Donald Metzler

Latest

• LAIT: Efficient Multi-Segment Encoding in Transformers with Layer-Adjustable Interaction
• UL2: Unifying Language Learning Paradigms
• Confident Adaptive Language Modeling
• Transformer Memory as a Differentiable Search Index
• Stretching Sentence-pair NLI Models to Reason over Long Documents and Clusters
• ExT5: Towards Extreme Multi-Task Scaling for Transfer Learning