Publications

(* means equal contribution)

Topic 1: LLM Reasoning & RL & Agent Memory

• Making Slow Thinking Faster: Compressing LLM Chain-of-Thought via Step Entropy. |PDF|Code|
  Zeju Li, Jianyuan Zhong, Ziyang Zheng, Xiangyu Wen, Zhijian Xu, Yingying Cheng, Fan Zhang, Qiang Xu. ICLR 2026
• Latent Context Compilation: Distilling Long Context into Compact Portable Memory. |PDF|Code|
  Zeju Li, Yizhou Zhou, Qiang Xu. COLM 2026
• HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning. |PDF|Code|
  Zeju Li, Ziyang Zheng, Yizhou Zhou, Qiang Xu. preprint, 2026
• Solve-Detect-Verify: Inference-Time Scaling with Flexible Generative Verifier. |PDF|Code|
  Jianyuan Zhong*, Zeju Li*, Zhijian Xu, Xiangyu Wen, Kezhi Li, Qiang Xu. ACL Main, 2026
• Dyve: Thinking Fast and Slow for Dynamic Process Verification. |PDF|Code|
  Jianyuan Zhong*, Zeju Li*, Zhijian Xu, Xiangyu Wen, Qiang Xu. EMNLP Main, 2025
• Context Distillation as Latent Memory Management. |PDF|Code|
  Ziyang Zheng, Ziyang_Zheng, Zeju Li, Xiangyu Wen, Jianyuan Zhong, Junhua Huang, Lei Chen, Mingxuan Yuan, Qiang Xu. preprint, 2026
• ThinkAnswer Loss: Balancing Semantic Similarity and Exact Matching for LLM Reasoning Enhancement.
  Shan Yang* , Kun Wu*† , Zeju Li*†, Linlin Zhang*, Xiangyu Pei, Leike An and Yu Liu. EMNLP Findings, 2025
• Reasoning Scaffolding: Distilling the Flow of Thought from LLMs.
  Xiangyu Wen, Junhua Huang, Zeju Li, Min Li, Jianyuan Zhong, Zhijian Xu, Mingxuan Yuan, Yongxiang Huang, Qiang Xu. ICLR, 2026
• Dependency Matters: Enhancing LLM Reasoning with Explicit Knowledge Grounding.
  Xiangyu Wen, Min Li, Junhua Huang, Jianyuan Zhong, Zhijian Xu, Zeju Li, Yongxiang Huang, Mingxuan Yuan, Qiang Xu. NeurIPS, 2025

Topic 2: LLM for Hardware Code Understanding and Generation

• DeepCircuitX: Repository-Level RTL Dataset for Code Understanding, Generation, and Multimodal Analysis. |PDF||Project Page|
  Zeju Li*, Changran Xu*, Zhengyuan Shi*, Zedong Peng, Yi Liu, Yunhao Zhou, Lingfeng Zhou, Chengyu Ma, Jianyuan Zhong, Xi Wang, Jieru Zhao, Zhufei Chu, Xiaoyan Yang, Qiang Xu. ICLAD, 2025
• ForgeEDA: A Comprehensive Multimodal Dataset for Advancing EDA. |Project|
  Zhengyuan Shi, Zeju Li, Chengyu Ma, Yunhao Zhou, Ziyang Zheng, Jiawei Liu, Hongyang Pan, Lingfeng Zhou, Kezhi Li, Jiaying Zhu, Lingwei Yan, Zhiqiang He, Chenhao Xue, Wentao Jiang, Fan Yang, Guangyu Sun, Xiaoyan Yang, Gang Chen, Chuan Shi, Zhufei Chu, Jun Yang, Qiang Xu. ISEDA, 2025
• DeepRTL: Bridging Verilog Understanding and Generation with a Unified Representation Model. |[PDF]|
  Yi Liu, Changran XU, Yunhao Zhou, Zeju Li, Qiang Xu. ICLR 2025, Spotlight
• ForgeHLS: A Large-Scale, Open-Source Dataset for High-Level Synthesis. |[PDF]|
  Zedong Peng, Zeju Li, Mingzhe Gao, Qiang Xu, Chen Zhang, Jieru Zhao. Preprint, 2025

Topic 3: MLLM for 3D World Understanding

• 3DMIT: 3D Multi-modal Instruction Tuning for Scene Understanding. |[PDF] [Code]|
  Zeju Li, Chao Zhang, Xiaoyan Wang, Ruilong Ren, Yifan Xu, Ruifei Ma, Xiangde Liu. ICME-3DMM 2024, Oral
• Spatial 3D-LLM : Exploring Spatial Awareness in 3D Vision-Language Models. |[PDF]|
  Xiaoyan Wang, Zeju Li, Yifan Xu, Jiaxing Qi, Zhifei Yang, Ruifei Ma, Xiangde Liu, Chao Zhang. ICME 2025, Oral
• Argus: Leveraging Multiview Images for Improved 3-D Scene Understanding With Large Language Models. |[PDF]|
  Yifan Xu, Chao Zhang, Hanqi Jiang, Xiaoyan Wang, Ruifei Ma, Yiwei Li, Zihao Wu, Zeju Li, Xiangde Liu. TNNLS, 2025

Undergraduate Research

• Multi-source Education Knowledge Graph Construction and Fusion for College Curricula. |[PDF]|
  Zeju Li, Linya Cheng, Chunhong Zhang, Xinning Zhu, Hui Zhao. ICALT 2023, Oral, Best Paper Nomination
• SemBAT: Physical Layer Black-box Adversarial Attacks for Deep Learning-based Semantic Communication Systems.|[PDF]|
  Zeju Li, Jinfei Zhou, Guoshun Nan*, Zhichun Li, Qimei Cui, Xiaofeng Tao. VTC 2022-fall, Oral
• Boosting Physical Layer Black-Box Attacks with Semantic Adversaries in Semantic Communications. |[PDF]|
  Zeju Li, Xinghan Liu, Guoshun Nan, Jinfei Zhou, Xinchen Lyu, Qimei Cui, Xiaofeng Tao. ICC 2023, Oral
• SemProtector: A Unified Framework for Semantic Protection in Deep Learning-based Semantic Communication Systems.| [PDF]|
  Xinghan Liu, Guoshun Nan*, Qimei Cui, Zeju Li, Peiyuan Liu, Zebin Xing, Hanqing Mu, Yuanwei Liu, Xiaofeng Tao. IEEE Communication Magazine, SCI Q1
• Exploring the Distributed Knowledge Congruence in Proxy-data-free Federated Distillation. |[PDF]|
  Zhiyuan Wu, Sheng Sun, Min Liu, Quyang Pan, Junbo Zhang, Yuwei Wang*, Zeju Li, and Qingxiang Liu. ACM Transactions on Intelligent Systems and Technology, SCI Q1