Data-Efficient Language Modelling

Thumbnail

Event details

Date 02.06.2025
Hour 10:0012:00
Speaker Vinko Sabolcec
Location
Category Conferences - Seminars
EDIC candidacy exam
Exam president: Prof. Volkan Cevher
Thesis advisor: Prof. Martin Jaggi
Co-examiner: Prof. Antoine Bosselut

Abstract
coming soon

Selected papers
1. DataComp-LM: In search of the next generation of training sets for language models; https://arxiv.org/abs/2406.11794
2. Metadata Conditioning Accelerates Language Model Pre-training; https://arxiv.org/abs/2501.01956
3. s1: Simple test-time scaling; https://arxiv.org/abs/2501.19393;

Practical information

  • General public
  • Free

Contact

  • edic@epfl.ch

Tags

EDIC candidacy exam

Share