REVIEW 2 cited by
RealFormer: Transformer Likes Residual Attention
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Transformer is the backbone of modern NLP models. In this paper, we propose RealFormer, a simple and generic technique to create Residual Attention Layer Transformer networks that significantly outperform the canonical Transformer and its variants (BERT, ETC, etc.) on a wide spectrum of tasks including Masked Language Modeling, GLUE, SQuAD, Neural Machine Translation, WikiHop, HotpotQA, Natural Questions, and OpenKP. We also observe empirically that RealFormer stabilizes training and leads to models with sparser attention. Source code and pre-trained checkpoints for RealFormer can be found at https://github.com/google-research/google-research/tree/master/realformer.
Forward citations
Cited by 2 Pith papers
-
Why Post-Norm Transformers Collapse: Attention Amplification and Gradient Repair Failure
Token similarity in Post-Norm decoders is amplified by causal attention at initialization, and RMSNorm backward contraction prevents gradients from repairing the resulting collapse.
-
EdgeGFL: Rethinking Edge Information in Graph Feature Preference Learning
EdgeGFL multiplies node messages by learned edge-type vectors to implement per-dimension feature preference in heterogeneous graph neural networks, reporting small gains over prior GNN baselines.
Discussion (0). Continue with ORCID to comment.