SAM-NER: Semantic Archetype Mediation for Zero-Shot Named Entity Recognition

TL;DR

SAM-NER introduces a three-stage framework that stabilizes zero-shot NER across domains by mediating entity recognition through universal semantic archetypes, improving cross-domain transfer performance.

Contribution

The paper proposes Semantic Archetype Mediation, a novel three-stage approach for zero-shot NER that enhances domain-invariant transfer and outperforms existing baselines.

Findings

SAM-NER outperforms prior ZS-NER methods on CrossNER benchmark.

The framework effectively stabilizes cross-domain entity recognition.

Open-source implementation available at the provided GitHub URL.

Abstract

Zero-shot Named Entity Recognition (ZS-NER) remains brittle under domain and schema shifts, where unseen label definitions often misalign with a large language model's (LLM's) intrinsic semantic organization. As a result, directly mapping entity mentions to fine-grained target labels can induce systematic semantic drift, especially when target schemas are novel or semantically overlapping. We propose \textbf{SAM-NER}, a three-stage framework based on \emph{Semantic Archetype Mediation} that stabilizes cross-domain transfer through an intermediate, domain-invariant archetype space. SAM-NER: (i) performs \emph{Entity Discovery} via cooperative extraction and consensus-based denoising to obtain high-coverage, high-fidelity entity spans; (ii) conducts \emph{Abstract Mediation} by projecting entities into a compact set of universal semantic archetypes distilled from high-level ontological abstractions; and (iii) applies \emph{Semantic Calibration} to resolve archetype-level predictions into target-domain types through constrained, definition-aligned inference with a frozen LLM. Experiments on the CrossNER benchmark show that SAM-NER consistently outperforms strong prior ZS-NER baselines in cross-domain settings. Our implementation will be open-sourced at https://github.com/DMIRLAB-Group/SAM-NER.