Lune

Profiles2023-2026

Dongfang Li

Harbin Institute of Technology (Shenzhen)

12Papers
0Last author
14Unique co-authors
4First author
0Solo

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

Recent papers12

  1. Is On-Policy Data always the Best Choice for Direct Preference Optimization-Based LM Alignment?Author 2 of 7ICLR 2026 · 1 citation
  2. Structured Episodic Event MemoryAuthor 2 of 6ACL 2026 · 1 citation
  3. Improving Value-based Process Verifier via Low-Cost Variance ReductionAuthor 2 of 4AAAI 2026
  4. LycheeCluster: Efficient Long-Context Inference with Structure-Aware Chunking and Hierarchical KV IndexingAuthor 1 of 5ACL Findings 2026
  5. SelectIT: Selective Instruction Tuning for LLMs via Uncertainty-Aware Self-ReflectionAuthor 4 of 7NeurIPS 2024 · 49 citations
  6. Separate the Wheat from the Chaff: Model Deficiency Unlearning via Parameter-Efficient Module OperationAuthor 2 of 6AAAI 2024 · 43 citations
  7. In-Context Learning State Vector with Inner and Momentum OptimizationAuthor 1 of 6NeurIPS 2024 · 19 citations
  8. Temporal Knowledge Question Answering via Abstract Reasoning InductionAuthor 2 of 5ACL 2024 · 9 citations
  9. SEER: Self-Aligned Evidence Extraction for Retrieval-Augmented GenerationAuthor 2 of 7EMNLP 2024 · 6 citations
  10. Improving Attributed Text Generation of Large Language Models via Preference LearningAuthor 1 of 7ACL Findings 2024 · 4 citations
  11. Take Off the Training Wheels! Progressive In-Context Learning for Effective AlignmentAuthor 2 of 7EMNLP 2024 · 1 citation
  12. ExplainCPE: A Free-text Explanation Benchmark of Chinese Pharmacist ExaminationAuthor 1 of 5EMNLP Findings 2023 · 10 citations
Dusk over the sea between two cliffs drawn in fine vertical lines