Paper 'ToolDelete' accepted to ICML 2025 — the first work on unlearning tools from tool-augmented LLMs
Awarded Computer Science Outstanding Graduate Research Award for Ph.D. Students at UMass Lowell (2025)
Two preprints released: 'Exploring Mode Connectivity in the context of Unlearning' and 'Fair Graph Unlearning'
Two papers accepted to NAACL 2025: main conference short paper 'EqualizeIR: Mitigating Linguistic Biases in Retrieval Models' and CMCL workshop paper 'Linguistic Blind Spots of Large Language Models'
MU-Bench accepted to NeurIPS 2024 SafeGenAI Workshop
FairFlow accepted to EMNLP 2024 main conference
MultiDelete accepted to ECCV 2024
Preprint of MU-Bench released — the first multitask multimodal comprehensive benchmark for Machine Unlearning
One paper accepted to ACL 2024 and one to INTERSPEECH 2024