Dr. Colin Raffel
University of Toronto / Vector Institute
Verified Proof of Work Artifacts
2 items catalogedEach artifact below represents an authenticated research publication, production code repository, or technical architectural framework directly authored or co-created by Dr. Colin Raffel. Every entry undergoes editorial source verification.
Parameter-Efficient Transfer Learning for NLP (Foundational PEFT Adapters)
Co-authored the landmark ICML 2019 paper introducing lightweight adapter modules injected into transformer layers, proving models could learn new tasks with less than 3% added parameters.
Serial adapter bottlenecks add minor inference latency compared to re-parameterization techniques like LoRA.
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer (T5)
First author of the seminal T5 paper establishing the unified text-to-text paradigm across all NLP tasks, introducing the C4 (Colossal Clean Crawled Corpus) dataset with over 25,000 academic citations.
Encoder-decoder architecture has higher KV-cache memory requirements during generation than pure autoregressive decoders.