Prompt-Augmented Linear Probing: Scaling beyond the Limit of Few-Shot In-Context Learners


  • Hyunsoo Cho Seoul National University
  • Hyuhng Joon Kim Seoul National University
  • Junyeob Kim Seoul National University
  • Sang-Woo Lee Naver Cloud KAIST
  • Sang-goo Lee Seoul National University
  • Kang Min Yoo Seoul National University Naver Cloud
  • Taeuk Kim Hanyang University



SNLP: Language Models, SNLP: Text Classification


Through in-context learning (ICL), large-scale language models are effective few-shot learners without additional model fine-tuning. However, the ICL performance does not scale well with the number of available training sample as it is limited by the inherent input length constraint of the underlying language model. Meanwhile, many studies have revealed that language models are also powerful feature extractors, allowing them to be utilized in a black-box manner and enabling the linear probing paradigm, where lightweight discriminators are trained on top of the pre-extracted input representations. This paper proposes prompt-augmented linear probing (PALP), a hybrid of linear probing and ICL, which leverages the best of both worlds. PALP inherits the scalability of linear probing and the capability of enforcing language models to derive more meaningful representations via tailoring input into a more conceivable form. Throughout in-depth investigations on various datasets, we verified that PALP significantly closes the gap between ICL in the data-hungry scenario and fine-tuning in the data-abundant scenario with little training overhead, potentially making PALP a strong alternative in a black-box scenario.




How to Cite

Cho, H., Kim, H. J., Kim, J., Lee, S.-W., Lee, S.- goo, Yoo, K. M., & Kim, T. (2023). Prompt-Augmented Linear Probing: Scaling beyond the Limit of Few-Shot In-Context Learners. Proceedings of the AAAI Conference on Artificial Intelligence, 37(11), 12709-12718.



AAAI Technical Track on Speech & Natural Language Processing