San Francisco, CA
I am a PhD student in Linguistics at UC Berkeley working on interpretability and on how deep learning can help us understand language and human cognition. I am also currently a research scientist intern at Snowflake, where I work on reinforcement learning, agent behavioral analysis, and model routing.
NeurIPS 2025 / PMLR
TL;DR: This study investigates the neural correlates between language models and human brain activity, demonstrating that the language representations underlying modern language models align specifically with human language processing patterns rather than general sequence processing. I also show shared geometrical properties of LM and brain representations.
Paper
AACL 2025
TL;DR: I proposed a novel application of SAEs to understand self-supervised speech models through interpretable sparse feature extraction. The method revealed linguistically meaningful patterns in SAE features, offering new insights into what these models learn about phonetic and phonological structure.
Paper
AAAI 2025
TL;DR: We introduced a conceptual-criticality framework for adaptive compute allocation in neural networks, enabling models to dynamically adjust computational resources based on input complexity. This approach achieves significant efficiency gains while maintaining or improving model performance.
Paper
NeurIPS 2024
TL;DR: This project explores how morphological typology affects BPE tokenization efficiency and language model performance across diverse languages. I demonstrated that morphology has a major effect on tokenization, and that this also affects training.
Paper
ACL 2024
TL;DR: I investigated whether speech enhancement techniques preserve or distort fine-grained linguistic features. My findings revealed that noise reduction improves overall intelligibility, and that enhancement methods do not significantly alter phonetic distinctions critical for downstream linguistic analysis.
Paper
EACL 2024
TL;DR: Using linguistically informed job market queries, I developed a methodology to quantify gender biases in masked language models. The analysis revealed systematic biases in occupational associations and provides insights into how language models perpetuate societal stereotypes.
PaperDemonstrating how sinusoidal positional encodings work in transformer models
Visualizing the Query, Key, and Value matrices in attention mechanisms
Breaking down the mathematical formulation of Generative Adversarial Networks
“All models are wrong, but some are useful.”
“Intuitions have been tacitly granted a privileged position in generative grammar. The result has been the construction of elaborate theoretical edifices supported by disturbingly shaky empirical evidence.”
“Being a native speaker doesn’t confer papal infallibility on one’s intuitive judgments.”