Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Small Models are Valuable Plug-ins for Large Language Models

About

Large language models (LLMs) such as GPT-3 and GPT-4 are powerful but their weights are often publicly unavailable and their immense sizes make the models difficult to be tuned with common hardware. As a result, effectively tuning these models with large-scale supervised data can be challenging. As an alternative, In-Context Learning (ICL) can only use a small number of supervised examples due to context length limits. In this paper, we propose Super In-Context Learning (SuperICL) which allows black-box LLMs to work with locally fine-tuned smaller models, resulting in superior performance on supervised tasks. Our experiments demonstrate that SuperICL can improve performance beyond state-of-the-art fine-tuned models while addressing the instability problem of in-context learning. Furthermore, SuperICL can enhance the capabilities of smaller models, such as multilinguality and interpretability.

Canwen Xu, Yichong Xu, Shuohang Wang, Yang Liu, Chenguang Zhu, Julian McAuley• 2023

Related benchmarks

TaskDatasetResultRank
Natural Language UnderstandingGLUE (dev)
SST-2 (Acc)96.79
529
Natural Language InferenceXNLI (test)
Average Accuracy74.11
167
Topic ClassificationAG News (test)
Accuracy88.79
116
Ontology ClassificationDBPedia (test)
Accuracy97.63
53
Readback Anomaly DetectionAPCP
Accuracy84.12
30
Natural Language InferenceANLI (test)
Overall Score47.44
28
Intent ClassificationBanking (BANK) (test)
Accuracy73.25
11
Medical Diagnosis ClassificationMedical Abstract (MA) (test)
Accuracy63.75
11
Topic ClassificationTREC (test)
Accuracy81.6
11
Content Type ClassificationRCT (test)
Accuracy67.82
11
Showing 10 of 16 rows

Other info

Code

Follow for update