What Does Disagreement Mean? A Semantic Identifiability Frontier for Pairwise Preference Learning

概要

Pairwise comparison data are a standard supervision signal in preference learning, but disagreement in such data can reflect stable latent heterogeneity, ephemeral stochasticity, or non-binary responses hidden by a forced-choice interface. We study which of these semantic distinctions are identifiable from the feedback channel itself. We introduce a nonparametric latent semantic model with three querywise quantities: aggregate directional preference $μ_q$, persistent disagreement $τ_q$, and non-binary mass $α_q$. Under the anonymous binary channel, we prove an exact semantic fiber theorem: $μ_q$ is point-identified, whereas $(τ_q, α_q)$ vary over an explicit feasible region compatible with the same binary law. Same-annotator repeats identify $τ_q$ through a covariance identity, abstention-enabled feedback identifies $α_q$ directly, and together they point-identify the full semantic triple. We also derive finite-sample estimators, nonasymptotic concentration bounds, and semantics-aware action-recovery guarantees, and validate the theory on theorem-matched synthetic data.

参考リンク

• Paper ↗

リンク先のライセンス・再現性・データ条件をご確認ください。自動分類や実装の目安は保証ではありません。

実装を始めるプロンプト

参考リンクを確認したうえで、次の内容をコピーして利用できます。

# 論文・記事
What Does Disagreement Mean? A Semantic Identifiability Frontier for Pairwise Preference Learning

# 概要
Pairwise comparison data are a standard supervision signal in preference learning, but disagreement in such data can reflect stable latent heterogeneity, ephemeral stochasticity, or non-binary…

# 原典
https://arxiv.org/abs/2610.20450

上記の手法を実装してください。実行前に依存関係とライセンスを確認してください。