[2602.00913] Do Schwartz Higher-Order Values Help Sentence-Level Human Value Detection? A Study of Hierarchical Gating and Calibration

Computer Science > Computation and Language arXiv:2602.00913 (cs) [Submitted on 31 Jan 2026 (v1), last revised 7 Apr 2026 (this version, v3)] Title:Do Schwartz Higher-Order Values Help Sentence-Level Human Value Detection? A Study of Hierarchical Gating and Calibration Authors:Víctor Yeste, Paolo Rosso View a PDF of the paper titled Do Schwartz Higher-Order Values Help Sentence-Level Human Value Detection? A Study of Hierarchical Gating and Calibration, by V\'ictor Yeste and 1 other authors View PDF HTML (experimental) Abstract:Human value detection from single sentences is a sparse, imbalanced multi-label task. We study whether Schwartz higher-order (HO) categories help this setting on ValueEval'24 / ValuesML (74K English sentences) under a compute-frugal budget. Rather than proposing a new architecture, we compare direct supervised transformers, hard HO$\rightarrow$values pipelines, Presence$\rightarrow$HO$\rightarrow$values cascades, compact instruction-tuned large language models (LLMs), QLoRA, and low-cost upgrades such as threshold tuning and small ensembles. HO categories are learnable: the easiest bipolar pair, Growth vs. Self-Protection, reaches Macro-$F_1=0.58$. The most reliable gains come from calibration and ensembling: threshold tuning improves Social Focus vs. Personal Focus from $0.41$ to $0.57$ ($+0.16$), transformer soft voting lifts Growth from $0.286$ to $0.303$, and a Transformer+LLM hybrid reaches $0.353$ on Self-Protection. In contrast, hard hierarch...