5 papers
One Retrieval to Cover Them All: Co-occurrence-Aware Knowledge Base Reorganization for Session-Level RAG
Shivam Ratnakar, Yixuan Zhu, Cecilia Cheng +1
RAG systems retrieve documents optimized for answering one query at a time. Yet enterprise users arrive with sessions, that is, coherent episodes of related questions that span sem…
The Geometry of Refusal: Linear Instability in Safety-Aligned LLMs
Shivam Ratnakar, Kartikeya Vats
Modern Large Language Models (LLMs) rely on extensive safety alignment, yet the mechanistic basis of refusal remains opaque. In this work, we investigate whether safety compliance…
The Chameleon Nature of LLMs: Quantifying Multi-Turn Stance Instability in Search-Enabled Language Models
Shivam Ratnakar, Sanjay Raghavendra
Integration of Large Language Models with search/retrieval engines has become ubiquitous, yet these systems harbor a critical vulnerability that undermines their reliability. We pr…
Beyond QA Pairs: Assessing Parameter-Efficient Fine-Tuning for Fact Embedding in LLMs
Shivam Ratnakar, Abhiroop Talasila, Raghav Chamadiya +2
This paper presents an extensive examination of Parameter-Efficient Fine-Tuning (PEFT) for embedding domain specific facts into Large Language Models (LLMs), focusing on improving…
STARLING: Self-supervised Training of Text-based Reinforcement Learning Agent with Large Language Models
Shreyas Basavatia, Keerthiram Murugesan, Shivam Ratnakar
Interactive fiction games have emerged as an important application to improve the generalization capabilities of language-based reinforcement learning (RL) agents. Existing environ…