# Avoiding Inference Heuristics in Few-shot Prompt-based Finetuning

arXiv: https://arxiv.org/abs/2109.04144 · HTML: https://mlsanity.com/paper/2109.04144/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Builds on 51 of its 63 references that are in the map.
- First posted on arXiv 9 Sep 2021.

## Builds on (most cited first, 30 of 51)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [Learning both Weights and Connections for Efficient Neural Networks](https://mlsanity.com/paper/1506.02626/) (2015)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [Don't Stop Pretraining: Adapt Language Models to Domains and Tasks](https://mlsanity.com/paper/2004.10964/) (2020)
- [Making Pre-trained Language Models Better Few-shot Learners](https://mlsanity.com/paper/2012.15723/) (2020)
- [BERT Rediscovers the Classical NLP Pipeline](https://mlsanity.com/paper/1905.05950/) (2019)
- [Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference](https://mlsanity.com/paper/2001.07676/) (2020)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Adversarial NLI: A New Benchmark for Natural Language Understanding](https://mlsanity.com/paper/1910.14599/) (2019)
- [It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners](https://mlsanity.com/paper/2009.07118/) (2020)
- [Green AI](https://mlsanity.com/paper/1907.10597/) (2019)
- [Adversarial Examples for Evaluating Reading Comprehension Systems](https://mlsanity.com/paper/1707.07328/) (2017)
- [New insights and perspectives on the natural gradient method](https://mlsanity.com/paper/1412.1193/) (2014)
- [Learning the Difference that Makes a Difference with Counterfactually-Augmented Data](https://mlsanity.com/paper/1909.12434/) (2019)
- [PAWS: Paraphrase Adversaries from Word Scrambling](https://mlsanity.com/paper/1904.01130/) (2019)
- [Fine-Tuning Pretrained Language Models: Weight Initializations, Data Orders, and Early Stopping](https://mlsanity.com/paper/2002.06305/) (2020)
- [Hypothesis Only Baselines in Natural Language Inference](https://mlsanity.com/paper/1805.01042/) (2018)
- [On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselines](https://mlsanity.com/paper/2006.04884/) (2020)
- [Get Your Vitamin C! Robust Fact Verification with Contrastive Evidence](https://mlsanity.com/paper/2103.08541/) (2021)
- [Probing Neural Network Comprehension of Natural Language Arguments](https://mlsanity.com/paper/1907.07355/) (2019)
- [Stress Test Evaluation for Natural Language Inference](https://mlsanity.com/paper/1806.00692/) (2018)
- [How Many Data Points is a Prompt Worth?](https://mlsanity.com/paper/2103.08493/) (2021)
- [Recall and Learn: Fine-tuning Deep Pretrained Language Models with Less Forgetting](https://mlsanity.com/paper/2004.12651/) (2020)

## Built on by (most cited first, 23 of 23)

- [Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution](https://mlsanity.com/paper/2202.10054/) (2022)
- [Few-shot Fine-tuning vs. In-context Learning: A Fair Comparison and Evaluation](https://mlsanity.com/paper/2305.16938/) (2023)
- [Shortcut Learning of Large Language Models in Natural Language Understanding](https://mlsanity.com/paper/2208.11857/) (2022)
- [Stretching Sentence-pair NLI Models to Reason over Long Documents and Clusters](https://mlsanity.com/paper/2204.07447/) (2022)
- [Fast Trainable Projection for Robust Fine-Tuning](https://mlsanity.com/paper/2310.19182/) (2023)
- [Towards Unified Prompt Tuning for Few-shot Text Classification](https://mlsanity.com/paper/2205.05313/) (2022)
- [Evaluating the Robustness of Discrete Prompts](https://mlsanity.com/paper/2302.05619/) (2023)
- [Debiasing Masks: A New Framework for Shortcut Mitigation in NLU](https://mlsanity.com/paper/2210.16079/) (2022)
- [Robustness of Demonstration-based Learning Under Limited Data Scenario](https://mlsanity.com/paper/2210.10693/) (2022)
- [Ada-Instruct: Adapting Instruction Generators for Complex Reasoning](https://mlsanity.com/paper/2310.04484/) (2023)
- [Feature-Level Debiased Natural Language Understanding](https://mlsanity.com/paper/2212.05421/) (2022)
- [Does Self-Rationalization Improve Robustness to Spurious Correlations?](https://mlsanity.com/paper/2210.13575/) (2022)
- [Are Sample-Efficient NLP Models More Robust?](https://mlsanity.com/paper/2210.06456/) (2022)
- [Fighting Bias with Bias: Promoting Model Robustness by Amplifying Dataset Biases](https://mlsanity.com/paper/2305.18917/) (2023)
- [Pathologies of Pre-trained Language Models in Few-shot Fine-tuning](https://mlsanity.com/paper/2204.08039/) (2022)
- [Looking at the Overlooked: An Analysis on the Word-Overlap Bias in Natural Language Inference](https://mlsanity.com/paper/2211.03862/) (2022)
- [Learning Dynamic Contextualised Word Embeddings via Template-based Temporal Adaptation](https://mlsanity.com/paper/2208.10734/) (2022)
- [Out-of-Distribution Generalization in Text Classification: Past, Present, and Future](https://mlsanity.com/paper/2305.14104/) (2023)
- [EXnet: Efficient In-context Learning for Data-less Text classification](https://mlsanity.com/paper/2305.14622/) (2023)
- [Multilingual Few-Shot Learning via Language Model Retrieval](https://mlsanity.com/paper/2306.10964/) (2023)
- [Confounder Balancing in Adversarial Domain Adaptation for Pre-Trained Large Models Fine-Tuning](https://mlsanity.com/paper/2310.16062/) (2023)
- [Self-Supervised Position Debiasing for Large Language Models](https://mlsanity.com/paper/2401.01218/) (2024)
- [Discovering the Hidden Role of Gini Index In Prompt-based Classification](https://mlsanity.com/paper/2603.15654/) (2026)

## Abstract (arXiv)

Recent prompt-based approaches allow pretrained language models to achieve strong performances on few-shot finetuning by reformulating downstream tasks as a language modeling problem. In this work, we demonstrate that, despite its advantages on low data regimes, finetuned prompt-based models for sentence pair classification tasks still suffer from a common pitfall of adopting inference heuristics based on lexical overlap, e.g., models incorrectly assuming a sentence pair is of the same meaning because they consist of the same set of words. Interestingly, we find that this particular inference heuristic is significantly less present in the zero-shot evaluation of the prompt-based model, indicating how finetuning can be destructive to useful knowledge learned during the pretraining. We then show that adding a regularization that preserves pretraining weights is effective in mitigating this destructive tendency of few-shot finetuning. Our evaluation on three datasets demonstrates promising improvements on the three corresponding challenge datasets used to diagnose the inference heuristics.