{"id":21006789,"url":"https://github.com/grig-guz/tree-content-structuring","last_synced_at":"2026-04-25T15:34:53.250Z","repository":{"id":129470056,"uuid":"317069291","full_name":"grig-guz/tree-content-structuring","owner":"grig-guz","description":"Content structuring for NLG with discourse dependency trees.","archived":false,"fork":false,"pushed_at":"2020-11-30T00:18:04.000Z","size":38,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-12-31T00:53:53.899Z","etag":null,"topics":["content-structuring","nlg","nlg-dataset","rst"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/grig-guz.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2020-11-30T00:14:35.000Z","updated_at":"2022-04-14T23:45:05.000Z","dependencies_parsed_at":"2023-06-11T05:00:27.460Z","dependency_job_id":null,"html_url":"https://github.com/grig-guz/tree-content-structuring","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/grig-guz/tree-content-structuring","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/grig-guz%2Ftree-content-structuring","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/grig-guz%2Ftree-content-structuring/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/grig-guz%2Ftree-content-structuring/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/grig-guz%2Ftree-content-structuring/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/grig-guz","download_url":"https://codeload.github.com/grig-guz/tree-content-structuring/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/grig-guz%2Ftree-content-structuring/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32267710,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-25T09:15:33.318Z","status":"ssl_error","status_checked_at":"2026-04-25T09:15:31.997Z","response_time":59,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"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":["content-structuring","nlg","nlg-dataset","rst"],"created_at":"2024-11-19T08:53:33.367Z","updated_at":"2026-04-25T15:34:53.211Z","avatar_url":"https://github.com/grig-guz.png","language":"Python","readme":"# Domain-Independent Neural Text Structuring \nThis is the code for our [paper](https://www.aclweb.org/anthology/2020.findings-emnlp.281/) on text structuring with silver-standard discourse trees from [MEGA-DT treebank](https://www.cs.ubc.ca/cs-research/lci/research-groups/natural-language-processing/mega_dt.html).\n## Requirements\n\n* Python (3.6+)\n* [Pytorch](https://pytorch.org/) (1.3.0+)\n* [dgl](https://www.dgl.ai/) (0.4.2 strictly)\n* [Transformers](https://huggingface.co/transformers/) (3.0.2)\n\n## Running experiments\n1. Create the folder named \"data\".\n2. Download the pickled versions of MEGA-DT [here](https://www.todo) (100k train, 250k train, 5k val, 15k test), and place it in the \"data\" folder.\n3. Run the train/testing script as described below. Each scripts accepts a single numeric (1 or 2) indicating whether the model should be trained on 100k or 250k version of MEGA-DT.\n\n#### Dependency Model\nTo train/evaluate the dependency model, \n ```bash\nbash scripts/train_dep.sh dataset_id\nbash scripts/eval_dep.sh dataset_id\n```\n#### Pointer Model\n ```bash\nbash scripts/train_pointer.sh dataset_id\nbash scripts/eval_pointer.sh dataset_id\n ```\n #### Dependency no-pointer Baseline\n ```bash\nbash scripts/train_dep_treetrain_baseline.sh  dataset_id\nbash scripts/eval_dep_treetrain_baseline.sh dataset_id\n ```\n#### Language Model Decoding Baseline\n ```bash\nbash scripts/eval_lm_baseline.sh\n ```\n\n## Configuration\nYou can set hyperparameters and device type in the training/testing scripts for each model individually. The parameter values used in our experiments are already specified there.\n\n## Citation\n ```\n@inproceedings{guz-carenini-2020-towards,\n    title = \"Towards Domain-Independent Text Structuring Trainable on Large Discourse Treebanks\",\n    author = \"Guz, Grigorii  and\n      Carenini, Giuseppe\",\n    booktitle = \"Findings of the Association for Computational Linguistics: EMNLP 2020\",\n    month = nov,\n    year = \"2020\",\n    address = \"Online\",\n    publisher = \"Association for Computational Linguistics\",\n    url = \"https://www.aclweb.org/anthology/2020.findings-emnlp.281\",\n    pages = \"3141--3152\",\n}\n ```\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgrig-guz%2Ftree-content-structuring","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgrig-guz%2Ftree-content-structuring","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgrig-guz%2Ftree-content-structuring/lists"}