Reproduce “Positional Encoding to Control Output Sequence Length” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Sho Takase, Naoaki Okazaki
The authors adapt Transformer positional encodings to tell the decoder how much length remains before a requested endpoint. The length-dependent signal replaces standard decoder positional encoding. For Japanese headline generation with a 10-character target, Transformer with LDPE reports a length-error variance of 0.000, versus 0.047 for LenInit; this result is limited to tested settings.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.