{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:C3NRL6W35V2H4QB6G7YSPW4OLD","short_pith_number":"pith:C3NRL6W3","schema_version":"1.0","canonical_sha256":"16db15fadbed747e403e37f127db8e58d9a0de63c5a2fd9aa9abb19603b517eb","source":{"kind":"arxiv","id":"2305.16192","version":1},"attestation_state":"computed","paper":{"title":"Explainability Techniques for Chemical Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","physics.chem-ph","q-bio.QM"],"primary_cat":"cs.LG","authors_text":"Stefan H\\\"odl, Tal Kachman, Wilhelm Huck, William Robinson, Yoram Bachrach","submitted_at":"2023-05-25T15:52:54Z","abstract_excerpt":"Explainability techniques are crucial in gaining insights into the reasons behind the predictions of deep learning models, which have not yet been applied to chemical language models. We propose an explainable AI technique that attributes the importance of individual atoms towards the predictions made by these models. Our method backpropagates the relevance information towards the chemical input string and visualizes the importance of individual atoms. We focus on self-attention Transformers operating on molecular string representations and leverage a pretrained encoder for finetuning. We show"},"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":"2305.16192","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-25T15:52:54Z","cross_cats_sorted":["cs.AI","physics.chem-ph","q-bio.QM"],"title_canon_sha256":"8350d74eba80cf878b3e3919a156a58d2e52c6aaa9f57c5d4acdfa9c6789588b","abstract_canon_sha256":"d5b553e0e40c859b7a22c57aa36d7d931b1a1ac774e23b3d875d19ba65e48953"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:14:11.629507Z","signature_b64":"GLf4qgP0KF2hiBwpHgOoWbx3EdZnj0XPPMltfScYnuUn0V+lqlupyGD7zvJteJq+Fi+9I1Pzj/Z10SN40A/3DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"16db15fadbed747e403e37f127db8e58d9a0de63c5a2fd9aa9abb19603b517eb","last_reissued_at":"2026-07-05T06:14:11.629069Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:14:11.629069Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Explainability Techniques for Chemical Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","physics.chem-ph","q-bio.QM"],"primary_cat":"cs.LG","authors_text":"Stefan H\\\"odl, Tal Kachman, Wilhelm Huck, William Robinson, Yoram Bachrach","submitted_at":"2023-05-25T15:52:54Z","abstract_excerpt":"Explainability techniques are crucial in gaining insights into the reasons behind the predictions of deep learning models, which have not yet been applied to chemical language models. We propose an explainable AI technique that attributes the importance of individual atoms towards the predictions made by these models. Our method backpropagates the relevance information towards the chemical input string and visualizes the importance of individual atoms. We focus on self-attention Transformers operating on molecular string representations and leverage a pretrained encoder for finetuning. We show"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.16192","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/2305.16192/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":"2305.16192","created_at":"2026-07-05T06:14:11.629131+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.16192v1","created_at":"2026-07-05T06:14:11.629131+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.16192","created_at":"2026-07-05T06:14:11.629131+00:00"},{"alias_kind":"pith_short_12","alias_value":"C3NRL6W35V2H","created_at":"2026-07-05T06:14:11.629131+00:00"},{"alias_kind":"pith_short_16","alias_value":"C3NRL6W35V2H4QB6","created_at":"2026-07-05T06:14:11.629131+00:00"},{"alias_kind":"pith_short_8","alias_value":"C3NRL6W3","created_at":"2026-07-05T06:14:11.629131+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.10273","citing_title":"Conditional Chemical Language Models are Versatile Tools in Drug Discovery","ref_index":2004,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C3NRL6W35V2H4QB6G7YSPW4OLD","json":"https://pith.science/pith/C3NRL6W35V2H4QB6G7YSPW4OLD.json","graph_json":"https://pith.science/api/pith-number/C3NRL6W35V2H4QB6G7YSPW4OLD/graph.json","events_json":"https://pith.science/api/pith-number/C3NRL6W35V2H4QB6G7YSPW4OLD/events.json","paper":"https://pith.science/paper/C3NRL6W3"},"agent_actions":{"view_html":"https://pith.science/pith/C3NRL6W35V2H4QB6G7YSPW4OLD","download_json":"https://pith.science/pith/C3NRL6W35V2H4QB6G7YSPW4OLD.json","view_paper":"https://pith.science/paper/C3NRL6W3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.16192&json=true","fetch_graph":"https://pith.science/api/pith-number/C3NRL6W35V2H4QB6G7YSPW4OLD/graph.json","fetch_events":"https://pith.science/api/pith-number/C3NRL6W35V2H4QB6G7YSPW4OLD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C3NRL6W35V2H4QB6G7YSPW4OLD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C3NRL6W35V2H4QB6G7YSPW4OLD/action/storage_attestation","attest_author":"https://pith.science/pith/C3NRL6W35V2H4QB6G7YSPW4OLD/action/author_attestation","sign_citation":"https://pith.science/pith/C3NRL6W35V2H4QB6G7YSPW4OLD/action/citation_signature","submit_replication":"https://pith.science/pith/C3NRL6W35V2H4QB6G7YSPW4OLD/action/replication_record"}},"created_at":"2026-07-05T06:14:11.629131+00:00","updated_at":"2026-07-05T06:14:11.629131+00:00"}