CiCi Yutong Cheng

Ph.D. candidate @Virginia Tech, Department of Computer Science.

prof_pic.jpg

About Me

Hi, I’m CiCi! I build self-evolving, continual-learning agents for long-horizon, open-ended tasks. My research focuses on:

  • Code-as-harness — framing harness evolution as a coding task, so agents self-improve by writing and rewriting their own harness as code.
  • Agent memory and retrieval — building long-term memory as self-organizing structure that links, consolidates, and ranks experience into reusable knowledge.
  • Test-time optimization — task-time adaptation by selecting among rollouts guided by execution-grounded verification.

Internships & Experience

  • Research intern, NEC Laboratories America, advised by Dr. Wei Cheng, 05/2026-present, building end-to-end long-horizon programming agent with self-evolving harness and memory.
  • Research intern, NEC Laboratories America, advised by Dr. Wei Cheng, 01/2026-3/2026, working on inference-time scaling for code documentation and test generation with execution feedback.
  • Research intern, DeepWisdom, 10/2024-12/2024, working on testing LLM-generated software with static and dynamic analysis in sandboxed environments.

Awards

  • 2026 ICML Golden Reviewer Award.
  • 2025 CCI SWVA Cyber Innovation Scholarship.
  • 2024 CCI SWVA Cyber Innovation Scholarship.
  • 2024 Bitshares Fellowship.

Services

  • Reviewer, ICML 2026, NeurIPS 2026, COLM 2026.
  • Student Organizer, 2024 DMV Security Workshop.

Selected Publications

  1. KDD 2026
    CTIConnect: A Benchmark for Retrieval-Augmented LLMs over Heterogeneous Cyber Threat Intelligence
    Yutong Cheng , Yang Liu, Changze Li, Dawn Song, and Peng Gao
    In Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining, SIGKDD, 2026
    Paper Code Website
  2. ICML 2026
    Escaping Whack-a-Mole: Code Documentation Optimization via Dependency-Guided Bi-level Search
    Yutong Cheng , Haifeng Chen, Wenchao Yu, Xujiang Zhao, Peng Gao, and Wei Cheng
    In Proceedings of the 43rd International Conference on Machine Learning, ICML, 2026
    Adopted by NEC for repo-level test and documentation generation across large-scale Go and Java legacy codebases.
    Paper Code
  3. EACL 2026
    NL2Logic: AST Guided Translation of Natural Language into First-Order Logic with Large Language Models
    Rizky Ramadhana Putra, Raihan Sultan Pasha Basuki,  Yutong Cheng , and Peng Gao
    In Findings of the 18th Conference of the European Chapter of the Association for Computational Linguistics, EACL, 2026
    Paper Code
  4. Euro S&P 2025
    CTINexus: Automatic Cyber Threat Intelligence Knowledge Graph Construction Using Large Language Models
    Yutong Cheng , Osama Bajaber, Saimon Amanuel Tsegai, Dawn Song, and Peng Gao
    In Proceedings of the 10th IEEE European Symposium on Security and Privacy, Euro S&P, 2025
    Adopted by Palo Alto Networks, ThreatConnect, and multiple other security companies for automated threat intelligence analysis
    Tutorial presented at the PRISM Workshop of NDSS 2026.
    Paper Code Website PyPI Downloads
  5. ESEC/FSE 2023
    Hue: A user-adaptive parser for hybrid logs
    Junjielong Xu, Qiuai Fu, Zhouruixing Zhu,  Yutong Cheng , Zhijing Li, Yuchi Ma, and Pinjia He
    In Proceedings of the 31st ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering, ESEC/FSE, 2023
    Paper Code