{"id":24283686,"url":"https://github.com/lengerichlab/context-review","last_synced_at":"2025-03-05T15:48:40.770Z","repository":{"id":253255317,"uuid":"842951661","full_name":"LengerichLab/context-review","owner":"LengerichLab","description":null,"archived":false,"fork":false,"pushed_at":"2024-10-23T20:02:11.000Z","size":1260,"stargazers_count":11,"open_issues_count":7,"forks_count":3,"subscribers_count":3,"default_branch":"main","last_synced_at":"2024-10-24T06:45:54.423Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"HTML","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"cc-by-4.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/LengerichLab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE-CC0.md","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-08-15T13:12:24.000Z","updated_at":"2024-10-23T19:59:23.000Z","dependencies_parsed_at":"2024-10-26T02:09:26.516Z","dependency_job_id":null,"html_url":"https://github.com/LengerichLab/context-review","commit_stats":null,"previous_names":["lengerichlab/context-review"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/LengerichLab%2Fcontext-review","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/LengerichLab%2Fcontext-review/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/LengerichLab%2Fcontext-review/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/LengerichLab%2Fcontext-review/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/LengerichLab","download_url":"https://codeload.github.com/LengerichLab/context-review/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":242058106,"owners_count":20065062,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-01-16T04:19:08.860Z","updated_at":"2025-03-05T15:48:40.743Z","avatar_url":"https://github.com/LengerichLab.png","language":"HTML","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Context-Adaptive Statistical Inference: Recent Progress, Open Problems, and Opportunities for Foundation Models\n\n[![HTML Manuscript](https://img.shields.io/badge/manuscript-HTML-blue.svg)](https://LengerichLab.github.io/context-review/)\n[![PDF Manuscript](https://img.shields.io/badge/manuscript-PDF-blue.svg)](https://LengerichLab.github.io/context-review/manuscript.pdf)\n[![GitHub Actions Status](https://github.com/LengerichLab/context-review/workflows/Manubot/badge.svg)](https://github.com/LengerichLab/context-review/actions)\n\n\nThis 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.\n\n\nThis manuscript is created automatically from the content in [content](https://github.com/LengerichLab/context-review/tree/main/content) using Manubot. Please [contribute](CONTRIBUTING.md)! Make a PR or file an issue, and see below for more information about Manubot. Live update versions of the manuscript are available at:\n\n+ **HTML manuscript** at https://LengerichLab.github.io/context-review/\n+ **PDF manuscript** at https://LengerichLab.github.io/context-review/manuscript.pdf\n\n---\n\n## Why are we writing this?\nAs statistical modeling evolves, we are witnessing two complementary approaches to integrating context. \nTraditional statistical models are being expanded to allow explicit parameter adjustments based on context, making their\nadaptations transparent and interpretable. \nMeanwhile, large foundation models are being built that how to implicitly adapt to context, enabling impressive \nperformance in a wide range of tasks including in-context learning. \nThis review seeks to unite these two perspectives, combining the explicit adaptability of statistical models with the \npowerful, implicit adjustments of foundation models. \nBy bringing these approaches together, we aim to provide a comprehensive overview of current progress, challenges, \nand opportunities in context-adaptive inference.\n\n\n### Key perspectives driving this review:\n\n- **Complex models contain multitudes of simpler, context-specific models** – Every complex model can be understood as a combination of many smaller models, each tailored to specific contexts.\n- **Explicit vs. implicit adaptation** – Statistical models explicitly adjust parameters based on context, while foundation models implicitly adapt to context. Combining these approaches offers new opportunities for robust and adaptive inference.\n- **Context reshapes inference** – Context-awareness enhances both statistical and foundation models, improving personalization and accuracy in predictions.\n- **In-context learning as a model of implicit adaptation** – Foundation models excel at tasks like in-context learning, showing how implicit adaptation can inform broader context-adaptive modeling efforts.\n- **Foundation models as context providers** – These models offer flexible, scalable ways to incorporate context, enhancing traditional methods with richer context integration.\n- **Challenging traditional assumptions** – Context-adaptive methods move beyond the assumption of homogeneity in data, enabling models to handle heterogeneous datasets more effectively.\n- **Personalization through adaptation** – Uniting explicit and implicit context-adaptive models provides a path to more nuanced, personalized predictions that reflect the complexities of real-world data.\n\n![./content/images/context_philosophies.png](./content/images/context_philosophies.png)\n\n## Table of Contents\n1. [Abstract](./content/01.abstract.md)\n2. [Introduction and Definitions](./content/02.introduction.md)\n3. [Theoretical Foundations and Advances in Varying-Coefficient Models](./content/03.vc.md) \n4. [Context-Adaptive Interpretations of Context-Invariant Functions](./content/04.interpretations.md) \n5. [Opportunities for Foundation Models](./content/05.foundation.md) \n6. [Applications, Case Studies, and Evaluations](./content/06.applications.md) \n7. [Technological and Software Tools](./content/07.tools.md) \n8. [Future Trends](./content/08.future.md) \n9. [Open Problems](./content/09.problems.md) \n10. [Conclusions](./content/10.conclusions.md)\n\n## How can you contribute?