Constantinos Karouzos

ckarouzos

https://ckarouzos.github.io/

AI & ML interests

Natural Language Processing, Machine Learning for NLP, NLP Applications, Financial and Legal NLP, Domain Adaptation, Unsupervised & Low-Resource Learning.

Recent Activity

authored a paper about 15 hours ago

Steerable Multi-Objective Group-Relative Preference Optimization with Jointly Trained Low-Rank Adapters

upvoted a paper about 18 hours ago

Steerable Multi-Objective Group-Relative Preference Optimization with Jointly Trained Low-Rank Adapters

submitted a paper about 18 hours ago

Steerable Multi-Objective Group-Relative Preference Optimization with Jointly Trained Low-Rank Adapters

View all activity

Organizations

None yet

authored a paper about 15 hours ago

Steerable Multi-Objective Group-Relative Preference Optimization with Jointly Trained Low-Rank Adapters

Paper • 2611.09634 • Published 4 days ago • 7

upvoted a paper about 18 hours ago

Steerable Multi-Objective Group-Relative Preference Optimization with Jointly Trained Low-Rank Adapters

Paper • 2611.09634 • Published 4 days ago • 7

submitted a paper to Daily Papers about 18 hours ago

Steerable Multi-Objective Group-Relative Preference Optimization with Jointly Trained Low-Rank Adapters

Paper • 2611.09634 • Published 4 days ago • 7

upvoted a collection about 1 month ago

efficient-reasoning

Collection

Models from "Efficient Reasoning Training Does Not Always Harm CoT Faithfulness and Monitorability". • 10 items • Updated Oct 5 • 2

upvoted 2 papers about 1 month ago

Efficient Reasoning Training Does Not Always Harm CoT Faithfulness and Monitorability

Paper • 2610.03509 • Published Oct 2 • 15

Synthetic Pre-pretraining Survives Scale, but Not as a Grammatical Prior

Paper • 2609.39827 • Published Sep 30 • 11

upvoted a paper 3 months ago

Logic Before Language: Pre-pretraining on Formal Derivations Fosters Skill Acquisition and Compressibility

Paper • 2608.03930 • Published Aug 4 • 10

upvoted a paper 4 months ago

MultiHashFormer: Hash-based Generative Language Models

Paper • 2606.28057 • Published Jun 26 • 23

authored a paper 7 months ago

Where does output diversity collapse in post-training?

Paper • 2604.16027 • Published Apr 17 • 21

upvoted 2 papers 7 months ago

Where does output diversity collapse in post-training?

Paper • 2604.16027 • Published Apr 17 • 21

UDALM: Unsupervised Domain Adaptation through Language Modeling

Paper • 2104.07078 • Published Apr 14, 2021 • 1

upvoted 2 papers 9 months ago

Fundamental Reasoning Paradigms Induce Out-of-Domain Generalization in Language Models

Paper • 2602.08658 • Published Feb 9 • 13

No Shortcuts to Culture: Indonesian Multi-hop Question Answering for Complex Cultural Understanding

Paper • 2602.03709 • Published Feb 3 • 8

authored 2 papers 10 months ago

UDALM: Unsupervised Domain Adaptation through Language Modeling

Paper • 2104.07078 • Published Apr 14, 2021 • 1

An Empirical Study on Preference Tuning Generalization and Diversity Under Domain Shift

Paper • 2601.05882 • Published Jan 9 • 21

upvoted 2 papers 10 months ago

An Empirical Study on Preference Tuning Generalization and Diversity Under Domain Shift

Paper • 2601.05882 • Published Jan 9 • 21

Enhancing Linguistic Competence of Language Models through Pre-training with Language Learning Tasks

Paper • 2601.03448 • Published Jan 6 • 13

upvoted a paper 11 months ago

Olmo 3

Paper • 2512.13961 • Published Dec 15, 2025 • 41

upvoted an article 11 months ago

Article

We Got Claude to Fine-Tune an Open Source LLM

burtenshaw, evalstate

•

Dec 4, 2025

• 636

upvoted a collection 11 months ago

Meta's Llama 3.1 models & evals

Collection

17 items • Updated Dec 13, 2024 • 642

Constantinos Karouzos

AI & ML interests

Recent Activity

Organizations

ckarouzos's activity

We Got Claude to Fine-Tune an Open Source LLM