Get started
Author

Shivam Ratnakar

Observed name: Ratnakar, Shivam
OpenAlex ID: A5139725988
STARLING: Self-supervised Training of Text-based Reinforcement Learning Agent with Large Language Models
2024 · Shreyas Basavatia, Keerthiram Murugesan, et al. · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · 5
PDF
One Retrieval to Cover Them All: Co-occurrence-Aware Knowledge Base Reorganization for Session-Level RAG
2026 · Ratnakar, Shivam, Zhu, Yixuan, et al. · arXiv (Cornell University)
PDF
The Geometry of Refusal: Linear Instability in Safety-Aligned LLMs
2026 · Ratnakar, Shivam, Vats, Kartikeya · arXiv (Cornell University)
PDF
One Retrieval to Cover Them All: Co-occurrence-Aware Knowledge Base Reorganization for Session-Level RAG
2026 · Association for Computational Linguistics 2026, Cheng, Cecilia, et al. · Underline Science Inc.
LatentGate: Low-Latency Semantic Routing via Frozen-Backbone Probing of Small Language Models
2026 · Association for Computational Linguistics 2026, K Doifode, Vinayak, et al. · Underline Science Inc.
One Retrieval to Cover Them All: Co-occurrence-Aware Knowledge Base Reorganization for Session-Level RAG
2026 · Shivam Ratnakar, Yixuan Zhu, et al. · Workshop on Towards Knowledgeable Foundation Models (KnowFM)
PDF
The Geometry of Refusal: Linear Instability in Safety-Aligned LLMs
2026 · Shivam Ratnakar, Kartikeya Vats · Workshop on Trustworthy NLP (TrustNLP)
PDF
LatentGate: Low-Latency Semantic Routing via Frozen-Backbone Probing of Small Language Models
2026 · Shivam Ratnakar, Abhiroop Talasila, et al. · Proceedings of the conference - Association for Computational Linguistics. Meeting
PDF
The Chameleon Nature of LLMs: Quantifying Multi-Turn Stance Instability in Search-Enabled Language Models
2025 · Ratnakar, Shivam, Raghavendra, Sanjay · arXiv (Cornell University)
PDF
Beyond QA Pairs: Assessing Parameter-Efficient Fine-Tuning for Fact Embedding in LLMs
2025 · Ratnakar, Shivam, Talasila, Abhiroop, et al. · arXiv (Cornell University)
PDF
Works count: 11
Citation count: 5
H-index: 1
I10-index:
Year
Topic
Type