LengerichLab / context-review

Creative Commons Attribution 4.0 International
11 stars 3 forks source link

Context-Adaptive Statistical Inference: Recent Progress, Open Problems, and Opportunities for Foundation Models

HTML Manuscript PDF Manuscript GitHub Actions Status

This is an open, collaborative review paper on context-adaptive statistical methods. We look at recent progress, identify open problems, and find practical opportunities for applying these methods. We are particularly excited by the opportunities for foundation models to provide context for statistical inference.

This manuscript is created automatically from the content in content using Manubot. Please contribute! Make a PR or file an issue, and see below for more information about Manubot. Live update versions of the manuscript are available at:


Why are we writing this?

As statistical modeling evolves, we are witnessing two complementary approaches to integrating context. Traditional statistical models are being expanded to allow explicit parameter adjustments based on context, making their adaptations transparent and interpretable. Meanwhile, large foundation models are being built that how to implicitly adapt to context, enabling impressive performance in a wide range of tasks including in-context learning. This review seeks to unite these two perspectives, combining the explicit adaptability of statistical models with the powerful, implicit adjustments of foundation models. By bringing these approaches together, we aim to provide a comprehensive overview of current progress, challenges, and opportunities in context-adaptive inference.

Key perspectives driving this review:

./content/images/context_philosophies.png

Table of Contents

  1. Abstract
  2. Introduction and Definitions
  3. Theoretical Foundations and Advances in Varying-Coefficient Models
  4. Context-Adaptive Interpretations of Context-Invariant Functions
  5. Opportunities for Foundation Models
  6. Applications, Case Studies, and Evaluations
  7. Technological and Software Tools
  8. Future Trends
  9. Open Problems
  10. Conclusions

How can you contribute?

We welcome contributions from the community. Please see our contribution guidelines for more information.


Manubot

Manubot is a system for writing scholarly manuscripts via GitHub. Manubot automates citations and references, versions manuscripts using git, and enables collaborative writing via GitHub. An [overview manuscript](https://greenelab.github.io/meta-review/ "Open collaborative writing with Manubot") presents the benefits of collaborative writing with Manubot and its unique features. The [rootstock repository](https://git.io/fhQH1) is a general purpose template for creating new Manubot instances, as detailed in [`SETUP.md`](SETUP.md). See [`USAGE.md`](USAGE.md) for documentation how to write a manuscript. Please open [an issue](https://git.io/fhQHM) for questions related to Manubot usage, bug reports, or general inquiries. ### Repository directories & files The directories are as follows: + [`content`](content) contains the manuscript source, which includes markdown files as well as inputs for citations and references. See [`USAGE.md`](USAGE.md) for more information. + [`output`](output) contains the outputs (generated files) from Manubot including the resulting manuscripts. You should not edit these files manually, because they will get overwritten. + [`webpage`](webpage) is a directory meant to be rendered as a static webpage for viewing the HTML manuscript. + [`build`](build) contains commands and tools for building the manuscript. + [`ci`](ci) contains files necessary for deployment via continuous integration. ### Local execution The easiest way to run Manubot is to use [continuous integration](#continuous-integration) to rebuild the manuscript when the content changes. If you want to build a Manubot manuscript locally, install the [conda](https://conda.io) environment as described in [`build`](build). Then, you can build the manuscript on POSIX systems by running the following commands from this root directory. ```sh # Activate the manubot conda environment (assumes conda version >= 4.4) conda activate manubot # Build the manuscript, saving outputs to the output directory bash build/build.sh # At this point, the HTML & PDF outputs will have been created. The remaining # commands are for serving the webpage to view the HTML manuscript locally. # This is required to view local images in the HTML output. # Configure the webpage directory manubot webpage # You can now open the manuscript webpage/index.html in a web browser. # Alternatively, open a local webserver at http://localhost:8000/ with the # following commands. cd webpage python -m http.server ``` Sometimes it's helpful to monitor the content directory and automatically rebuild the manuscript when a change is detected. The following command, while running, will trigger both the `build.sh` script and `manubot webpage` command upon content changes: ```sh bash build/autobuild.sh ``` ### Continuous Integration Whenever a pull request is opened, CI (continuous integration) will test whether the changes break the build process to generate a formatted manuscript. The build process aims to detect common errors, such as invalid citations. If your pull request build fails, see the CI logs for the cause of failure and revise your pull request accordingly. When a commit to the `main` branch occurs (for example, when a pull request is merged), CI builds the manuscript and writes the results to the [`gh-pages`](https://github.com/LengerichLab/context-review/tree/gh-pages) and [`output`](https://github.com/LengerichLab/context-review/tree/output) branches. The `gh-pages` branch uses [GitHub Pages](https://pages.github.com/) to host the following URLs: + **HTML manuscript** at https://LengerichLab.github.io/context-review/ + **PDF manuscript** at https://LengerichLab.github.io/context-review/manuscript.pdf For continuous integration configuration details, see [`.github/workflows/manubot.yaml`](.github/workflows/manubot.yaml).

License

[![License: CC BY 4.0](https://img.shields.io/badge/License%20All-CC%20BY%204.0-lightgrey.svg)](http://creativecommons.org/licenses/by/4.0/) [![License: CC0 1.0](https://img.shields.io/badge/License%20Parts-CC0%201.0-lightgrey.svg)](https://creativecommons.org/publicdomain/zero/1.0/) Except when noted otherwise, the entirety of this repository is licensed under a CC BY 4.0 License ([`LICENSE.md`](LICENSE.md)), which allows reuse with attribution. Please attribute by linking to https://github.com/LengerichLab/context-review. Since CC BY is not ideal for code and data, certain repository components are also released under the CC0 1.0 public domain dedication ([`LICENSE-CC0.md`](LICENSE-CC0.md)). All files matched by the following glob patterns are dual licensed under CC BY 4.0 and CC0 1.0: + `*.sh` + `*.py` + `*.yml` / `*.yaml` + `*.json` + `*.bib` + `*.tsv` + `.gitignore` All other files are only available under CC BY 4.0, including: + `*.md` + `*.html` + `*.pdf` + `*.docx` Please open [an issue](https://github.com/LengerichLab/context-review/issues) for any question related to licensing.