Current applications of generative language models (GLMs) for portfolio construction utilize the chat functionality of such models to forecast expected returns. The current literature classify outputs of these models using discrete labels that ignore the magnitude of sentiment. We show that this procedure is not optimal for cross-sectional portfolio construction. In this paper, we introduce Logit Extraction , a methodology for extracting the model’s assigned probabilities to sentiment labels, allowing for the formulation of a continuous-valued ranking variable for portfolio construction. We demonstrate that Logit Extraction significantly enhances risk-adjusted returns over previous discrete label approaches. We make available an open source implementation of Logit Extraction in the python package “TokenProbs”.
Fabozzi, Francesco A., Florescu, Ionut. Cut the chit-chat: a new framework for the application of generative language models for portfolio construction. // Annals of Operations Research. – 2026. – DOI: 10.1007/s10479-026-07418-0. – URL: http://link.springer.com/openurl/fulltext?id=doi:10.1007/s10479-026-07418-0
Fabozzi, Francesco A, Florescu, & Ionut (2026). Cut the chit-chat: a new framework for the application of generative language models for portfolio construction. Annals of Operations Research. https://doi.org/10.1007/s10479-026-07418-0
Fabozzi, et al. “Cut the chit-chat: a new framework for the application of generative language models for portfolio construction.” Annals of Operations Research, 2026. doi:10.1007/s10479-026-07418-0.
@article{Fabozzi2026,
title = {Cut the chit-chat: a new framework for the application of generative language models for portfolio construction},
author = {Fabozzi and A., Francesco and Florescu and Ionut},
journal = {Annals of Operations Research},
year = {2026},
doi = {10.1007/s10479-026-07418-0},
url = {http://link.springer.com/openurl/fulltext?id=doi:10.1007/s10479-026-07418-0}
}
In this paper, we consider cost-sharing problems defined on rooted trees. We refer to these problems as cost-tree problems and to the induced transferable utility cooperative games as irrigation games. We introduce a formal notion of...
In this paper, we focus on nonlinear programming problems with cardinality constraints that restrict the number of nonzero components of the decision vector. We start with a straightforward mixed-integer programming reformulation....
Portfolio selection is a critical issue in financial management under uncertainty. In this paper, we propose a complex approach for portfolio selection with quantile return approximation. In particular, we propose a practical workflow...
The transcendental logarithmic (translog) production function is recognised as a flexible functional form for modelling production processes. It accommodates varying elasticities of input substitution and diverse types of...
This study investigates mothership-drone route determination using two optimization approaches: a Mixed Integer Linear Programming (MILP) model based on a discrete hexagonal grid, and a Mixed Integer Nonlinear Programming (MINLP) model...