PPaperPicks

Yu Cheng

Chinese University of Hong Kong, Department of Computer Science and Engineering, Shatin, Hong Kong

54 papers at tracked venues · 44 at CORE A* · active 2024–2026

DBLP profile ↗ORCID 0000-0002-2315-5641 ↗Google Scholar ↗Homepage ↗

Venues

ACL ×14ICML ×9NeurIPS ×8EMNLP ×6AAAI ×3ICCV ×3ICLR ×3CVPR ×2ECCV ×2NAACL ×2ACM MM ×1SIGKDD ×1

Frequent coauthors

Chenghao Fan
DBLP profile ↗ORCID search ↗
×3
Yucheng Zhou
DBLP profile ↗ORCID search ↗
×2
Yixiao Zhou
DBLP profile ↗ORCID search ↗
×2
Yafu Li
DBLP profile ↗ORCID search ↗
×2
Jihai Zhang
DBLP profile ↗ORCID search ↗
×2
Tong Zhu
DBLP profile ↗ORCID search ↗
×2
Mingyang Song
DBLP profile ↗ORCID search ↗
×2
Jiaqi Liao
DBLP profile ↗ORCID search ↗
×2
Zhilin Wang
DBLP profile ↗ORCID search ↗
×2
Guanjie Chen
DBLP profile ↗ORCID search ↗
×2
Zhaochen Su
DBLP profile ↗ORCID search ↗
×2
Zhenyi Lu
DBLP profile ↗ORCID search ↗
×2

