This paper discusses the challenges of annotating predicate-argument structures in Chinese light verb constructions (LVCs) within the Uniform Meaning Representation (UMR) framework, a cross-linguistic extension of Abstract Meaning Representation (AMR). A central challenge lies in reliably identifying LVCs in Chinese and determining their appropriate representation in UMR. We analyze the linguistic properties of Chinese LVCs, outline annotation difficulties for these structures and related constructions, and illustrate these issues through concrete examples. Our analysis focuses specifically on LVC.full types, where the light verb serves solely to convey morphological features and aspectual information. We exclude LVC.cause types, in which the light verb introduces an additional argument (e.g., a causal agent or source) to the event or state denoted by the nominal predicate. To address the practical challenge of semantic role assignment in Chinese LVCs, we propose a dual-path annotation approach: due to the compositional nature of these constructions, we recommend independently annotating the argument structure of the nominal predicate while systematically encoding the grammatical attributes and relations introduced by the light verb.

Abstractive summarisation is notoriously hard to evaluate since standard word-overlap-based metrics are insufficient. We introduce a new evaluation metric which is based on fact-level content weighting, i.e. relating the facts of the document to the facts of the summary. We fol- low the assumption that a good summary will reflect all relevant facts, i.e. the ones present in the ground truth (human-generated refer- ence summary). We confirm this hypothe- sis by showing that our weightings are highly correlated to human perception and compare favourably to the recent manual highlight- based metric of Hardy et al. (2019).