pith. sign in

Sebastian Lapuschkin

Identifiers

  • name variant Sebastian Lapuschkin 0.60 · backfill

Papers (15)

  1. Fast & Faithful Function Vectors cs.CL · 2026 · author #5
  2. Playing the network backward: A Game Theoretic Attribution Framework cs.LG · 2026 · author #4
  3. Contrastive Semantic Projection: Faithful Neuron Labeling with Contrastive Examples cs.CV · 2026 · author #4
  4. From Weights to Activations: Is Steering the Next Frontier of Adaptation? cs.CL · 2026 · author #5
  5. From Attribution to Action: A Human-Centered Application of Activation Steering cs.AI · 2026 · author #5
  6. X-SYS: A Reference Architecture for Interactive Explanation Systems cs.AI · 2026 · author #8
  7. Atlas-Alignment: Making Interpretability Transferable Across Language Models cs.LG · 2025 · author #3
  8. Attribution-Guided Pruning for Insight and Control: Circuit Discovery and Targeted Correction in Small-scale LLMs cs.LG · 2025 · author #8
  9. FADE: Why Bad Descriptions Happen to Good Features cs.LG · 2025 · author #7
  10. PINNfluence: Interpreting PINNs through Influence Functions cs.LG · 2024 · author #6
  11. Unmasking Clever Hans Predictors and Assessing What Machines Really Learn cs.AI · 2019 · author #1
  12. Explaining the Unique Nature of Individual Gait Patterns with Deep Learning cs.LG · 2018 · author #2
  13. iNNvestigate neural networks! cs.LG · 2018 · author #2
  14. Understanding and Comparing Deep Neural Networks for Age and Gender Classification stat.ML · 2017 · author #1
  15. Interpreting the Predictions of Complex ML Models by Layer-wise Relevance Propagation stat.ML · 2016 · author #4

Mentions

  • 2606.05079 #5 · arxiv_oai · confidence 0.70 Sebastian Lapuschkin
  • 2409.08958 #6 · arxiv_oai · confidence 0.70 Sebastian Lapuschkin
  • 2604.11467 #5 · arxiv_oai · confidence 0.70 Sebastian Lapuschkin

Frequent Coauthors