3 papers
cs.LG2025
Learning Obfuscations Of LLM Embedding Sequences: Stained Glass Transform
Jay Roberts, Kyle Mylonakis, Sidhartha Roy +1
The high cost of ownership of AI compute infrastructure and challenges of robust serving of large language models (LLMs) has led to a surge in managed Model-as-a-service deployment…
cs.CL2025
THELMA: Task Based Holistic Evaluation of Large Language Model Applications-RAG Question Answering
Udita Patel, Rutu Mulkar, Jay Roberts +6
We propose THELMA (Task Based Holistic Evaluation of Large Language Model Applications), a reference free framework for RAG (Retrieval Augmented generation) based question answerin…
cs.CR2025
BeamClean: Language Aware Embedding Reconstruction
Kaan Kale, Kyle Mylonakis, Jay Roberts +1
In this work, we consider an inversion attack on the obfuscated input embeddings sent to a language model on a server, where the adversary has no access to the language model or th…