About
Research
Research Interests:
Machine Learning
Natural Language Processing
Large Language Models
Synthetic Data
Unsupervised Learning
Text Style Transfer
AI Applications
- FineInstructions: A Web-Scale Instructions Dataset by Ajay Patel, Colin Raffel, Chris Callison-Burch (Working Manuscript)
- Scaling Text-Rich Image Understanding via Code-Guided Synthetic Multimodal Data Generation by Yue Yang*, Ajay Patel*, Matt Deitke, Luca Weihs, Tanmay Gupta, Ranjay Krishna, Andrew Head, Chris Callison-Burch, Mark Yatskar, Aniruddha Kembhavi, Christopher Clark (ACL 2025)
- Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models by Matt Deitke et al. (CVPR 2025)
- Quantifying Misattribution Unfairness in Authorship Attribution by Pegah Alipoormolabashi, Ajay Patel, Niranjan Balasubramanian (ACL 2025)
- mStyleDistance: Multilingual Style Embeddings and their Evaluation by Justin Qiu*, Jiacheng Zhu*, Ajay Patel, Marianna Apidianaki, Chris Callison-Burch (Findings of ACL 2025)
- StyleDistance: Stronger Content-Independent Style Embeddings with Synthetic Parallel Examples by Ajay Patel*, Jiacheng Zhu*, Justin Qiu, Zachary Horvitz, Marianna Apidianaki, Kathleen McKeown, and Chris Callison-Burch (NAACL 2025)
- TinyStyler: Efficient Few-Shot Text Style Transfer with Authorship Embeddings by Zachary Horvitz, Ajay Patel, Kanishk Singh, Chris Callison-Burch, Kathleen McKeown, Zhou Yu (Findings of EMNLP 2024)
- Large Language Models Can Self-Improve At Web Agent Tasks by Ajay Patel, Markus Hofmarcher, Claudiu Leoveanu-Condrei, Marius-Constantin Dinu, Chris Callison-Burch, Sepp Hochreiter (arXiv 2024)
- DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflows by Ajay Patel, Colin Raffel, Chris Callison-Burch (ACL 2024)
- ParaGuide: Guided Diffusion Paraphrasers for Plug-and-Play Textual Style Transfer by Zachary Horovitz, Ajay Patel, Chris Callison-Burch, Zhou Yu, Kathleen McKeown (AAAI 2024)
- Learning Interpretable Style Embeddings via Prompting LLMs by Ajay Patel, Delip Rao, Chris Callison-Burch (Findings of EMNLP 2023)
- Low-Resource Authorship Style Transfer: Can Non-Famous Authors Be Imitated? by Ajay Patel, Nicholas Andrews, Chris Callison-Burch (arXiv 2022)
- Bidirectional language models are also few-shot learners by Ajay Patel, Bryan Li, Mohammad Sadegh Rasooli, Noah Constant, Colin Raffel, Chris Callison-Burch (ICLR 2023)
- Magnitude: A fast, efficient universal vector embedding utility package by Ajay Patel, Alexander Sands, Chris Callison-Burch, Marianna Apidianaki (EMNLP 2018)
Other Work
Press
- The Most Capable Open Source AI Model Yet Could Supercharge AI Agents (WIRED)
- A tiny new open-source AI model performs as well as powerful big ones (MIT Technology Review)
- AI vision, reinvented: The power of synthetic data (Penn Today)
- CoSyn: The open-source tool that's making GPT-4V-level vision AI accessible to everyone (VentureBeat)
- How fake accounts pushing inflammatory content went viral – with the help of YouTube's algorithms (CNN)
- Uncovering YouTube disinformation campaign with NLP and AI (CNN TV Coverage)
- Elizabeth Warren’s Exit Spurs Biden, Sanders to Vie for Her Voters (Wall Street Journal)
- Sanders Aims for Positive Campaign, but Allies Don’t Always Follow (Wall Street Journal)
- Plasticity wants to help chatbots seem less robotic (TechCrunch)
- Y Combinator takes machine intelligence startups to school and learns a thing or two (TechCrunch)
- Hack Siri To Control Spotify, A Nest, A Tesla, And Give Directions Through Google Maps (TechCrunch)
- College kids gave Siri new powers and now you can too (Engadget)
- This Simple Siri Hack Lets You Control Anything With Your iPhone (Gizmodo)
- College Kids Hack Siri to Unlock Teslas and Heat Up Your House (WIRED)
- Crazy Siri Hack Brings Voice Commands to Google Maps, Spotify (TIME Magazine)
- Apple Reportedly Plans To Open Siri To Third Parties (Just As Hackers Force It Open) (Forbes)
Copyright © 2025 Ajay Patel. All rights reserved.