Lukas Thoma, BA MA
1090 Wien
Room : 05.20
Publications
Influence-driven Curriculum Learning for Pre-training on Limited Data. / Schoenegger, Loris; Thoma, Lukas; Blevins, Terra et al.
2025. 356–379 Paper presented at The First BabyLM Workshop at the Conference
on Empirical Methods in Natural Language Processing 2025, Suzhou, China.
Publications: Contribution to conference › Paper › Peer Reviewed
RecombiText: Compositional Data Augmentation for Enhancing LLM Pre-Training Datasets in Low-Resource Scenarios. / Tampier, Alexander; Thoma, Lukas; Schoenegger, Loris et al.
2025. Paper presented at The First BabyLM Workshop at the Conference
on Empirical Methods in Natural Language Processing 2025, Suzhou, China.
Publications: Contribution to conference › Paper › Peer Reviewed
CogMemLM: Human-Like Memory Mechanisms Improve Performance and Cognitive Plausibility of LLMs. / Thoma, Lukas; Weyers, Ivonne; Çano, Erion et al.
The BabyLM Challenge at the 27th Conference on Computational Natural Language Learning: Proceedings of the BabyLM Challenge. ed. / Alex Warstadt; Aaron Mueller; Leshem Choshen; Ethan Wilcox; Chengxu Zhuang; Juan Ciro; Rafael Mosquera; Bhargavi Paranjabe; Adina Williams; Tal Linzen; Ryan Cotterell. Stroudsburg: Association for Computational Linguistics (ACL), 2023. p. 180-185.
Publications: Contribution to book › Contribution to proceedings › Peer Reviewed
Projects
Cognitive Plausibility of Deep Learning Language Models
Thoma, L. (Project Lead), Blevins, T. (Co-Lead), Schoenegger, L. (Project Staff), Weyers, I. (Project Lead) & Çano, E. (Affiliated Project Staff)
1/08/22 → 31/01/25
Project: Research funding
More: Management & Team | Staff directory | Home