Lune

学者档案2020-2026

Deyi Xiong

Tianjin University

69论文数
56末位作者
91合作者人数
0一作
0独著

一作与末位作者按论文署名顺序统计,部分顶会采用姓氏字母顺序署名。

近期论文69

  1. From Curated Data to Scalable Models: Continual Pre-training of Dense and MoE Large Language Models for Tibetan第 17 位,共 17 位作者ACL 2026 · 5 次引用
  2. Finding the Translation Switch: Discovering and Exploiting the Task-Initiation Features in LLMs第 7 位,共 9 位作者AAAI 2026 · 2 次引用
  3. Incentivizing Parametric Knowledge via Reinforcement Learning with Verifiable Rewards for Cross-Cultural Entity Translation第 11 位,共 11 位作者ACL 2026 · 2 次引用
  4. Revisiting Entropy in Reinforcement Learning for Large Reasoning Models第 9 位,共 9 位作者ACL Findings 2026 · 2 次引用
  5. EvoSci: A Bio-Inspired Multi-Agent Framework for the Evolution of Scientific Discovery第 3 位,共 3 位作者ACL 2026 · 1 次引用
  6. Why Does Reinforcement Learning Generalize? A Feature-Level Mechanistic Study of Post-Training in Large Language Models第 6 位,共 6 位作者ACL 2026 · 1 次引用
  7. AdaDPI: Document-level Translation Adaptive Agent via Dynamic Parametric Internalization第 4 位,共 4 位作者ACL 2026
  8. APPSI-139: A Parallel Corpus of English Application Privacy Policy Summarization and Interpretation第 7 位,共 10 位作者ACL 2026
  9. Beyond Value Benchmarks: Measuring Value-Structure Alignment in Large Language Models via Symmetric Q-Sorts第 5 位,共 5 位作者ACL 2026
  10. DVMap: Fine-Grained Pluralistic Value Alignment via High-Consensus Demographic-Value Mapping第 5 位,共 5 位作者ACL 2026
  11. ERRV: Eliciting Efficient Reasoning through Reasoning Vectors for Policy Optimization in Large Language Models第 6 位,共 6 位作者ACL Findings 2026
  12. ExpertIVS: Sociological Expert Driven Individual Value Simulation in Large Language Models第 10 位,共 10 位作者ACL Findings 2026
  13. From Insight to Action: A Novel Framework for Interpretability-Guided Data Selection in Large Language Models第 9 位,共 10 位作者ACL 2026
  14. MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation第 2 位,共 2 位作者ACL Findings 2026
  15. Multilingual Safety Alignment via Representation-Space Separability第 3 位,共 3 位作者ICML 2026
  16. Neuronal Insights into LLM Attacks: Targeted Neuron Tuning for Precise and Robust Vulnerability Patching第 7 位,共 7 位作者ACL Findings 2026
  17. SARA: Unlocking Multilingual Knowledge in Mixture-of-Experts via Semantically Anchored Routing Alignment第 12 位,共 12 位作者ACL Findings 2026
  18. TENP: Trapezoidal Expert Neuron Pruning For Mixture-of-Experts第 3 位,共 3 位作者ACL Findings 2026
  19. ChatSOP: An SOP-Guided MCTS Planning Framework for Controllable LLM Dialogue Agents第 16 位,共 16 位作者ACL 2025 · 9 次引用
  20. CRiskEval: A Chinese Multi-Level Risk Evaluation Benchmark Dataset for Large Language Models第 2 位,共 2 位作者ACL 2025 · 5 次引用
  21. Debate4MATH: Multi-Agent Debate for Fine-Grained Reasoning in Math第 2 位,共 2 位作者ACL Findings 2025 · 5 次引用
  22. C²RBench: A Chinese Complex Reasoning Benchmark for Large Language Models第 4 位,共 4 位作者ACL Findings 2025 · 2 次引用
  23. Evaluating Multimodal Large Language Models on Video Captioning via Monte Carlo Tree Search第 9 位,共 9 位作者ACL 2025 · 2 次引用
  24. DCIS: Efficient Length Extrapolation of LLMs via Divide-and-Conquer Scaling Factor Search第 5 位,共 5 位作者EMNLP 2025
  25. DiplomacyAgent: Do LLMs Balance Interests and Ethical Principles in International Events?第 7 位,共 7 位作者EMNLP 2025
  26. HighMATH: Evaluating Math Reasoning of Large Language Models in Breadth and Depth第 10 位,共 10 位作者EMNLP Findings 2025
  27. MLAS-LoRA: Language-Aware Parameters Detection and LoRA-Based Knowledge Transfer for Multilingual Machine Translation第 5 位,共 5 位作者ACL 2025
  28. Praetor: A Fine-Grained Generative LLM Evaluator with Instance-Level Customizable Evaluation Criteria第 9 位,共 9 位作者ACL 2025
  29. Towards a Unified Paradigm of Concept Editing in Large Language Models第 5 位,共 5 位作者EMNLP 2025
  30. Watermarking Conditional Text Generation for AI Detection: Unveiling Challenges and a Semantic-Aware Watermark Remedy第 2 位,共 3 位作者AAAI 2024 · 66 次引用

展示最近收录的 30 篇论文。

黄昏的海面,两侧是细线勾勒的悬崖