LabCompass

Siyuan Cheng

Computer Science · Purdue University West Lafayette

OpenAlex ORCID
Mid career · publishing since 2009Rising activity
Topic ModelingNatural Language Processing TechniquesAdversarial Robustness in Machine LearningSemantic Web and OntologiesMetaheuristic Optimization Algorithms Research

Publications

57

Citations

441

Est. group size

~4

Recurring co-author estimate

Active years

17

Publishing since 2009

Research summary
AI-generated

Siyuan Cheng's recent work focuses on how to update, correct, or remove specific pieces of knowledge stored inside large language models and knowledge graphs, a topic often called 'knowledge editing' or 'unlearning.' This includes building benchmarks, tools (like EasyEdit), and methods for editing both text-based and multimodal AI models, as well as related work on knowledge graph representations. Earlier and adjacent publications also touch on unrelated applied topics such as fault detection and image processing, suggesting some collaborative or cross-disciplinary work.

Knowledge editing in large language modelsKnowledge graphs and graph representation learningMultimodal AI modelsMachine unlearning and model safetyApplied signal/image processing (occasional)

Publication output grew sharply from near-zero before 2022 to a peak of 16-17 papers per year in 2023-2024, indicating a strongly increasing and currently active publication pace.

Generated by claude-sonnet-5 from public bibliographic data · Jul 20, 2026

Publication cadence
Publications per year over the last 10 years — averaging 9.4/year recently
172018: 1 publication182019: 1 publication192020: 1 publication20212022: 9 publications222023: 16 publications232024: 17 publications17242025: 5 publications2526
Recent publications
Publishes in
  • arXiv (Cornell University)×25
  • Proceedings of the AAAI Conference on Artificial Intelligence×2
  • Neurocomputing×1
  • SSRN Electronic Journal×1
  • Lecture notes in computer science×1
Similar researchers
By research-topic overlap

This profile was generated automatically from public scholarly data (OpenAlex). Group size and activity levels are estimates derived from co-authorship patterns.

Last updated Jul 20, 2026.

Claim or correct this profile