Find a story
Search Spins
Search titles, summaries, and missing voices across published articles — press releases, announcements, and media coverage.
0 results for “transformer models”
Transformer Models for Text Summarization: A Comparative Study of BART, BERT, and RoBERTa
A new arXiv preprint presents a comparative review of transformer-based models—BART, BERT, and RoBERTa—for text summarization tasks, analyzing architectures, pretraining strategies, and suitability for extractive versus abstractive approaches.
Aug 21, 2026
Data Fusion and Contrastive Alignment for Unconstrained IR Molecular Structure Elucidation
A new AI method improves automated molecular structure prediction from infrared spectroscopy data by replacing additive aggregation with non-additive operators and adding contrastive alignment, achieving >10pp Top-K accuracy gain over IR-only baselines.
Jul 30, 2026
Multi-level context Modeling for consistent expert selection in Mixture-of-Experts
Researchers propose MCF-MOE, a new Mixture-of-Experts routing framework that improves expert selection consistency by fusing multi-level contextual signals across Transformer layers, addressing instability in existing MoE models.
Jul 21, 2026