Zhuowen Tu, Yifan Xu, San Diego, Weijian Xu, Tyler Chang
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. If it links a repository, it is listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
In this paper, we detail the relationship between convolutions and self-attention in natural language tasks. We show that relative position embeddings in self-attention layers are equivalent to recently-proposed dynamic lightweight convolutions, and we consider multiple new ways of integrating convolutions into Transformer self-attention. Specifically, we propose composite attention, which unites previous relative position embedding methods under a convolutional framework. We conduct experiments by training BERT with composite attention, finding that convolutions consistently improve performance on multiple downstream tasks, replacing absolute position embeddings. To inform future work, we present results comparing lightweight convolutions, dynamic convolutions, and depthwiseseparable convolutions in language model pretraining, considering multiple injection points for convolutions in self-attention layers.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2106.05505")
get_code_for_paper("2106.05505")
have("2106.05505")
Connect an agent — have() is free.