Jonathan Uesato
10 papers · 2017–2022 · 6 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+7 more ↓ Show less ↑
πΊοΈ Taxonomy Completionist (21) π Conference Polyglot (6) π Academic Marathon (5) π Interdisciplinary Bridge π§ Keyword Pioneer
π
Interdisciplinary Bridge
π
Academic Marathon
(5)
π§¬
Topic Evolution
π
Trend Setter
π
Century Club
(10)
π₯
Unstoppable
(6)
β
The Questioner
Conferences
NIPS (4)
ICML (2)
CVPR (1)
EMNLP (1)
ICCV (1)
ICLR (1)
Top co-authors
Keywords
adversarial robustness
(6)
neural network verification
(3)
adversarial training
(2)
large language model
(2)
toxicity detection
(2)
language model evaluation
(1)
robust classification
(1)
semidefinite programming
(1)
harmful content
(1)
bayesian neural network
(1)
convex relaxation
(1)
interval bound propagation
(1)
bias mitigation
(1)
language model
(1)
loss landscape
(1)
first-order method
(1)
out-of-distribution detection
(1)
decision boundary
(1)
responsible ai
(1)
text generation
(1)
Papers
Characteristics of Harmful Text: Towards Rigorous Benchmarking of Language Models
NIPS 2022
Challenges in Detoxifying Language Models
EMNLP 2021
Make Sure You're Unsure: A Framework for Verifying Probabilistic Specifications
NIPS 2021
Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming
NIPS 2020
Are Labels Required for Improving Adversarial Robustness?
NIPS 2019
Robustness via Curvature Regularization, and Vice Versa
CVPR 2019
Scalable Verified Training for Provably Robust Image Classification
ICCV 2019
Verification of Non-Linear Specifications for Neural Networks
ICLR 2019
Adversarial Risk and the Dangers of Evaluating Against Weak Attacks
ICML 2018
RobustFill: Neural Program Learning under Noisy I/O
ICML 2017