500AI
← People

Simin Chen

Source-listed: Postdoctoral Researcher, Computer Science Department · Columbia University

  • Language models
  • Evaluation

Selected work

5
  1. Benchmarking Large Language Models Under Data Contamination: A Survey from Static to Dynamic EvaluationFirst author · 2025
  2. Your Compiler is Backdooring Your Model: Understanding and Exploiting Compilation Inconsistency Vulnerabilities in Deep Learning CompilersFirst author · 2026
  3. DyCodeEval: Dynamic Benchmarking of Reasoning Capabilities in Code Large Language Models Under Data ContaminationFirst author · 2025
  4. PPM: Automated Generation of Diverse Programming Problems for Benchmarking Code Generation ModelsFirst author · 2024
  5. The Dark Side of Dynamic Routing Neural Networks: Towards Efficiency Backdoor InjectionFirst author · 2023