\nWe welcome contributions from the community. Please see our [contribution guidelines](CONTRIBUTING.md) for more information.\n\n--- \n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003ch2\u003eManubot\u003c/h2\u003e\u003c/summary\u003e\n  \n\u003c!-- usage note: do not edit this section --\u003e\n\nManubot is a system for writing scholarly manuscripts via GitHub.\nManubot automates citations and references, versions manuscripts using git, and enables collaborative writing via GitHub.\nAn [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.\nThe [rootstock repository](https://git.io/fhQH1) is a general purpose template for creating new Manubot instances, as detailed in [`SETUP.md`](SETUP.md).\nSee [`USAGE.md`](USAGE.md) for documentation how to write a manuscript.\n\nPlease open [an issue](https://git.io/fhQHM) for questions related to Manubot usage, bug reports, or general inquiries.\n\n### Repository directories \u0026 files\n\nThe directories are as follows:\n\n+ [`content`](content) contains the manuscript source, which includes markdown files as well as inputs for citations and references.\n  See [`USAGE.md`](USAGE.md) for more information.\n+ [`output`](output) contains the outputs (generated files) from Manubot including the resulting manuscripts.\n  You should not edit these files manually, because they will get overwritten.\n+ [`webpage`](webpage) is a directory meant to be rendered as a static webpage for viewing the HTML manuscript.\n+ [`build`](build) contains commands and tools for building the manuscript.\n+ [`ci`](ci) contains files necessary for deployment via continuous integration.\n\n### Local execution\n\nThe easiest way to run Manubot is to use [continuous integration](#continuous-integration) to rebuild the manuscript when the content changes.\nIf you want to build a Manubot manuscript locally, install the [conda](https://conda.io) environment as described in [`build`](build).\nThen, you can build the manuscript on POSIX systems by running the following commands from this root directory.\n\n```sh\n# Activate the manubot conda environment (assumes conda version \u003e= 4.4)\nconda activate manubot\n\n# Build the manuscript, saving outputs to the output directory\nbash build/build.sh\n\n# At this point, the HTML \u0026 PDF outputs will have been created. The remaining\n# commands are for serving the webpage to view the HTML manuscript locally.\n# This is required to view local images in the HTML output.\n\n# Configure the webpage directory\nmanubot webpage\n\n# You can now open the manuscript webpage/index.html in a web browser.\n# Alternatively, open a local webserver at http://localhost:8000/ with the\n# following commands.\ncd webpage\npython -m http.server\n```\n\nSometimes it's helpful to monitor the content directory and automatically rebuild the manuscript when a change is detected.\nThe following command, while running, will trigger both the `build.sh` script and `manubot webpage` command upon content changes:\n\n```sh\nbash build/autobuild.sh\n```\n\n### Continuous Integration\n\nWhenever a pull request is opened, CI (continuous integration) will test whether the changes break the build process to generate a formatted manuscript.\nThe build process aims to detect common errors, such as invalid citations.\nIf your pull request build fails, see the CI logs for the cause of failure and revise your pull request accordingly.\n\nWhen 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.\nThe `gh-pages` branch uses [GitHub Pages](https://pages.github.com/) to host the following URLs:\n\n+ **HTML manuscript** at https://LengerichLab.github.io/context-review/\n+ **PDF manuscript** at https://LengerichLab.github.io/context-review/manuscript.pdf\n\nFor continuous integration configuration details, see [`.github/workflows/manubot.yaml`](.github/workflows/manubot.yaml).\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003ch2\u003eLicense\u003c/h2\u003e\u003c/summary\u003e\n\n\u003c!--\nusage note: edit this section to change the license of your manuscript or source code changes to this repository.\nWe encourage users to openly license their manuscripts, which is the default as specified below.\n--\u003e\n\n[![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/)\n[![License: CC0 1.0](https://img.shields.io/badge/License%20Parts-CC0%201.0-lightgrey.svg)](https://creativecommons.org/publicdomain/zero/1.0/)\n\nExcept 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.\nPlease attribute by linking to https://github.com/LengerichLab/context-review.\n\nSince 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)).\nAll files matched by the following glob patterns are dual licensed under CC BY 4.0 and CC0 1.0:\n\n+ `*.sh`\n+ `*.py`\n+ `*.yml` / `*.yaml`\n+ `*.json`\n+ `*.bib`\n+ `*.tsv`\n+ `.gitignore`\n\nAll other files are only available under CC BY 4.0, including:\n\n+ `*.md`\n+ `*.html`\n+ `*.pdf`\n+ `*.docx`\n\nPlease open [an issue](https://github.com/LengerichLab/context-review/issues) for any question related to licensing.\n\n\u003c/details\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flengerichlab%2Fcontext-review","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flengerichlab%2Fcontext-review","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flengerichlab%2Fcontext-review/lists"}