AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

Parameterizing Context: Unleashing the Power of Parameter-Efficient Fine-Tuning and In-Context Tuning for Continual Table Semantic Parsing

Yongrui Chen, Shenyu Zhang, Guilin Qi, Xinnan Guo

NeurIPS 2023 · Continual learning › Continual fine-tuning of LLMs · approach: Low-rank and model-merging adaptation · 9 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add teacher (ICT) with student distillation to prompt-tuning (ablation)PEFT (T5-base) · Spider-StreamTA: 65.7 -> 67.7Improved
add teacher (ICT) with student distillation to prompt-tuning (ablation)PEFT (T5-base) · Combined-StreamTA: 63.8 -> 66.4Improved
add student prompt-tuning (compress teacher into prompts) (ablation)C3 · Combined-StreamTA: 59.6 -> 66.4Improved
add student prompt-tuning (compress teacher into prompts) (ablation)C3 · Spider-StreamTA: 64.9 -> 67.7Improved
use in-context tuning for the teacher instead of fine-tuning (ablation)C3 · Spider-StreamTA: 65.6 -> 67.7Improved
use in-context tuning for the teacher instead of fine-tuning (ablation)C3 · Combined-StreamTA: 65.3 -> 66.4Improved
add initial task adaptation (fine-tune backbone on first task) (ablation)C3 · Spider-StreamTA: 19.4 -> 67.7Improved
add initial task adaptation (fine-tune backbone on first task) (ablation)C3 · Combined-StreamTA: 35.8 -> 66.4Improved
initialize prompts from previous task prompt instead of shared P* (ablation)C3 · Spider-StreamTA: 67.7 -> 67.0Hurt
initialize prompts from previous task prompt instead of shared P* (ablation)C3 · Combined-StreamTA: 66.4 -> 66.0Hurt
replace T5-large teacher with text-davinci-003 (no tuning)C3 (T5-large student) · Spider-StreamTA: 71.1 -> 71.3No clear effect
replace T5-large teacher with text-davinci-003 (no tuning)C3 (T5-large student) · Combined-StreamTA: 68.3 -> 67.6Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy