TransformerFAM: Feedback attention is working memory
Paper
•
2404.09173
•
Published
•
44
Madison May
pragmaticml
·
AI & ML interests
Scanned documents, long-term context, cheaper attention alternatives, model compression.
Organizations
upvoted
a
paper
about 1 year ago
upvoted
3
papers
almost 2 years ago
SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inference
Paper
•
2307.02628
•
Published
•
10
Focused Transformer: Contrastive Training for Context Scaling
Paper
•
2307.03170
•
Published
•
11
Lost in the Middle: How Language Models Use Long Contexts
Paper
•
2307.03172
•
Published
•
40