AI Paper Atlas

Research map / Continual learning / Domain and prompt-based

Progressive Prompts: Continual Learning for Language Models

Anastasia Razdaibiedina, Yuning Mao, Rui Hou, Madian Khabsa, Mike Lewis, Amjad Almahairi

ICLR 2023 · Continual learning › Domain and prompt-based · approach: Prompt-based continual learning · 15 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
progressively concatenate a new task prompt with frozen prompts from earlier tasksT5-large · T5 CL benchmark, average across three task ordersaccuracy: 52.7 -> 75.1Improved
progressively concatenate a new task prompt with frozen prompts from earlier tasksBERT-base · BERT CL benchmark, average across four task ordersaccuracy: 76.3 -> 77.9Improved
use progressive prompts instead of LFPT5T5-Large · 15-task CL, 20 samples/classaverage test performance: 54.3 -> 76.2Improved
use progressive prompts instead of IDBRBERT-base · 15-task CL, 20 samples/classaverage test performance: 36.8 -> 53.5Improved
transfer a learned source-task prompt to a related target task by concatenationT5-Large · IMDb → SST2, all training dataperformance: 93.6 -> 95.8Improved
transfer a learned source-task prompt to a related target task by concatenationT5-Large · MRPC → QQP, all training dataperformance: 91.1 -> 90.9Hurt
transfer a learned source-task prompt to a related target task by concatenationT5-Large · six related task pairs, 5 samples/classaverage performance: 57.5 -> 68.2Mixed
add GLUE prompts before task-specific SuperGLUE promptsT5 · SuperGLUE validationSuperGLUE score: 74.5 -> 77.2Improved
add residual MLP reparameterization to prompt tuning (ablation)BERT-base · IMDb, prompt length 5validation performance: 85.8_{0.7} -> 91.4_{0.8}Improved
use a plain MLP reparameterization without the residual connection (ablation)BERT-base prompt tuning · IMDb, prompt length 5validation performance: 85.8_{0.7} -> 83.4_{0.9}Hurt
add residual MLP reparameterization to prompt tuning (ablation)BERT-base · QQP, prompt length 5validation performance: 72.9_{0.8} -> 76.6_{1.2}Improved
add residual MLP reparameterization to prompt tuning (ablation)BERT-base · RTE, prompt length 5validation performance: 65.0_{2.3} -> 65.0_{3.5}No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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