Work & Research

Working on the theoretical analysis of randomised search heuristics and genetic programming, supervised by Prof. Pietro Simone Oliveto in collaboration with Mingxuan Yin. The current direction focuses on runtime analysis and generalisation bounds, with a submission target including Algorithmica.

My contribution includes deriving runtime bounds, formalising proof arguments, and exploring how algorithmic parameters affect generalisation in evolutionary search.

Keywords / skills involved: probabilistic methods, drift theorems, runtime analysis, evolutionary algorithms, mathematical proofs.

Second author of a small-target recognition project currently under review at NeurIPS 2026. The work focuses on improving detection and representation learning for low-resolution or weakly salient objects under challenging visual conditions.

My contribution involves model design choices, training pipeline engineering, and extensive experiments on benchmark datasets to validate the proposed approach.

Keywords / skills involved: object detection, low-resolution vision, attention mechanisms, representation learning, PyTorch.

Co-first author of a context-engineering work currently under review at NeurIPS 2026. The study explores how structured context design affects reasoning and decision-making in large language models, with an emphasis on theoretical understanding and reproducible empirical protocols.

My contribution focused on the initial problem formulation, core idea validation, and large-scale empirical design. I also led the theoretical framing of how context structure influences model behavior.

Keywords / skills involved: prompting theory, empirical protocol design, LLM evaluation, reproducible experiments.

Co-first author of a safety-focused reinforcement-learning project targeting AAAI 2027. The work addresses safety in RL through a risk-sensitive framework inspired by quantitative finance, using tail-risk estimation (e.g., CVaR-style measures) to bound rare but catastrophic failures without relying on dense reward engineering.

My contribution centers on algorithmic design, convergence analysis, and the theoretical justification of safety guarantees under tail-risk constraints.

Keywords / skills involved: reinforcement learning theory, safety constraints, tail risk estimation, CVaR, risk-sensitive RL, convergence analysis, regret bounds.