Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Sarah M. Erfani

Sarah M. Erfani

2 indexed papers

Recent (6 mo)
2
With code
0
Influential cites
0
Benchmarked
0

Publications per year

2
26

Top categories

ML×2AI×1Vision×1Audio and Speech Processing×1Crypto×1

Frequent co-authors

Hesam Asadollahzadeh1×
Feng Liu1×
Christopher Leckie1×
Sandra Arcos-Holzinger1×
James Bailey1×
Sanjeev Khudanpur1×

Research Timeline

2026
Dimensionality-Aware Anomaly Detection in Learned Representations of Self-Supervised Speech Models

The paper introduces GRIDS, a framework using Local Intrinsic Dimensionality (LID) to detect anomalies in self-supervised speech model representations, showing that LID elevation correlates with ASR degradation and enables transcript-free monitoring.

TRACER: Persistent Regularization for Robust Multimodal Finetuning

The paper introduces TRACER, a novel regularization framework that uses Weighted Moving Average (WMA) distillation to robustly finetune multimodal models, mitigating catastrophic forgetting and improving out-of-distribution performance.

Highlighted terms show continued research focus across papers

Papers

cs.LGcs.AIcs.CVRecentMay 28, 2026

TRACER: Persistent Regularization for Robust Multimodal Finetuning

Hesam Asadollahzadeh, Feng Liu, Christopher Leckie, Sarah M. Erfani

The paper introduces TRACER, a novel regularization framework that uses Weighted Moving Average (WMA) distillation to robustly finetune multimodal models, mitigating catastrophic forgetting and improv…

View →
eess.AScs.CRcs.LGRecentMay 4, 2026

Dimensionality-Aware Anomaly Detection in Learned Representations of Self-Supervised Speech Models

Sandra Arcos-Holzinger, Sarah M. Erfani, James Bailey, Sanjeev Khudanpur

The paper introduces GRIDS, a framework using Local Intrinsic Dimensionality (LID) to detect anomalies in self-supervised speech model representations, showing that LID elevation correlates with ASR d…

View →