Exploring Principled Learning and Optimization Research for Artificial Intelligence!
People
Collaborate w/ Us
We welcome faculty, research labs, and industry R&D teams seeking to co-develop theory or apply principled optimization and learning to real-world LLMs and agents.
Tracks
- Theory Co-Development: Partner with us on convergence analysis, oracle models, variance reduction, structure-aware optimization, etc. We aim to build unified principles that shape the next decade of optimization and learning theory.
- Application to LLMs & Agents: Collaborate on scaling and adapting LLMs at inference time, designing efficient training/deployment pipelines, building reliable agentic AI, etc. We validate theory through impactful benchmarks, open-source artifacts, and real-world deployments.
How to Start
- Step 1: Email a 1–2 page brief (problem, goals, constraints).
- Step 2: We arrange a short alignment call.
- Step 3: Launch collaboration (joint papers, benchmarks, or deployments).
Contact: [email protected], subject [XPLOR-Collab-2025] + organization name.
Join Us
We welcome PhD students and research interns excited to work at the intersection of optimization, learning theory, and AI applications.
Tracks
- Theory-First Research: Work on black-box optimization and decision-making (ZOO, BO, bandits, RL) and learning theory (generalization, stability, adaptivity, robustness).
- Application to LLMs & Agents: Apply principled methods to efficient training, inference-time scaling, reliable agentic AI, etc.
Requirements
- Solid background in math (probability, optimization, linear algebra).