Beyond Independent Labels: Schwartz-Geometry Decoding for Human Value Detection

2026-07-06

AI Summary

This work addresses a key limitation in existing value detection methods, which treat Schwartz’s 19 human values as independent labels and ignore their underlying circular motivational structure. Building upon DeBERTa-v3-base, the authors propose a theory-aware energy-based decoder and introduce geometry-informed training objectives that incorporate value compatibility and opposition as soft biases during prediction. The proposed approach maintains competitive performance in standard Macro-F1 and Micro-F1 metrics while significantly improving scores on a newly introduced theoretical consistency metric. Crucially, this improvement is observed only when using the authentic Schwartz circumplex ordering; it vanishes under random permutations or co-occurrence-based graphs, demonstrating the model’s specific capacity to capture the theoretically grounded structural relationships among values.

Abstract

Human value detection is commonly formulated as sentence-level multi-label classification over the 19 refined Schwartz values, typically predicted as independent labels. Schwartz theory, however, describes them as a circular motivational continuum, in which adjacent values are compatible and opposing values are in tension. We ask whether this structure can be operationalized as an explicit output-space geometry and used as a soft bias rather than a hard constraint. On a DeBERTa-v3-base classifier, we compare two ways of injecting it: training-time geometry-aware objectives and a post-hoc Schwartz-aware energy decoder that scores whole label sets jointly. Across five seeds, training-time geometry gives only limited gains-no larger for the true continuum than for a random ordering-whereas the decoder makes label sets more coherent with the continuum-on theory-aware coherence metrics we introduce-at no cost to Macro-F1 or Micro-F1 (held fixed by its selection rule). The gain is specific to the true Schwartz ordering: it does not appear for a random permutation or an empirical co-occurrence graph through the identical decoder. A bounded Qwen2.5-72B-Instruct diagnostic shows that supplying the continuum at inference shifts behavior but does not match supervised structured prediction. Theory-aware decoding thus offers a lightweight, controllable way to make value detection faithful to its label space.

Problem

• human value detection
• Schwartz theory
• multi-label classification
• label space structure
• motivational continuum

Innovation

• Schwartz values
• structured prediction
• energy-based decoding
• output-space geometry
• value detection