# Exploring Universal Intrinsic Task Subspace via Prompt Tuning

arXiv: https://arxiv.org/abs/2110.07867 · HTML: https://mlsanity.com/paper/2110.07867/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- Builds on 86 of its 127 references that are in the map.
- First posted on arXiv 15 Oct 2021.

## Builds on (most cited first, 30 of 86)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering](https://mlsanity.com/paper/1809.02789/) (2018)
- [Language Models as Knowledge Bases?](https://mlsanity.com/paper/1909.01066/) (2019)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [FEVER: a large-scale dataset for Fact Extraction and VERification](https://mlsanity.com/paper/1803.05355/) (2018)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [Making Pre-trained Language Models Better Few-shot Learners](https://mlsanity.com/paper/2012.15723/) (2020)
- [Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task](https://mlsanity.com/paper/1809.08887/) (2018)
- [How Can We Know What Language Models Know?](https://mlsanity.com/paper/1911.12543/) (2019)
- [Neural Network Acceptability Judgments](https://mlsanity.com/paper/1805.12471/) (2018)
- [Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference](https://mlsanity.com/paper/2001.07676/) (2020)
- [RACE: Large-scale ReAding Comprehension Dataset From Examinations](https://mlsanity.com/paper/1704.04683/) (2017)
- [Adversarial NLI: A New Benchmark for Natural Language Understanding](https://mlsanity.com/paper/1910.14599/) (2019)

## Built on by (most cited first, 25 of 25)

- [Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey](https://mlsanity.com/paper/2403.14608/) (2024)
- [LLM-Adapters: An Adapter Family for Parameter-Efficient Fine-Tuning of Large Language Models](https://mlsanity.com/paper/2304.01933/) (2023)
- [Knowledge Editing for Large Language Models: A Survey](https://mlsanity.com/paper/2310.16218/) (2023)
- [Scaling Down to Scale Up: A Guide to Parameter-Efficient Fine-Tuning](https://mlsanity.com/paper/2303.15647/) (2023)
- [Parameter-Efficient Fine-Tuning for Medical Image Analysis: The Missed Opportunity](https://mlsanity.com/paper/2305.08252/) (2023)
- [VB-LoRA: Extreme Parameter Efficient Fine-Tuning with Vector Banks](https://mlsanity.com/paper/2405.15179/) (2024)
- [Parameter-Efficient Fine-Tuning in Large Models: A Survey of Methodologies](https://mlsanity.com/paper/2410.19878/) (2024)
- [HyperTuning: Toward Adapting Large Language Models without Back-propagation](https://mlsanity.com/paper/2211.12485/) (2022)
- [When Do Prompting and Prefix-Tuning Work? A Theory of Capabilities and Limitations](https://mlsanity.com/paper/2310.19698/) (2023)
- [Hyperbolic Fine-Tuning for Large Language Models](https://mlsanity.com/paper/2410.04010/) (2024)
- [Configurable Foundation Models: Building LLMs from a Modular Perspective](https://mlsanity.com/paper/2409.02877/) (2024)
- [Keeping Yourself is Important in Downstream Tuning Multimodal Large Language Model](https://mlsanity.com/paper/2503.04543/) (2025)
- [Task Prompt Vectors: Effective Initialization through Multi-Task Soft-Prompt Transfer](https://mlsanity.com/paper/2408.01119/) (2024)
- [Improving Complex Reasoning with Dynamic Prompt Corruption: A soft prompt Optimization Approach](https://mlsanity.com/paper/2503.13208/) (2025)
- [Parameter-Efficient Tuning with Special Token Adaptation](https://mlsanity.com/paper/2210.04382/) (2022)
- [Understanding Prompt Tuning and In-Context Learning via Meta-Learning](https://mlsanity.com/paper/2505.17010/) (2025)
- [SOEDiff: Efficient Distillation for Small Object Editing](https://mlsanity.com/paper/2405.09114/) (2024)
- [CoLA: Collaborative Low-Rank Adaptation](https://mlsanity.com/paper/2505.15471/) (2025)
- [Sparsity- and Hybridity-Inspired Visual Parameter-Efficient Fine-Tuning for Medical Diagnosis](https://mlsanity.com/paper/2405.17877/) (2024)
- [SIBO: A Simple Booster for Parameter-Efficient Fine-Tuning](https://mlsanity.com/paper/2402.11896/) (2024)
- [Learning Intrinsic Dimension via Information Bottleneck for Explainable Aspect-based Sentiment Analysis](https://mlsanity.com/paper/2402.18145/) (2024)
- [Question-Aware Knowledge Graph Prompting for Enhancing Large Language Models](https://mlsanity.com/paper/2503.23523/) (2025)
- [Memory-Efficient Transfer Learning with Fading Side Networks via Masked Dual Path Distillation](https://mlsanity.com/paper/2604.09088/) (2026)
- [Labeling supervised fine-tuning data with the scaling law](https://mlsanity.com/paper/2405.02817/) (2024)
- [Exploring Intrinsic Language-specific Subspaces in Fine-tuning Multilingual Neural Machine Translation](https://mlsanity.com/paper/2409.05224/) (2024)

## Abstract (arXiv)

Why can pre-trained language models (PLMs) learn universal representations and effectively adapt to broad NLP tasks differing a lot superficially? In this work, we empirically find evidence indicating that the adaptations of PLMs to various few-shot tasks can be reparameterized as optimizing only a few free parameters in a unified low-dimensional intrinsic task subspace, which may help us understand why PLMs could easily adapt to various NLP tasks with small-scale data. To find such a subspace and examine its universality, we propose an analysis pipeline called intrinsic prompt tuning (IPT). Specifically, we resort to the recent success of prompt tuning and decompose the soft prompts of multiple NLP tasks into the same low-dimensional nonlinear subspace, then we learn to adapt the PLM to unseen data or tasks by only tuning parameters in this subspace. In the experiments, we study diverse few-shot NLP tasks and surprisingly find that in a 250-dimensional subspace found with 100 tasks, by only tuning 250 free parameters, we can recover 97% and 83% of the full prompt tuning performance for 100 seen tasks (using different training data) and 20 unseen tasks, respectively, showing great generalization ability of the found intrinsic task subspace. Besides being an analysis tool, IPT could further help us improve the prompt tuning stability.