Sim, Robert

Hybrid LLM: Cost-Efficient and Quality-Aware Query Routing

Large language models (LLMs) excel in most NLP tasks but also require expensive cloud servers for deployment due to their size, while smaller models that can be deployed on lower cost (e.g., edge)...

Simulating Environments with Reasoning Models for Agent Training

LLM agents excel in compact environments requiring deep reasoning but remain brittle when operating in broader, more complex contexts that demand robustness across diverse tools and schemas....

Privacy-Preserving In-Context Learning with Differentially Private Few-Shot Generation

We study the problem of in-context learning (ICL) with large language models (LLMs) on private datasets. This scenario poses privacy risks, as LLMs may leak or regurgitate the private examples...

Memory Makes the Difference: Evaluating How Different Memory Roles Shape Conversational Agents

Leveraging Structured Metadata for Improving Question Answering on the Web

Stereotyping Norwegian Salmon: An Inventory of Pitfalls in Fairness Benchmark Datasets

UserIdentifier: Implicit User Representations for Simple and Effective Personalized Sentiment Analysis

Privacy Leakage in Text Classification A Data Extraction Approach

Synthetic Text Generation with Differential Privacy: A Simple and Practical Recipe

CI-Work: Benchmarking Contextual Integrity in Enterprise LLM Agents