{"url":"/task/unconditional-molecule-generation","name":"Unconditional Molecule Generation","slug":"unconditional-molecule-generation","description_markdown":"This task evaluates the ability of generative models to sample valid and realistic molecular structures.\r\n\r\nThe training dataset can be:\r\n- QM9 (Wu et al., 2018) - consists of 130,000 stable small organic molecules containing up to nine heavy atoms (C, N, O, F) along with hydrogens.\r\n- GEOM-DRUGS (Axelrod and Gómez-Bombarelli, 2022) - consistes of 430,000 large organic molecules of up to 180 atoms. \r\n\r\nFollowing prior work (Hoogeboom et al., 2022), we generally sample 10,000 molecules and compute validity, uniqueness and Posebusters sanity checks (Buttenschoen et al., 2024) for molecules.  Data is generally split following prior work (Hoogeboom et al., 2022, Vignac et al., 2023) to ensure fair comparisons.","categories":[{"name":"Miscellaneous","url":"/area/miscellaneous"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":8,"papers_with_code":6,"benchmarks":2,"benchmark_tables_in_archive":2,"benchmark_tables_shown":2,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":2,"subtasks":0,"parent_tasks":0},"benchmarks":[{"leaderboard":"/sota/unconditional-molecule-generation-on-geom","slug":"unconditional-molecule-generation-on-geom","dataset":"GEOM-DRUGS","dataset_url":"/dataset/geom-drugs","rows_in_archive":5,"metrics":["PoseBusters Validity","Validity","PoseBusters Atoms Connected"],"first_row_in_archive_order":{"model":"TABASCO","paper_title":"TABASCO: A Fast, Simplified Model for Molecular Generation with Improved Physical Quality","paper_url":"/paper/tabasco-a-fast-simplified-model-for-molecular-1","paper_date":"2025-07-01","arxiv_id":"2507.00899","code_links":[{"title":"carlosinator/tabasco","url":"https://github.com/carlosinator/tabasco"}],"syntology":null}},{"leaderboard":"/sota/unconditional-molecule-generation-on-qm9","slug":"unconditional-molecule-generation-on-qm9","dataset":"QM9","dataset_url":"/dataset/qm9","rows_in_archive":4,"metrics":["Validity","PoseBusters Internal Energy"],"first_row_in_archive_order":{"model":"ADiT","paper_title":"All-atom Diffusion Transformers: Unified generative modelling of molecules and materials","paper_url":"/paper/all-atom-diffusion-transformers-unified","paper_date":"2025-03-05","arxiv_id":"2503.03965","code_links":[{"title":"facebookresearch/all-atom-diffusion-transformer","url":"https://github.com/facebookresearch/all-atom-diffusion-transformer"}],"syntology":{"n":7,"n_ran":6,"n_unverified":1,"n_pointer_only":7}}}],"datasets":[{"url":"/dataset/qm9","name":"QM9","full_name":"","num_papers_in_archive":76},{"url":"/dataset/geom-drugs","name":"GEOM-DRUGS","full_name":"","num_papers_in_archive":20}],"subtasks":[],"parent_tasks":[],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":6,"of":6,"tagged_in_all":8,"items":[{"url":"/paper/equivariant-diffusion-for-molecule-generation","title":"Equivariant Diffusion for Molecule Generation in 3D","date":"2022-03-31","arxiv_id":"2203.17003","repositories_listed":4,"syntology":{"n":13,"n_ran":7,"n_unverified":6,"n_pointer_only":2}},{"url":"/paper/geometric-latent-diffusion-models-for-3d","title":"Geometric Latent Diffusion Models for 3D Molecule Generation","date":"2023-05-02","arxiv_id":"2305.01140","repositories_listed":2,"syntology":{"n":9,"n_ran":4,"n_unverified":5,"n_pointer_only":0}},{"url":"/paper/all-atom-diffusion-transformers-unified","title":"All-atom Diffusion Transformers: Unified generative modelling of molecules and materials","date":"2025-03-05","arxiv_id":"2503.03965","repositories_listed":1,"syntology":{"n":7,"n_ran":6,"n_unverified":1,"n_pointer_only":7}},{"url":"/paper/geometric-representation-condition-improves","title":"Geometric Representation Condition Improves Equivariant Molecule Generation","date":"2024-10-04","arxiv_id":"2410.03655","repositories_listed":1,"syntology":{"n":13,"n_ran":9,"n_unverified":4,"n_pointer_only":13}},{"url":"/paper/symphony-symmetry-equivariant-point-centered","title":"Symphony: Symmetry-Equivariant Point-Centered Spherical Harmonics for 3D Molecule Generation","date":"2023-11-27","arxiv_id":"2311.16199","repositories_listed":1,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/midi-mixed-graph-and-3d-denoising-diffusion","title":"MiDi: Mixed Graph and 3D Denoising Diffusion for Molecule Generation","date":"2023-02-17","arxiv_id":"2302.09048","repositories_listed":1,"syntology":{"n":13,"n_ran":2,"n_unverified":11,"n_pointer_only":0}}],"syntology_records":6,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}