Modeling Perspectives in NLP: Parameter-Efficient Perspective Conditioning for Span Extraction and Summarization
Year: 2026
Type: conference-paper
Abstract: Understanding text through multiple perspectives is essential in healthcare community question answering, where answers frequently contain heterogeneous viewpoints, including experiences, suggestions, causes, follow-up questions, and informational claims.We present a unified perspective-conditioned framework for both span identification and perspective-aware summarization on the PerAnsSumm dataset.We approach explicit perspective samples in transformer models using two parameter-efficient mechanisms: prefix-conditioned representations and perspective-aware attention layers.First, we use multi-label perspective classification to identify relevant viewpoints, which serve as conditioning signals for downstream tasks.Span identification for perspective-specific extraction is modeled as a conditioned binary sequence labeling problem.Summarization, finally, is guided by perspective-enriched encoder representations.Experiments demonstrate that explicit perspective conditioning substantially improves span detection performance while achieving competitive summarization quality.Notably, perspective-aware attention achieves strong results using only a small fraction of the trainable parameters required by full fine-tuning.Our findings highlight the importance of structured viewpoint modeling and show that explicit perspective control enables efficient and interpretable multi-perspective text understanding.
Institution: Concordia University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166511121