Piotr Sankowski
9 papers · 2015–2025 · 6 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+1 more ↓ Show less ↑
🌍 Conference Polyglot (6) 🏃 Academic Marathon (10) 🌉 Interdisciplinary Bridge 🗺️ Taxonomy Completionist (15) 🧭 Keyword Pioneer
🐝
Cross-Pollinator
(15)
Conferences
ICML (4)
AAAI (1)
ACL (1)
EMNLP (1)
IJCAI (1)
UAI (1)
Top co-authors
Keywords
large language model
(2)
feature importance
(2)
shapley value
(2)
content moderation
(1)
instruction following
(1)
feature attribution
(1)
model interpretability
(1)
discrete optimization
(1)
maximum flow
(1)
tree models
(1)
parametric optimization
(1)
tree ensemble
(1)
decision tree
(1)
model explanation
(1)
robustness evaluation
(1)
feature ranking
(1)
banzhaf value
(1)
a/b testing
(1)
user engagement
(1)
hate speech detection
(1)
Papers
Accurate Estimation of Feature Importance Faithfulness for Tree Models
AAAI 2025
Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient
ICML 2025
Wait, that’s not an option: LLMs Robustness with Incorrect Multiple-Choice Options
ACL 2025
Since Faithfulness Fails: The Performance Limits of Neural Causal Discovery
ICML 2025
Scaling Laws for Fine-Grained Mixture of Experts
ICML 2024
LLM generated responses to mitigate the impact of hate speech
EMNLP 2024
Improved feature importance computation for tree models based on the Banzhaf value
UAI 2022
Decomposable Submodular Function Minimization via Maximum Flow
ICML 2021
Revenue Maximization Envy-Free Pricing for Homogeneous Resources
IJCAI 2015