Jos Rozen
9 papers · 2021–2025 · 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 ↓+2 more ↓ Show less ↑
🗺️ Taxonomy Completionist (29) 🐣 Hot Topic Early Bird 🌍 Conference Polyglot · Silver 🌉 Interdisciplinary Bridge · Silver 👥 Mega-Team · Squad
❓
The Questioner
🔥
Unstoppable · Bronze
Top co-authors
Research topics
Natural Language Processing
(6)
Artificial Intelligence
(5)
Machine Learning
(5)
Core AI
(5)
Deep Learning
(4)
Generation
(3)
Learning Types · Machine Learning
(3)
Resources & Methods
(3)
Application Areas
(2)
Applications
(2)
Learning Types · Deep Learning
(2)
Models
(2)
Optimization & Theory
(2)
Mathematics & Optimization
(1)
Core Methods
(1)
Keywords
preference alignment
(2)
large language model
(2)
language model
(2)
language model alignment
(2)
controlled generation
(2)
reinforcement learning from human feedback
(1)
self-supervised learning
(1)
multi-task learning
(1)
autoregressive generation
(1)
opinion summarization
(1)
automatic speech recognition
(1)
policy gradient
(1)
question answering
(1)
parameter efficient
(1)
distribution matching
(1)
importance sampling
(1)
constraint satisfaction
(1)
marginal probability
(1)
distribution learning
(1)
kl divergence
(1)
Papers
ELITR-Bench: A Meeting Assistant Benchmark for Long-Context Language Models
COLING 2025
FaST: Feature-aware Sampling and Tuning for Personalized Preference Alignment with Limited Data
EMNLP 2025
Guaranteed Generation from Large Language Models
ICLR 2025
Compositional Preference Models for Aligning LMs
ICLR 2024
Aligning Language Models with Preferences through f-divergence Minimization
ICML 2023
disco: a toolkit for Distributional Control of Generative Models
ACL 2023
Should you marginalize over possible tokenizations?
ACL 2023
Multitask Prompted Training Enables Zero-Shot Task Generalization
ICLR 2022
Self-Supervised and Controlled Multi-Document Opinion Summarization
EACL 2021