Publications

This page lists my full publication record. The homepage highlights a smaller set of representative works by research direction.

Selected Preprints

• [P4] NeoHorse Team et al., “NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness”, technical report, 2026. (paper) (code and models)

• [P3] Zehua Pei, Hui-Ling Zhen, Shixiong Kai, Sinno Jialin Pan, Yunhe Wang, Mingxuan Yuan, Bei Yu, “SCOPE: Prompt Evolution for Enhancing Agent Effectiveness”, arXiv preprint, 2025. (paper) (code)

• [P2] Zehua Pei, Hui-Ling Zhen, Xianzhi Yu, Sinno Jialin Pan, Mingxuan Yuan, Bei Yu, “FocuSFT: Bilevel Optimization for Dilution-Aware Long-Context Fine-Tuning”, arXiv preprint, 2026. (paper) (code)

• [P1] Zehua Pei, Hui-Ling Zhen, Xianzhi Yu, Sinno Jialin Pan, Mingxuan Yuan, Bei Yu, “From Pruning to Grafting: Dynamic Knowledge Redistribution via Learnable Layer Fusion”, arXiv preprint, 2024. (paper) (code)

Conference Papers

• [C12] Zehua Pei, Hui-Ling Zhen, Weizhe Lin, Sinno Jialin Pan, Yunhe Wang, Mingxuan Yuan, Bei Yu, “MemDLM: Memory-Enhanced DLM Training”, Annual Conference on Neural Information Processing Systems (NeurIPS), Sydney/Atlanta/Paris, Dec. 6–13, 2026. (paper) (code)

• [C11] Zehua Pei, Ying Zhang, Hui-Ling Zhen, Tao Yuan, Xianzhi Yu, Zhenhua Dong, Sinno Jialin Pan, Mingxuan Yuan, Bei Yu, “PreMoE: Proactive Inference for Efficient Mixture-of-Experts”, Conference on Language Modeling (COLM), San Francisco, Oct. 6–9, 2026. (paper) (code)

• [C10] Zehua Pei, Hui-Ling Zhen, Lancheng Zou, Xianzhi Yu, Wulong Liu, Sinno Jialin Pan, Mingxuan Yuan, Bei Yu, “Analytical FFN-to-MoE Restructuring via Activation Pattern Analysis”, Association for Computational Linguistics (ACL Main), San Diego, July 2-7, 2026. (paper) (code)

• [C9] Yu Zhang, Hui-Ling Zhen, Zehua Pei, Yingzhao Lian, Lihao Yin, Mingxuan Yuan, Bei Yu, “Sola: Solver-Layer Adaptation of LLM for Better Logic Reasoning”, SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Jeju, Korea, Aug. 9-13, 2026.

• [C8] Zehua Pei, Hui-Ling Zhen, Ying Zhang, Zhiyuan Yang, Xing Li, Xianzhi Yu, Mingxuan Yuan, Bei Yu, “Behavioral Fingerprinting of Large Language Models”, 1st Open Conference of AI Agents for Science (Agents4Science), Oct. 22, 2025. (paper) (code)

• [C7] Lancheng Zou, Shuo Yin, Zehua Pei, Tsung-Yi Ho, Farzan Farnia, Bei Yu, “PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models”, Neural Information Processing Systems (NeurIPS), San Diego, Dec. 02-07, 2025. (paper)

• [C6] Zehua Pei, Hui-Ling Zhen, Mingxuan Yuan, Yu Huang, Bei Yu, “BetterV: Controlled Verilog Generation with Discriminative Guidance”, International Conference on Machine Learning (ICML), Vienna, Austria, Jul. 21-27, 2024. (paper)

• [C5] Fangzhou Liu, Zehua Pei, Ziyang Yu, Haisheng Zheng, Zhuolun He, Tinghuan Chen, Bei Yu, “CBTune: Contextual Bandit Tuning for Logic Synthesis”, IEEE/ACM Proceedings Design, Automation and Test in Europe (DATE), Valencia, Spain, Mar. 25-27, 2024.

• [C4] Haisheng Zheng, Zhuolun He, Fangzhou Liu, Zehua Pei, Bei Yu, “LSTP: A Logic Synthesis Timing Predictor”, IEEE/ACM Asian and South Pacific Design Automation Conference (ASPDAC), South Korea, Jan. 22-25, 2024. (paper)

• [C3] Zehua Pei, Fangzhou Liu, Zhuolun He, Guojin Chen, Haisheng Zheng, Keren Zhu, Bei Yu, “AlphaSyn: Logic Synthesis Optimization with Efficient Monte Carlo Tree Search”, IEEE/ACM International Conference on Computer-Aided Design (ICCAD), San Francisco, Oct. 29-Nov. 02, 2023. (paper)

• [C2] Guojin Chen, Zehua Pei, Haoyu Yang, Yuzhe Ma, Bei Yu, Martin Wong, “Physics-Informed Optical Kernel Regression Using Complex-valued Neural Fields”, ACM/IEEE Design Automation Conference (DAC), San Francisco, Jul. 09-13, 2023. (paper)

• [C1] Zehua Pei, Wenqian Zhao, Zhuolun He, Bei Yu, “Bit-Level Quantization for Efficient Layout Hotspot Detection”, International Symposium of EDA (ISEDA), May 9-11, 2023. (paper)

Journal Papers

• [J3] Fangzhou Liu, Wuqian Tang, Zehua Pei, Ziyang Yu, Haisheng Zheng, Zhuolun He, Mengjia Dai, Tinghuan Chen, Bei Yu, “CB-EVO: Contextual Bandit Tuning with Evolutionary Search for Logic Synthesis”, ACM Transactions on Design Automation of Electronic Systems (TODAES).

• [J2] Juncheng Li, Zehua Pei, Wenjie Li, Guangwei Gao, Longguang Wang, Yingqian Wang, Tieyong Zeng, “A Systematic Survey of Deep Learning-based Single-Image Super-Resolution”, ACM Computing Surveys (ACM COMPUT SURV).

• [J1] Zehua Pei, Xufeng Yao, Wenqian Zhao, Bei Yu, “Quantization via Distillation and Contrastive Learning”, IEEE Transactions on Neural Networks and Learning Systems (TNNLS).