Event-Centric Human Value Understanding in News-Domain Texts: An Actor-Conditioned, Multi-Granularity Benchmark

Existing human value datasets do not directly support value understanding in factual news: many are actor-agnostic, rely on isolated utterances or synthetic scenarios, and lack explicit event structure or value direction. We present $\mathbf{\text{NEVU}}$ ($\mathbf{\text{N}}$ews $\mathbf{\text{E}}$vent-centric $\mathbf{\text{V}}$alue $\mathbf{\text{U}}$nderstanding), a benchmark for \emph{actor-conditioned}, \emph{event-centric}, and \emph{direction-aware} human value recognition in factual news. NEVU evaluates whether models can identify value cues, attribute them to the correct actor, and determine value direction from grounded evidence. Built from 2{,}865 English news articles, NEVU organizes annotations at four semantic unit levels ($\mathbf{\text{Subevent}}$, $\mathbf{\text{behavior-based}}\mathbf{\text{composite}}\mathbf{\text{event}}$, $\mathbf{\text{story-based}}\mathbf{\text{composite}}\mathbf{\text{event}}$, and $\mathbf{\text{Article}}$) and labels \mbox{(unit, actor)} pairs for fine-grained evaluation across local and composite contexts. The annotations are produced through an LLM-assisted pipeline with staged verification and targeted human auditing. Using a hierarchical value space with $\mathbf{\text{54}}$ fine-grained values and $\mathbf{\text{20}}$ coarse-grained categories, NEVU covers 45{,}793 unit–actor pairs and 168{,}061 directed value instances. We provide unified baselines for proprietary and open-source LLMs, and find that lightweight adaptation (LoRA) consistently improves open-source models, showing that although NEVU is designed primarily as a benchmark, it also supports supervised adaptation beyond prompting-only evaluation. Data availability is described in Appendix~\ref{app:data_code_availability}.