4 papers
ExecTune: Effective Steering of Black-Box LLMs with Guide Models
Vijay Lingam, Aditya Golatkar, Anwesan Pal +6
For large language models deployed through black-box APIs, recurring inference costs often exceed one-time training costs. This motivates composed agentic systems that amortize exp…
ELLA: Efficient Lifelong Learning for Adapters in Large Language Models
Shristi Das Biswas, Yue Zhang, Anwesan Pal +2
Large Language Models (LLMs) suffer severe catastrophic forgetting when adapted sequentially to new tasks in a continual learning (CL) setting. Existing approaches are fundamentall…
Tagging-Augmented Generation: Assisting Language Models in Finding Intricate Knowledge In Long Contexts
Anwesan Pal, Karen Hovsepian, Tinghao Guo +5
Recent investigations into effective context lengths of modern flagship large language models (LLMs) have revealed major limitations in effective question answering (QA) and reason…
The Empirical Impact of Data Sanitization on Language Models
Anwesan Pal, Radhika Bhargava, Kyle Hinsz +2
Data sanitization in the context of language modeling involves identifying sensitive content, such as personally identifiable information (PII), and redacting them from a dataset c…