Lune

学者档案2023-2026

Mariya Toneva

Max Planck Institute for Software Systems

11论文数
11末位作者
16合作者人数
0一作
0独著

一作与末位作者按论文署名顺序统计,部分顶会采用姓氏字母顺序署名。

近期论文11

  1. When Language Models Lose Their Mind: The Consequences of Brain Misalignment第 2 位,共 2 位作者ICLR 2026 · 3 次引用
  2. Tracking Equivalent Mechanistic Interpretations Across Neural Networks第 2 位,共 2 位作者ICLR 2026 · 1 次引用
  3. Fine-grained Analysis of Brain-LLM Alignment through Input Attribution第 3 位,共 3 位作者ICML 2026
  4. Brain-tuning Improves Generalizability and Efficiency of Brain Alignment in Speech Models第 2 位,共 2 位作者NeurIPS 2025 · 7 次引用
  5. Hints Help Finding and Fixing Bugs Differently in Python and Text-Based Program Representations第 5 位,共 5 位作者ICSE 2025 · 1 次引用
  6. Large Language Models as Model Organisms for Human Associative Learning第 3 位,共 3 位作者NeurIPS 2025 · 1 次引用
  7. Improving Semantic Understanding in Speech Language Models via Brain-tuning第 3 位,共 3 位作者ICLR 2025
  8. Language models and brains align due to more than next-word prediction and word-level information第 2 位,共 2 位作者EMNLP 2024 · 2 次引用
  9. Speech language models lack important brain-relevant semantics第 4 位,共 4 位作者ACL 2024
  10. Joint processing of linguistic properties in brains and language models第 3 位,共 3 位作者NeurIPS 2023 · 64 次引用
  11. Training language models to summarize narratives improves brain alignment第 2 位,共 2 位作者ICLR 2023 · 11 次引用
黄昏的海面,两侧是细线勾勒的悬崖