Om Dipakbhai Thakkar
4 papers · 2018–2022 · 3 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+2 more ↓ Show less ↑
π Interdisciplinary Bridge π§ Keyword Pioneer π£ Hot Topic Early Bird π Conference Polyglot (3) π Cross-Pollinator (11)
π
Renaissance Researcher
(5)
π
Trend Setter
Conferences
INTERSPEECH (2)
ICML (1)
NAACL (1)
Top co-authors
Research topics
Keywords
language model
(2)
speech recognition
(2)
differential privacy
(2)
privacy preservation
(1)
gradient descent
(1)
matrix completion
(1)
gradient clipping
(1)
privacy leakage
(1)
frank-wolfe method
(1)
training data extraction
(1)
singular vector computation
(1)
next word prediction
(1)
unintended memorization
(1)
federated learning
(1)
memorization detection
(1)
collaborative filtering
(1)
data augmentation
(1)
Papers
Extracting Targeted Training Data from ASR Models, and How to Mitigate It
INTERSPEECH 2022
Detecting Unintended Memorization in Language-Model-Fused ASR
INTERSPEECH 2022
Understanding Unintended Memorization in Language Models Under Federated Learning
NAACL 2021
Differentially Private Matrix Completion Revisited
ICML 2018