Lune

Profiles2022-2024

Mark Niklas Müller

Department of Computer Science ETH Zurich

14Papers
0Last author
18Unique co-authors
3First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers14

  1. SWT-Bench: Testing and Validating Real-World Bug-Fixes with Code AgentsAuthor 2 of 4NeurIPS 2024 · 172 citations
  2. ConStat: Performance-Based Contamination Detection in Large Language ModelsAuthor 2 of 3NeurIPS 2024 · 40 citations
  3. DAGER: Exact Gradient Inversion for Large Language ModelsAuthor 4 of 8NeurIPS 2024 · 29 citations
  4. SPEAR: Exact Gradient Inversion of Batches in Federated LearningAuthor 3 of 7NeurIPS 2024 · 29 citations
  5. Understanding Certified Training with Interval Bound PropagationAuthor 2 of 4ICLR 2024 · 26 citations
  6. Expressivity of ReLU-Networks under Convex RelaxationsAuthor 2 of 4ICLR 2024 · 7 citations
  7. Prompt Sketching for Large Language ModelsAuthor 2 of 4ICML 2024 · 6 citations
  8. Connecting Certified and Adversarial TrainingAuthor 2 of 4NeurIPS 2023 · 14 citations
  9. Certified Training: Small Boxes are All You NeedAuthor 1 of 4ICLR 2023 · 10 citations
  10. Automated Classification of Model Errors on ImageNetAuthor 2 of 4NeurIPS 2023 · 9 citations
  11. Abstract Interpretation of Fixpoint Iterators with Applications to Neural NetworksAuthor 1 of 4PLDI 2023 · 5 citations
  12. Efficient Certified Training and Robustness Verification of Neural ODEsAuthor 2 of 4ICLR 2023
  13. Complete Verification via Multi-Neuron Relaxation Guided Branch-and-BoundAuthor 2 of 4ICLR 2022 · 117 citations
  14. PRIMA: general and precise neural network certification via scalable convex hull approximationsAuthor 1 of 6POPL 2022 · 75 citations
Dusk over the sea between two cliffs drawn in fine vertical lines