Starting Working at Minnesota NLP Lab
I began working at Minnesota NLP Lab under Prof. Dongyeop Kang.
News
Starting My Internship at Piper Sandler
I’m excited to announce that I’m starting my internship at Piper Sandler as a Technology Intern.
Publishing Our Work on Evaluating the Effectiveness of LLMs, CoBBLEr
Our study on evaluating the effectiveness of LLMs as automatic evaluators using a new benchmark, CoBBLEr has been published on ArXiv.
Introducing SelectLLM: A New Method for High-Quality Instruction Selection Using Large Language Models
Our work on selecting high-quality instructions from unlabeled datasets, called “SelectLLM”, has been published on ArXiv.
Submitting Our New Paper About Changes in Advertisement on AI-powered Search Engines
Our work exploring potential changes in search tactics and advertising engagement influenced by generative AI-powered search engines has been submitted to th...
Completing My Internship at Piper Sandler
I completed my internship at Piper Sandler.
Announcing the Acceptance of Our Paper on Evaluating the Effectiveness of LLMs to ACL 2024 Findings
Our research on Benchmarking Cognitive Biases in Large Language Models as Evaluators was accepted to appear at ACL 2024 Findings.
Announcing the Acceptance of Our Paper on Changes in Advertisement on AI-powered Search Engines to AEJMC Conference
Our research on “Consumer Engagement With AI-Powered Search Engines and Implications for the Future of Search Advertising” was accepted for presentation in A...
Graduating with an MS in Data Science from the University of Minnesota
I’m thrilled to announce that I have graduated with a Master’s degree in Data Science from the University of Minnesota.
Starting a New Position at University of Minnesota as Full-Stack Web Developer & Research Engineer
I’m happy to announce that I’m starting two new roles.
New Paper Drop: A Framework for Robust Cognitive Evaluation of LLMs
Just released CognitivEval, an open framework for systematically probing LLM cognition with prompt permutations and probability testing — arXiv.
New Paper Alert: Stealing Creator’s Workflow: A Creator-Inspired Agentic Framework with Iterative Feedback Loop for Improved Scientific Short-form Generation
SciTalk, a creator-inspired, multi-LLM workflow that converts complex papers into accurate, engaging short-form videos, was released on arXiv. Check out our ...
Announcing the Acceptance of Our Paper on Evaluating Cognitive Capabilities of LLMs to ACL 2025 Main
Our work on “How LLMs Comprehend Temporal Meaning in Narratives: A Case Study in Cognitive Evaluation of LLMs” was accepted to appear at ACL 2025 Main!
Farewell, MinnesotaNLP Lab
I wrapped up my role as a Research Engineer at MinnesotaNLP Lab.