Publications

* Equal contribution

TLAPS-Bench: Formally Proving Correctness of Real-World System Specifications with TLA+ Proof System

Q Cheng, R Tang, SMR Pial, ZC Cai, Y Su, J Clark, M Thahmid, MR Sukanya, C Newcombe, I Konnov, S Merz, X Sun, M Demirbas, Y Huang, MA Kuppe, T Xu arXiv 2026
arXiv Code

Specula: Scaling Formal Specifications for Autonomous Model Checking of System Code

Q Cheng, SMR Pial, R Tang, Y Su, E Ma, F Hackett, I Beschastnikh, Y Huang, T Xu arXiv 2026
arXiv

EduPresenta: A Conversational AI Agent for Pedagogically Sound Presentation Generation for Instructors

SK Shantanu*, SMR Pial*, S Sharmin HCII 2026
DOI

SREGym: A Live Benchmark for AI SRE Agents with High-Fidelity Failure Scenarios

J Clark*, Y Su*, SMR Pial, Y Tian, L Gniedziejko, H Jacobsen, Y Chen, T Xu NeurIPS (Evaluations and Datasets Track) 2026
arXiv

SREGym: A Live Training Ground for AI SRE Agents with High-Fidelity Failure Drills

J Clark, Y Su, SMR Pial, L Gniedziejko, T Xu ACM CAIS 2026
DOI

NuevAI: Streamlined Dataset Generator and Human Evaluator System for Development of Pedagogical Conversational Agents

S Sayeed, SMR Pial, A Iqbal CHItaly 2025
DOI