Papers

  1. CoTEvol: Self-Evolving Chain-of-Thoughts for Data Synthesis in Mathematical Reasoning
    ACL 2026 ·
    Zhuo Wang
    DBLP profile ↗ORCID search ↗
  2. Each Rank Could be an Expert: Single-Ranked Mixture of Experts LoRA for Multi-task Learning
    SIGKDD 2026 ·
    Ziyu Zhao
    DBLP profile ↗ORCID search ↗
  3. Less Is More: Vision Representation Compression for Efficient Video Generation with Large Language Models
    AAAI 2026 ·
    Yucheng Zhou
    DBLP profile ↗ORCID search ↗
  4. Multi-LLM Collaborative Search for Complex Problem Solving
    ACL 2026 ·
    Sen Yang
    DBLP profile ↗ORCID search ↗
  5. Native Hybrid Attention for Efficient Sequence Modeling
    ACL 2026 ·
    Jusen Du
    DBLP profile ↗ORCID search ↗
  6. Nirvana: A Specialized Generalist Model With Task-Aware Memory Mechanism
    ACL 2026 ·
    Yuhua Jiang
    DBLP profile ↗ORCID search ↗
  7. One Refiner to Unlock Them All: Inference-Time Reasoning Elicitation via Reinforcement Query Refinement
    ACL 2026 ·
    Yixiao Zhou
    DBLP profile ↗ORCID search ↗
  8. Scaling Reasoning, Losing Control: Evaluating Instruction Following in Large Reasoning Models
    ACL 2026 ·
    Tingchen Fu
    DBLP profile ↗ORCID search ↗
  9. The Best of Both Worlds: Combining Parallel and Sequential Inference Scaling via Aggregation Fine-Tuning
    ACL 2026 ·
    Yafu Li
    DBLP profile ↗ORCID search ↗
  10. Bit-Flip Error Resilience in LLMs: A Comprehensive Analysis and Defense Framework
    EMNLP 2025 ·
    Yuhang Chen
    DBLP profile ↗ORCID search ↗
  11. CLIP-MoE: Towards Building Mixture of Experts for CLIP with Diversified Multiplet Upcycling
    EMNLP 2025 ·
    Jihai Zhang
    DBLP profile ↗ORCID search ↗
  12. Can MLLMs Reason in Multimodality? EMMA: An Enhanced MultiModal ReAsoning Benchmark
    ICML 2025 ·
    Yunzhuo Hao
    DBLP profile ↗ORCID search ↗
  13. Cooperative or Competitive? Understanding the Interaction between Attention Heads From A Game Theory Perspective
    ACL 2025 ·
    Xiaoye Qu
    DBLP profile ↗ORCID search ↗
  14. Divide and Conquer: Grounding LLMs as Efficient Decision-Making Agents via Offline Hierarchical Reinforcement Learning
    ICML 2025 ·
    Zican Hu
    DBLP profile ↗ORCID search ↗
  15. Diving into Self-Evolving Training for Multimodal Reasoning
    ICML 2025 ·
    Wei Liu
    DBLP profile ↗ORCID search ↗
  16. Dropping Experts, Recombining Neurons: Retraining-Free Pruning for Sparse Mixture-of-Experts LLMs
    EMNLP 2025 ·
    Yixiao Zhou
    DBLP profile ↗ORCID search ↗
  17. Dynamic Data Mixing Maximizes Instruction Tuning for Mixture-of-Experts
    NAACL 2025 ·
    Tong Zhu
    DBLP profile ↗ORCID search ↗
  18. Extrapolating and Decoupling Image-to-Video Generation Models: Motion Modeling is Easier Than You Think
    CVPR 2025 ·
    Jie Tian
    DBLP profile ↗ORCID search ↗
  19. From Head to Tail: Towards Balanced Representation in Large Vision-Language Models through Adaptive Data Calibration
    CVPR 2025 ·
    Mingyang Song
    DBLP profile ↗ORCID search ↗
  20. ImageGen-CoT: Enhancing Text-to-Image in-context Learning with Chain-of-Thought Reasoning
    ICCV 2025 ·
    Jiaqi Liao
    DBLP profile ↗ORCID search ↗
  21. LangBridge: Interpreting Image as a Combination of Language Embeddings
    ICCV 2025 ·
    Jiaqi Liao
    DBLP profile ↗ORCID search ↗
  22. Learning to Reason under Off-Policy Guidance
    NeurIPS 2025 ·
    Jianhao Yan
    DBLP profile ↗ORCID search ↗
  23. Liger: Linearizing Large Language Models to Gated Recurrent Structures
    ICML 2025 ·
    Disen Lan
    DBLP profile ↗ORCID search ↗
  24. Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment
    ICML 2025 ·
    Chenghao Fan
    DBLP profile ↗ORCID search ↗
  25. Occult: Optimizing Collaborative Communications across Experts for Accelerated Parallel MoE Training and Inference
    ICML 2025 ·
    Shuqing Luo
    DBLP profile ↗ORCID search ↗
  26. PRMBench: A Fine-grained and Challenging Benchmark for Process-Level Reward Models
    ACL 2025 ·
    Mingyang Song
    DBLP profile ↗ORCID search ↗
  27. SEE: Continual Fine-tuning with Sequential Ensemble of Experts
