Hi! I am a PhD student in Computer Science at MILA and the University of Montreal, advised by Aaron Courville and Navin Goyal (Microsoft Research). Previously, I spent three wonderful years as a Research Fellow at Microsoft Research, where I worked with Navin Goyal, Adam Tauman Kalai (OpenAI), Lester Mackey and Siddhartha Gadgil (IISc Bangalore).

Currently, I am interested in building AI systems that can discover novel mathematical constructs, such as conjectures. I am also interested in the brittleness of AI safety techniques, and in the emergent capabilities and limitations of multi-agent systems. My research is generously supported by the AI for Math Fund from Renaissance Philanthropy and XTX Markets.

Research Interests: automated mathematical discovery (conjecturing), LLM reasoning and post-training, formal theorem proving, AI safety, hallucination, and multi-agent systems.

I graduated with BE in Computer Science and MSc in Economics from BITS Pilani, India in 2022. For more details about my background, refer to my CV. If you'd like to chat with me about my work or research in general, feel free to reach out. I am always open to exploring opportunities for collaboration!
Publications

Automated Conjecturing: Generation, Selection, and Evaluation
Ayush Agrawal, Vedant Shah, Navin Goyal, Ankit Anand, Aaron Courville
Preprint pdf

Shape of Thought: When Distribution Matters More than Correctness in Reasoning Tasks
Abhranil Chandra*, Ayush Agrawal*, Arian Hosseini, Sebastian Fischmeister, Rishabh Agarwal, Navin Goyal, Aaron Courville (*equal contribution)
Preprint pdf

Towards Demystifying the Optimization Landscape of RLVR Methods
Samyak Jain, Ayush Agrawal, Navin Goyal
FoRLM Workshop @ NeurIPS'25 pdf

Language Models' Factuality Depends on the Language of Inquiry
Tushar Aggarwal*, Kumar Tanmay*, Ayush Agrawal*, Kumar Ayush, Hamid Palangi, Paul Pu Liang (*equal contribution)
Preprint pdf

Learn Globally, Speak Locally: Bridging the Gaps in Multilingual Reasoning
Jaedong Hwang, Kumar Tanmay, Seok-Jin Lee, Ayush Agrawal, Hamid Palangi, Kumar Ayush, Ila Fiete, Paul Pu Liang
Preprint pdf

Do Language Models Know When They're Hallucinating References?
Ayush Agrawal, Mirac Suzgun, Lester Mackey, Adam Tauman Kalai
EACL'24, presented @ICBINB NeurIPS'23 pdf
Covered in Nature News (Jan 2025)

Aesop Gets the Fables: Towards a Hybrid Proof Search Using Ingredient Prediction by Language Models
Siddhartha Gadgil, Ayush Agrawal, Anand Rao Tadipatri, Navin Goyal
Preprint pdf

Ambient Intelligence for Securing Intelligent Vehicular Networks: Edge-Enabled Intrusion and Anomaly Detection Strategies
Tejasvi Alladi, Ayush Agrawal, Bhavya Gera, Vinay Chamola, F Richard Yu
IEEE IoT'23 pdf

Towards Automating Formalisation of Theorem Statements Using Large Language Models
Siddhartha Gadgil, Anand Rao Tadipatri, Ayush Agrawal, Ashvni Narayanan, Navin Goyal
MATH-AI NeurIPS'22 pdf

Towards a Mathematics Formalisation Assistant using Large Language Models
Ayush Agrawal, Siddhartha Gadgil, Navin Goyal, Ashvni Narayanan, Anand Tadipatri
Preprint pdf

NovelADS: A Novel Anomaly Detection System for Intra-vehicular Networks
Kushagra Agrawal, Tejasvi Alladi, Ayush Agrawal, Vinay Chamola, Abderrahim Benslimane
IEEE ITS'22 pdf

DeepADV: A Deep Neural Network Framework for Anomaly Detection in VANETs
Tejasvi Alladi, Bhavya Gera, Ayush Agrawal, Vinay Chamola, Fei Richard Yu
IEEE TVT'21 pdf

Deep Neural Networks for Securing IoT Enabled Vehicular Ad-hoc Networks
Tejasvi Alladi, Ayush Agrawal, Bhavya Gera, Vinay Chamola, Biplab Sikdar, Mohsen Guizani
IEEE ICC'21 pdf

Talks
AI for Math Fund Convening, Renaissance Philanthropy (2026)   Automated Conjecturing: Generation, Selection, and Evaluation
Microsoft Research Montreal (2025)   Shape of Thought: When Distribution Matters More than Correctness in Reasoning Tasks
TIFR-CAM (2023)   Large Language Models: Better Reasoners or Storehouses of Knowledge? | abstract

  Template: Sebastin