Noah Goodman, Robert Hawkins, Julia White
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Language differs dramatically from context to context. To some degree, large language models like GPT-3 account for such variation by conditioning on strings of initial input text, or prompts. However, prompting can be ineffective when contexts are sparse, out-of-sample, or extra-textual. In this paper, we introduce the mixed-effects transformer (MET), a novel approach for learning hierarchically-structured prefixes-lightweight modules prepended to an input sequence-to account for structured variation in language use. Specifically, we show how the popular class of mixedeffects regression models may be extended to transformer-based architectures using a regularized prefix-tuning procedure with dropout. We evaluate this approach on several domainadaptation benchmarks, finding that it learns contextual variation from minimal data while generalizing well to unseen contexts.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2205.01749")
get_code_for_paper("2205.01749")
have("2205.01749")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2205.01749.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2205.01749)
Connect an agent — have() is free.