research

My research explores how intelligent systems can plan, verify their decisions, and adapt over long horizons.

interests

• Large language models for planning and decision-making
• Agentic systems and self-evolving agents
• Long-horizon consistency and verification-guided repair
• Constrained optimization and network scheduling
• Efficient large language models

current questions

• How can language models satisfy explicit constraints while solving complex planning problems?
• How can an agent detect and repair inconsistencies before they compound over a long horizon?
• How can structured search, verification, and optimization make agent behavior more reliable?

selected work

Hierarchical Reinforcement Learning with Topology-Aware Exploration

An end-to-end framework for constructing multiple network paths and allocating traffic under real-time states and hard constraints.

AAAI 2026 · Project summary · Paper