Mariya Toneva

Assistant Professor@Max Planck Institute for Software Systems·德国·OpenReview

研究方向

speech models · natural language processing · interpretability · language models · machine learning · neuroimaging (fMRI · MEG · EEG) · computational cognitive neuroscience

6.0

Brain-tuning Improves Generalizability and Efficiency of Brain Alignment in Speech Models

NeurIPS 2025Poster

二作

7.3

Large Language Models as Model Organisms for Human Associative Learning

NeurIPS 2025Poster

三作

6.5

Improving Semantic Understanding in Speech Language Models via Brain-tuning

ICLR 2025Poster

三作

6.0

Positional Biases Shift as Inputs Approach Context Window Limits

COLM 2025Poster

三作

5.0

Assessing Episodic Memory in LLMs with Sequence Order Recall Tasks

合作者 (20)

Mariya Toneva

Tracking Equivalent Mechanistic Interpretations Across Neural Networks

When Language Models Lose Their Mind: The Consequences of Brain Misalignment

Fine-grained Analysis of Brain-LLM Alignment through Input Attribution

Brain-tuning Improves Generalizability and Efficiency of Brain Alignment in Speech Models

Large Language Models as Model Organisms for Human Associative Learning

Improving Semantic Understanding in Speech Language Models via Brain-tuning

Positional Biases Shift as Inputs Approach Context Window Limits

Assessing Episodic Memory in LLMs with Sequence Order Recall Tasks