The Impact of Positional Encoding on Length Generalization in Transformers
Paper
•
2305.19466
•
Published
•
2
Papers - Encodings - No Positional Encodings - NoPE
-
-
Transformers Can Do Arithmetic with the Right Embeddings
Paper • 2405.17399 • Published • 52 -
Teaching Transformers Causal Reasoning through Axiomatic Training
Paper • 2407.07612 • Published • 2 -
Round and Round We Go! What makes Rotary Positional Encodings useful?
Paper • 2410.06205 • Published • 1 -
Byte Latent Transformer: Patches Scale Better Than Tokens
Paper • 2412.09871 • Published • 85