2026 PESGM
Fine-Tuning LLMs for Grid Actions
A fine-tuning framework for language models that generate corrective power-grid actions while balancing cost and reliability.
Publications
Publications
Peer-reviewed publications and preprints in reliable machine learning, reinforcement learning, LLMs, optimization, and power systems applications.
7 conference papers
1 preprint
8 paper pages
Visual Index
Paper previews
Selected papers with visuals cropped from their PDFs.
2026 PESGM
Carbon-Aware Building Energy Markets
A data-driven market-participation study for building energy systems that weighs energy cost against carbon emissions.
2026 HICSS
NEO-Grid
A neural digital-twin approach for approximating distribution-grid physics while keeping optimization and control tractable.
2025 ICML
LEVIS
A verification method for constructing large exact input spaces over which neural-network behavior can be certified.
2025 ICML
SITAlign
An inference-time alignment method that treats acceptable behavior through satisficing rather than single-objective maximization.
2024 NAPS
Battery Control
A case study comparing model-free and model-based control choices for battery control problems.
2024 arXiv
TRAM
A test-time risk-adaptation method that composes source agents according to deployment-time risk specifications.
Peer Reviewed
Conference publications
7 papers
2026 PESGM
Fine-Tuning LLMs to Generate Economical and Reliable Actions for the Power Grid
Accepted to 2026 IEEE Power & Energy Society General Meeting (PESGM), 2026
2026 PESGM
Carbon-aware Market Participation for Building Energy Management Systems
Accepted to 2026 IEEE Power & Energy Society General Meeting (PESGM), 2026
2026 HICSS
NEO-Grid: A Neural Approximation Framework for Optimization and Control in Distribution Grids
Proceedings of the 59th Hawaii International Conference on System Sciences (HICSS), 2026
2025 ICML
LEVIS: Large Exact Verifiable Input Spaces for Neural Networks
Proceedings of the 42nd International Conference on Machine Learning (ICML), 2025
2025 ICML
Bounded Rationality for LLMs: Satisficing Alignment at Inference-Time
Proceedings of the 42nd International Conference on Machine Learning (ICML), 2025
2025 TPEC
BOOST: Microgrid Sizing using Ordinal Optimization
2025 Texas Power and Energy Conference (TPEC), IEEE, pp. 1-4, Feb. 2025
2024 NAPS
Should We Use Model-Free or Model-Based Control? A Case Study of Battery Control
2024 56th North American Power Symposium (NAPS), IEEE, pp. 1-5, Oct. 2024
Preprints
Working papers
2024 arXiv
TRAM: Test-Time Risk Adaptation with Mixture of Agents
arXiv, 2024