5 papers
When to Review: Spaced Repetition for Continual Pre-Training of Language Models
Alankar Atreya, Devesh Batra, Yoages Kumar Mantri +3
Continual pre-training of large language models must acquire new information without erasing old knowledge. Existing replay methods often choose a global old/new mixture and sample…
Large-Scale ChatBot Validation Through Customer Digital Twin Simulations
Cristovao Iglesias, Devesh Batra, Alankar Atreya +7
LLM-based chatbots are transforming customer service in regulated domains such as banking, but scalable and cost-effective validation remains a critical barrier to safe deployment.…
Helping Customers in Distress: An LLM-powered Agent that Converses, Probes, and Routes
Alankar Atreya, Stefan Sylvius Wanger, Devesh Batra +7
Banks receive millions of reports of fraud, scams, and disputed transactions every year, making it challenging to accurately direct customers to the appropriate specialist teams fo…
Evaluating Performance Drift from Model Switching in Multi-Turn LLM Systems
Raad Khraishi, Iman Zafar, Katie Myles +1
Deployed multi-turn LLM systems routinely switch models mid-interaction due to upgrades, cross-provider routing, and fallbacks. Such handoffs create a context mismatch: the model g…
Evaluating the Sensitivity of LLMs to Prior Context
Robert Hankache, Kingsley Nketia Acheampong, Liang Song +3
As large language models (LLMs) are increasingly deployed in multi-turn dialogue and other sustained interactive scenarios, it is essential to understand how extended context affec…