{"id":13690010,"url":"https://github.com/samacqua/LARC","last_synced_at":"2025-05-02T06:31:53.218Z","repository":{"id":49126097,"uuid":"334748446","full_name":"samacqua/LARC","owner":"samacqua","description":"Language-annotated Abstraction and Reasoning Corpus","archived":false,"fork":false,"pushed_at":"2023-05-20T01:12:42.000Z","size":9727,"stargazers_count":78,"open_issues_count":1,"forks_count":10,"subscribers_count":4,"default_branch":"main","last_synced_at":"2024-11-12T15:43:18.357Z","etag":null,"topics":["arc"],"latest_commit_sha":null,"homepage":"https://samacquaviva.com/LARC/explore/","language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/samacqua.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2021-01-31T20:19:30.000Z","updated_at":"2024-10-22T16:35:08.000Z","dependencies_parsed_at":"2023-01-20T09:36:09.412Z","dependency_job_id":"429751e3-8112-444d-a18e-d473fe1d3c14","html_url":"https://github.com/samacqua/LARC","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/samacqua%2FLARC","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/samacqua%2FLARC/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/samacqua%2FLARC/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/samacqua%2FLARC/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/samacqua","download_url":"https://codeload.github.com/samacqua/LARC/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":251998615,"owners_count":21678005,"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":["arc"],"created_at":"2024-08-02T16:00:41.643Z","updated_at":"2025-05-02T06:31:49.928Z","avatar_url":"https://github.com/samacqua.png","language":"JavaScript","funding_links":[],"categories":["Papers"],"sub_categories":["ARC-based"],"readme":"# Language-complete Abstraction and Reasoning Corpus (LARC)\n\nThis repository contains the LARC dataset and supporting assets\n\nThe entire dataset can be browsed at [the explorer interface](https://samacqua.github.io/LARC/explore) take a look!\n\nHere's a [quick 5 minutes slideslive video explaining this work](https://recorder-v3.slideslive.com/#/share?share=75868\u0026s=71a5a42e-dceb-4055-9ae7-9bfccd4ffc1f)\n\n*\"How can we build intelligent systems that achieve human-level performance on challenging and structured domains (like ARC), with or without additional human guidance? We posit the answer may be found in studying natural programs - instructions humans give to each other to communicate how to solve a task. Like a computer program, these instructions can be reliably \"executed\" by others to produce intended outputs.\"*\n\nA comprehensive view of this dataset and its goals can be found in [Communicating Natural Programs to Humans and Machines (Neurips Dataset and Benchmark, 2022)](https://arxiv.org/abs/2106.07824)\n\nLARC is curated from a communication game, where \none participant, the *describer* solves an [ARC task](https://github.com/fchollet/ARC) and describes the solution to a different participant, \nthe *builder*, who must solve the task on the new input using the description alone. \nThe successful descriptions are \"language-complete\" in a sense that it fully captures the underlying ARC task in the absence of the original input-output examples.\n\n\u003cp align=\"center\"\u003e\n\u003cimg src=\"https://raw.githubusercontent.com/samacqua/LARC/main/assets/collection.jpg\" alt=\"drawing\" width=\"75%\"/\u003e\n\u003c/p\u003e\n\n\nCitation\n```\n@article{acquaviva2021communicating,\n  title={Communicating Natural Programs to Humans and Machines},\n  author={Acquaviva, Samuel and Pu, Yewen and Kryven, Marta and Wong, Catherine and Ecanow, Gabrielle E and Nye, Maxwell and Sechopoulos, Theodoros and Tessler, Michael Henry and Tenenbaum, Joshua B},\n  journal={arXiv preprint arXiv:2106.07824},\n  year={2021}\n}\n```\n\nThe original ARC data can be found here [The Abstraction and Reasoning Corpus](https://github.com/fchollet/ARC)\n\n## Contents\n- `dataset` contains the language-complete ARC tasks and successful natural program phrase annotations\n- `explorer` contains the explorer code that allows for easy browsing of the annotated tasks\n- `collection` contains the source code used to curate the dataset\n- `bandit` contains the formulation and environment for bandit algorithm used for collection\n\nlanguage-guided program synthesis code can be found [here](https://github.com/theosech/ec/tree/language-guided_program_synthesis_for_larc)\n\nGPT4 (vision only) program induction results can be found [here](https://github.com/evanthebouncy/larc_gpt4)\n\n## License\n\nThe [dataset](https://github.com/samacqua/LARC/tree/main/dataset) is licensed under the [Creative Commons Attribution 4.0 International License](http://creativecommons.org/licenses/by/4.0/)\n\nAll supporting code follows the [MIT License](https://opensource.org/licenses/MIT)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsamacqua%2FLARC","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsamacqua%2FLARC","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsamacqua%2FLARC/lists"}