John Thickstun
8 papers · 2020–2023 · 6 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓+1 more ↓ Show less ↑
๐ Conference Polyglot ยท Silver ๐ Interdisciplinary Bridge ยท Silver ๐บ๏ธ Taxonomy Completionist (23) ๐ฃ Hot Topic Early Bird ๐ Keyword Champion
๐
Renaissance Researcher
(5)
Top co-authors
Research topics
Machine Learning
(7)
Deep Learning
(6)
Optimization & Theory
(6)
Models
(5)
Natural Language Processing
(3)
Generation
(3)
Learning Types
(3)
Artificial Intelligence
(2)
Application Areas
(2)
Core AI
(2)
Core Methods
(2)
Computer Vision
(1)
Reinforcement Learning
(1)
Speech & Audio
(1)
Applications
(1)
Keywords
text generation
(3)
generative model
(3)
controllable generation
(2)
langevin dynamics
(2)
source separation
(2)
language modeling
(2)
language model
(2)
information bottleneck
(1)
bayesian inference
(1)
semi-supervised learning
(1)
feature selection
(1)
rationale extraction
(1)
policy learning
(1)
speech enhancement
(1)
image restoration
(1)
distribution matching
(1)
posterior distribution
(1)
model interpretability
(1)
adaptive discretization
(1)
neural architecture
(1)
Papers
Backpack Language Models
ACL 2023
MAUVE Scores for Generative Models: Theory and Practice
JMLR 2023
Diffusion-LM Improves Controllable Text Generation
NeurIPS 2022
Faster Policy Learning with Continuous-Time Gradients
L4DC 2021
MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontiers
NeurIPS 2021
Parallel and Flexible Sampling from Autoregressive Models via Langevin Dynamics
ICML 2021
An Information Bottleneck Approach for Controlling Conciseness in Rationale Extraction
EMNLP 2020
Source Separation with Deep Generative Priors
ICML 2020