{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:4AXXQZPHDKAYCKHPDMDPZQNZZ6","short_pith_number":"pith:4AXXQZPH","schema_version":"1.0","canonical_sha256":"e02f7865e71a818128ef1b06fcc1b9cfa196dd5d12fe2ff889556babc15354c1","source":{"kind":"arxiv","id":"2011.13230","version":1},"attestation_state":"computed","paper":{"title":"Molecular representation learning with language models and domain-relevant auxiliary tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Benedek Fabian, H\\'el\\'ena Gaspar, Joshua Meyers, Marco Fiscato, Marwin Segler, Mohamed Ahmed, Thomas Edlich","submitted_at":"2020-11-26T10:55:05Z","abstract_excerpt":"We apply a Transformer architecture, specifically BERT, to learn flexible and high quality molecular representations for drug discovery problems. We study the impact of using different combinations of self-supervised tasks for pre-training, and present our results for the established Virtual Screening and QSAR benchmarks. We show that: i) The selection of appropriate self-supervised task(s) for pre-training has a significant impact on performance in subsequent downstream tasks such as Virtual Screening. ii) Using auxiliary tasks with more domain relevance for Chemistry, such as learning to pre"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2011.13230","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-11-26T10:55:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"84f173be46c90ac9c5823299d60aff14e4808645ad231fdbdd569a655df8e9ca","abstract_canon_sha256":"a445f2515573b21d7866b4fac3aac37cd8735912949eeddfeb0d2f4b2d03c17d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:54:52.205581Z","signature_b64":"4+ufZm8PdnToJkxaakFG04LTBRFP45u6tClCwVdzG/RyR+GNRDHhD2Ikyu84ItXBqBe6wvEQmnWqYSYUgpKeBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e02f7865e71a818128ef1b06fcc1b9cfa196dd5d12fe2ff889556babc15354c1","last_reissued_at":"2026-07-05T01:54:52.205020Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:54:52.205020Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Molecular representation learning with language models and domain-relevant auxiliary tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Benedek Fabian, H\\'el\\'ena Gaspar, Joshua Meyers, Marco Fiscato, Marwin Segler, Mohamed Ahmed, Thomas Edlich","submitted_at":"2020-11-26T10:55:05Z","abstract_excerpt":"We apply a Transformer architecture, specifically BERT, to learn flexible and high quality molecular representations for drug discovery problems. We study the impact of using different combinations of self-supervised tasks for pre-training, and present our results for the established Virtual Screening and QSAR benchmarks. We show that: i) The selection of appropriate self-supervised task(s) for pre-training has a significant impact on performance in subsequent downstream tasks such as Virtual Screening. ii) Using auxiliary tasks with more domain relevance for Chemistry, such as learning to pre"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2011.13230","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2011.13230/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2011.13230","created_at":"2026-07-05T01:54:52.205083+00:00"},{"alias_kind":"arxiv_version","alias_value":"2011.13230v1","created_at":"2026-07-05T01:54:52.205083+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2011.13230","created_at":"2026-07-05T01:54:52.205083+00:00"},{"alias_kind":"pith_short_12","alias_value":"4AXXQZPHDKAY","created_at":"2026-07-05T01:54:52.205083+00:00"},{"alias_kind":"pith_short_16","alias_value":"4AXXQZPHDKAYCKHP","created_at":"2026-07-05T01:54:52.205083+00:00"},{"alias_kind":"pith_short_8","alias_value":"4AXXQZPH","created_at":"2026-07-05T01:54:52.205083+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.02140","citing_title":"Probing Chemical Language Models: Effects of Pre-training and Fine-tuning","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12651","citing_title":"Physics-Aware Auxiliary Losses Improve Out-of-Distribution Generalization of a GNN Synthesizability Filter","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2504.16559","citing_title":"Synergistic Benefits of Joint Molecule Generation and Property Prediction","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2602.22822","citing_title":"FlexMS is a flexible framework for benchmarking deep learning-based mass spectrum prediction tools in metabolomics","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13262","citing_title":"Chem-GMNet: A Sphere-Native Geometric Transformer for Molecular Property Prediction","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06558","citing_title":"When Does Context Help? A Systematic Study of Target-Conditional Molecular Property Prediction","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4AXXQZPHDKAYCKHPDMDPZQNZZ6","json":"https://pith.science/pith/4AXXQZPHDKAYCKHPDMDPZQNZZ6.json","graph_json":"https://pith.science/api/pith-number/4AXXQZPHDKAYCKHPDMDPZQNZZ6/graph.json","events_json":"https://pith.science/api/pith-number/4AXXQZPHDKAYCKHPDMDPZQNZZ6/events.json","paper":"https://pith.science/paper/4AXXQZPH"},"agent_actions":{"view_html":"https://pith.science/pith/4AXXQZPHDKAYCKHPDMDPZQNZZ6","download_json":"https://pith.science/pith/4AXXQZPHDKAYCKHPDMDPZQNZZ6.json","view_paper":"https://pith.science/paper/4AXXQZPH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2011.13230&json=true","fetch_graph":"https://pith.science/api/pith-number/4AXXQZPHDKAYCKHPDMDPZQNZZ6/graph.json","fetch_events":"https://pith.science/api/pith-number/4AXXQZPHDKAYCKHPDMDPZQNZZ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4AXXQZPHDKAYCKHPDMDPZQNZZ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4AXXQZPHDKAYCKHPDMDPZQNZZ6/action/storage_attestation","attest_author":"https://pith.science/pith/4AXXQZPHDKAYCKHPDMDPZQNZZ6/action/author_attestation","sign_citation":"https://pith.science/pith/4AXXQZPHDKAYCKHPDMDPZQNZZ6/action/citation_signature","submit_replication":"https://pith.science/pith/4AXXQZPHDKAYCKHPDMDPZQNZZ6/action/replication_record"}},"created_at":"2026-07-05T01:54:52.205083+00:00","updated_at":"2026-07-05T01:54:52.205083+00:00"}