Maksim Kulik

Maksim Kulik

Future MS in Human Language Technology, interested in Large Language Models, machine translation, and text generation

Portfolio

XRI Global: LLMs for low-resource languages

At XRI global, I helped develop a scoring system for assessing the impact of digital divide on any low-resource language. I also built a solution for aggregating LLMs that support low-resource languages, curated a collection of benchmarks for evaluating LLM performance on low-resource languages and measured the performance of the models using the Global MMLU dataset

A light ReAct agent based on GPT-4o-mini

As part of my work for the Applied NLP course at the University of Arizona, I tested the effectiveness of lightweight language models in common sense reasoning tasks by comparing 1-step and 2-step ReAct agents using GPT-4o mini and the TextWorldExpress Commonsense environment.