agent Frameworks in Smart Personal AssistantAuthor 2 of 2EMNLP Findings 2025
  19. Beyond Content Relevance: Evaluating Instruction Following in Retrieval ModelsAuthor 8 of 8ICLR 2025
  20. MCM-DPO: Multifaceted Cross-Modal Direct Preference Optimization for Alt-text GenerationAuthor 5 of 7ACM MM 2025
  21. MultiConIR: Towards Multi-Condition Information RetrievalAuthor 9 of 9EMNLP Findings 2025
  22. Multimodal Language Models See Better When They Look ShallowerAuthor 9 of 10EMNLP 2025
  23. SkipGPT: Each Token is One of a KindAuthor 8 of 8ICML 2025
  24. StableMask: Refining Causal Masking in Decoder-only TransformerAuthor 6 of 7ICML 2024 · 23 citations
  25. Assessing "Implicit" Retrieval Robustness of Large Language ModelsAuthor 1 of 5EMNLP 2024 · 3 citations
  26. Fine-Tuning Large Language Models to Translate: Will a Touch of Noisy Data in Misaligned Languages Suffice?Author 5 of 6EMNLP 2024 · 3 citations
  27. The Impact of Demonstrations on Multilingual In-Context Learning: A Multidimensional AnalysisAuthor 5 of 7ACL Findings 2024 · 3 citations
  28. Unveiling In-Context Learning: A Coordinate System to Understand Its Working MechanismAuthor 4 of 4EMNLP 2024 · 2 citations
  29. To Preserve or To Compress: An In-Depth Study of Connector Selection in Multimodal Large Language ModelsAuthor 4 of 4EMNLP 2024 · 1 citation
Dusk over the sea between two cliffs drawn in fine vertical lines