3 papers
cs.CL2026
Generation-Step-Aware Framework for Cross-Modal Representation and Control in Multilingual Speech-Text Models
Toshiki Nakai, Varsha Suresh, Vera Demberg
Multilingual speech-text models rely on cross-modal language alignment to transfer knowledge between speech and text, but it remains unclear whether this reflects shared computatio…
cs.CL2025
TRepLiNa: Layer-wise CKA+REPINA Alignment Improves Low-Resource Machine Translation in Aya-23 8B
Toshiki Nakai, Ravi Kiran Chikkala, Lena Sophie Oberkircher +4
The 2025 Multimodal Models for Low-Resource Contexts and Social Impact (MMLoSo) Language Challenge addresses one of India's most pressing linguistic gaps: the lack of resources for…
cs.CL2025
What Language(s) Does Aya-23 Think In? How Multilinguality Affects Internal Language Representations
Katharina Trinley, Toshiki Nakai, Tatiana Anikina +1
Large language models (LLMs) excel at multilingual tasks, yet their internal language processing remains poorly understood. We analyze how Aya-23-8B, a decoder-only LLM trained on…