3 papers
q-fin.GN2026
Interpretable Systematic Risk around the Clock
Songrun He
In this paper, I present the first comprehensive, around-the-clock analysis of systematic jump risk by combining high-frequency market data with contemporaneous news narratives ide…
cs.LG2025
Instruction Tuning Chronologically Consistent Language Models
Songrun He, Linying Lv, Asaf Manela +1
We introduce a family of chronologically consistent, instruction-tuned large language models to eliminate lookahead bias. Each model is trained only on data available before a clea…
q-fin.GN2025
Chronologically Consistent Large Language Models
Songrun He, Linying Lv, Asaf Manela +1
Large language models are increasingly used in social sciences, but their training data can introduce lookahead bias and training leakage. A good chronologically consistent languag…