    ACL 2025 ·
    Zhilin Wang
    DBLP profile ↗ORCID search ↗
  28. Scaling Physical Reasoning with the PHYSICS Dataset
    NeurIPS 2025 ·
    Shenghe Zheng
    DBLP profile ↗ORCID search ↗
  29. Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedback
    ICML 2025 ·
    Yafu Li
    DBLP profile ↗ORCID search ↗
  30. Text-to-Decision Agent: Offline Meta-Reinforcement Learning from Natural Language Supervision
    NeurIPS 2025 ·
    Shilin Zhang
    DBLP profile ↗ORCID search ↗
  31. Towards Stabilized and Efficient Diffusion Transformers Through Long-Skip-Connections With Spectral Constraints
    ICCV 2025 ·
    Guanjie Chen
    DBLP profile ↗ORCID search ↗
  32. Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation
    ICML 2025 ·
    Fanqing Meng
    DBLP profile ↗ORCID search ↗
  33. Unveiling Attractor Cycles in Large Language Models: A Dynamical Systems View of Successive Paraphrasing
    ACL 2025 ·
    Zhilin Wang
    DBLP profile ↗ORCID search ↗
  34. VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models
    NeurIPS 2025 ·
    Xiangdong Zhang
    DBLP profile ↗ORCID search ↗
  35. Weak to Strong Generalization for Large Language Models with Multi-capabilities
    ICLR 2025 ·
    Yucheng Zhou
    DBLP profile ↗ORCID search ↗
  36. Confidence is not Timeless: Modeling Temporal Validity for Rule-based Temporal Knowledge Graph Forecasting
    ACL 2024 ·
    Rikui Huang
    DBLP profile ↗ORCID search ↗
  37. ConflictBank: A Benchmark for Evaluating the Influence of Knowledge Conflicts in LLMs
    NeurIPS 2024 ·
    Zhaochen Su
    DBLP profile ↗ORCID search ↗
  38. Enhancing Low-Resource Relation Representations through Multi-View Decoupling
    AAAI 2024 ·
    Chenghao Fan
    DBLP profile ↗ORCID search ↗
  39. LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-Training
    EMNLP 2024 ·
    Tong Zhu
    DBLP profile ↗ORCID search ↗
  40. Learning the Unlearned: Mitigating Feature Suppression in Contrastive Learning
    ECCV 2024 ·
    Jihai Zhang
    DBLP profile ↗ORCID search ↗
  41. Living in the Moment: Can Large Language Models Grasp Co-Temporal Reasoning?
    ACL 2024 ·
    Zhaochen Su
    DBLP profile ↗ORCID search ↗
  42. MAGIS: LLM-Based Multi-Agent Framework for GitHub Issue Resolution
    NeurIPS 2024 ·
    Wei Tao
    DBLP profile ↗ORCID search ↗
  43. Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy
    ICLR 2024 ·
    Pingzhi Li
    DBLP profile ↗ORCID search ↗
  44. Mitigating Boundary Ambiguity and Inherent Bias for Text Classification in the Era of Large Language Models
    ACL 2024 ·
    Zhenyi Lu
    DBLP profile ↗ORCID search ↗
  45. MoE-RBench: Towards Building Reliable Language Models with Sparse Mixture-of-Experts
    ICML 2024 ·
    Guanjie Chen
    DBLP profile ↗ORCID search ↗
  46. Not All Inputs Are Valid: Towards Open-Set Video Moment Retrieval using Language
    ACM MM 2024 ·
    Xiang Fang
    DBLP profile ↗ORCID search ↗
  47. On Giant's Shoulders: Effortless Weak to Strong by Dynamic Logits Fusion
    NeurIPS 2024 ·
    Chenghao Fan
    DBLP profile ↗ORCID search ↗
  48. On the Universal Truthfulness Hyperplane Inside LLMs
    EMNLP 2024 ·
    Junteng Liu
    DBLP profile ↗ORCID search ↗
  49. Reinforcement Learning with Token-level Feedback for Controllable Text Generation
    NAACL 2024 ·
    Wendi Li
    DBLP profile ↗ORCID search ↗
  50. Rethinking Weakly-Supervised Video Temporal Grounding From a Game Perspective
    ECCV 2024 ·
    Xiang Fang
    DBLP profile ↗ORCID search ↗
  51. SURf: Teaching Large Vision-Language Models to Selectively Utilize Retrieved Information
    EMNLP 2024 ·
    Jiashuo Sun
    DBLP profile ↗ORCID search ↗
  52. Sparse MoE with Language Guided Routing for Multilingual Machine Translation
    ICLR 2024 ·
    Xinyu Zhao
    DBLP profile ↗ORCID search ↗
  53. Twin-Merging: Dynamic Integration of Modular Expertise in Model Merging
    NeurIPS 2024 ·
    Zhenyi Lu
    DBLP profile ↗ORCID search ↗
  54. Unsupervised Domain Adaptative Temporal Sentence Localization with Mutual Information Maximization
    AAAI 2024 ·
    Daizong Liu
    DBLP profile ↗ORCID search ↗