Selected papers and ongoing work.

2026

1. 

   Preprint arXiv Code

   Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics

   Maoqi Liu, Junwei He, Bowen Zhang, and 5 more authors

   arXiv preprint arXiv:2609.38847, 2026

2. 

   EMNLP 2026 arXiv Code

   CoDe-LoRA: Mitigating the Orthogonality Dilemma in Continual Learning of LLMs via Knowledge Consolidation and Decoupling

   Maoqi Liu, Quan Fang, and Yufei He

   Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, 2026

3. 

   AAAI 2026 arXiv Code

   NOTAM-Evolve: A Knowledge-Guided Self-Evolving Optimization Framework with LLMs for NOTAM Interpretation

   Maoqi Liu, Quan Fang, Yuhao Wu, and 3 more authors

   Proceedings of the AAAI Conference on Artificial Intelligence, 2026

2025

1. 

   ADVEI 2025 arXiv Code

   Knots: A Large-Scale Multi-Agent Enhanced Expert-Annotated Dataset and LLM Prompt Optimization for NOTAM Semantic Parsing

   Maoqi Liu, Quan Fang, Yang Yang, and 2 more authors

   Advanced Engineering Informatics, 2025

2. 

   PRCV 2025

   R²A²-MoE: Ridge Regression-based Analytic Adaptation with Mixture of Experts for Continual Learning with Vision-Language Models

   Maoqi Liu, Quan Fang, Junyu Gao, and 1 more author

   Pattern Recognition and Computer Vision, 2025