Grigorios G. Chrysos
Scholar

Grigorios G. Chrysos

Google Scholar ID: 1bU041kAAAAJ
Assistant Professor at University of Wisconsin-Madison
Machine LearningReliable MLLearning efficiency
Citations & Impact
All-time
Citations
2,011
 
H-index
21
 
i10-index
35
 
Publications
20
 
Co-authors
23
list available
Resume (English only)
Academic Achievements
  • Paper 'Multilinear Mixture of Experts: Scalable Expert Specialization through Factorization' accepted at NeurIPS 2024
  • Paper 'Revisiting character-level adversarial attacks for Language Models' accepted at ICML 2024
  • Paper 'Going beyond compositional generalization, DDPMs can produce zero-shot interpolation' accepted at ICML 2024
  • Paper 'Towards Interpretability Without Sacrifice: Faithful Dense Layer Decomposition with Mixture of Decoders' accepted at NeurIPS 2025
  • Area Chair for NeurIPS (2024, 2025), ICLR (2025, 2026), ICML (2025)
  • Action Editor for TMLR
  • Outstanding reviewer awards at NeurIPS'23, NeurIPS'22, ICML'22, ICLR'22, ICML'21
  • Organizing tutorial 'Foundations of Tensor Computations for AI' at NeurIPS'25
  • Organizing IJCV special issue on 'Post-training in Large Language Models for Computer Vision'
  • Organized multiple tutorials and workshops at top conferences including CVPR, ICLR, AAAI, NeurIPS, DSAA, and ISIT
Background
  • Assistant Professor at University of Wisconsin-Madison
  • Research focuses on trustworthy machine learning
  • Passionate about developing innovative algorithms to understand and interpret data in complex real-world scenarios
  • Recruiting motivated PhD students from ECE or CS programs
  • Teaching course ECE/Stat/Math 888 with a significant component on trustworthy machine learning