# Enhancing Robustness of Foundation Model Representations under Provenance-related Distribution Shifts

arXiv: https://arxiv.org/abs/2312.05435 · HTML: https://mlsanity.com/paper/2312.05435/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 31 references that are in the map.
- First posted on arXiv 9 Dec 2023.

## Builds on (most cited first, 18 of 18)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [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)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [BLOOM: A 176B-Parameter Open-Access Multilingual Language Model](https://mlsanity.com/paper/2211.05100/) (2022)
- [A Survey on Evaluation of Large Language Models](https://mlsanity.com/paper/2307.03109/) (2023)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [Measuring Robustness to Natural Distribution Shifts in Image Classification](https://mlsanity.com/paper/2007.00644/) (2020)
- [Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classification](https://mlsanity.com/paper/1903.04561/) (2019)
- [Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection](https://mlsanity.com/paper/2012.15761/) (2020)
- [Hate Speech Dataset from a White Supremacy Forum](https://mlsanity.com/paper/1809.04444/) (2018)
- [Directions in Abusive Language Training Data: Garbage In, Garbage Out](https://mlsanity.com/paper/2004.01670/) (2020)
- [AlexaTM 20B: Few-Shot Learning Using a Large-Scale Multilingual Seq2Seq Model](https://mlsanity.com/paper/2208.01448/) (2022)
- [Annotating Social Determinants of Health Using Active Learning, and Characterizing Determinants Using Neural Event Extraction](https://mlsanity.com/paper/2004.05438/) (2020)
- [The 2022 n2c2/UW Shared Task on Extracting Social Determinants of Health](https://mlsanity.com/paper/2301.05571/) (2023)
- [Backdoor Adjustment of Confounding by Provenance for Robust Text Classification of Multi-institutional Clinical Notes](https://mlsanity.com/paper/2310.02451/) (2023)

## Built on by (most cited first, 1 of 1)

- [DeconDTN-Toolkit: A Library for Evaluation and Enhancement of Robustness to Provenance Shift](https://mlsanity.com/paper/2605.11237/) (2026)

## Abstract (arXiv)

Foundation models are a current focus of attention in both industry and academia. While they have shown their capabilities in a variety of tasks, in-depth research is required to determine their robustness to distribution shift when used as a basis for supervised machine learning. This is especially important in the context of clinical data, with particular limitations related to data accessibility, lack of pretraining materials, and limited availability of high-quality annotations. In this work, we examine the stability of models based on representations from foundation models under distribution shift. We focus on confounding by provenance, a form of distribution shift that emerges in the context of multi-institutional datasets when there are differences in source-specific language use and class distributions. Using a sampling strategy that synthetically induces varying degrees of distribution shift, we evaluate the extent to which representations from foundation models result in predictions that are inherently robust to confounding by provenance. Additionally, we examine the effectiveness of a straightforward confounding adjustment method inspired by Pearl's conception of backdoor adjustment. Results indicate that while foundation models do show some out-of-the-box robustness to confounding-by-provenance related distribution shifts, this can be considerably improved through adjustment. These findings suggest a need for deliberate adjustment of predictive models using representations from foundation models in the context of source-specific distributional differences.