Nazia Tasnim
Scholar

Nazia Tasnim

Google Scholar ID: _9JYPbsAAAAJ
Boston University
PEFT & Model EditingComputer VisionExplainable AIMultimodal Systems
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
87
 
H-index
4
 
i10-index
4
 
Publications
10
 
Co-authors
0
 
Contact
Emailnazia@bu.eduTwitterOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
9 items
Decompose, Mix, Adapt: A Unified Framework for Parameter-Efficient Neural Network Recombination and Compression
2026
Cited
0
Seeing Isn't Orienting: A Cognitively Grounded Benchmark Reveals Systematic Orientation Failures in MLLMs Supplementary
2026
Cited
0
Representation-Aware Unlearning via Activation Signatures: From Suppression to Knowledge-Signature Erasure
2026
Cited
1
RegSpeech12: A Regional Corpus of Bengali Spontaneous Speech Across Dialects
2025
Cited
0
Are ASR foundation models generalized enough to capture features of regional dialects for low-resource languages?
2025
Cited
0
From Facts to Folklore: Evaluating Large Language Models on Bengali Cultural Knowledge
2025
Cited
0
Right Side Up? Disentangling Orientation Understanding in MLLMs with Fine-grained Multi-axis Perception Tasks
2025
Cited
0
RECAST: Reparameterized, Compact weight Adaptation for Sequential Tasks
arXiv.org · 2024
Cited
0
Resume
Academic Achievements
  • Published in multiple A* conferences and workshops, including Interspeech, NeurIPS, EMNLP, and CVPR.
  • Two papers accepted at IJCNLP'25 (October 2025).
  • Second PhD paper posted on arXiv in May 2025.
  • Presented paper at ICLR'25 in Singapore (April 2025).
  • Attended CVPR'25 as a WiCV mentee in June 2025.
  • Develops open-source tools and datasets to support the research community.