PinnerFormerLite: A Transformer Based Architecture for Focused Recommendations via Sequences
This paper introduces a novel training methodology for transformer-based sequence models that creates specialized user representations within a single, universal architecture. Built on PinnerFormer, the approach addresses diluted recommendations for
Authors
- Vimala Venkatesh (ORCID: https://orcid.org/0000-0003-0322-4248)
- Varun Joshi (ORCID: https://orcid.org/0000-0003-0251-4745)
- Akshay Mittal (ORCID: https://orcid.org/0009-0008-5233-9248)
Institutions
- University of the Cumberlands (US)
Publication Details
- Published
- 2025-09-03
- DOI
- https://doi.org/10.36227/techrxiv.175693612.26984277/v1
- Primary Topic
- Image Retrieval and Classification Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00