{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HNFL3RPJ6HFLMMGG33WRMOSOR4","short_pith_number":"pith:HNFL3RPJ","schema_version":"1.0","canonical_sha256":"3b4abdc5e9f1cab630c6deed163a4e8f22a72af1acd247cfc879b35aff6f9946","source":{"kind":"arxiv","id":"2405.16727","version":3},"attestation_state":"computed","paper":{"title":"Disentangling and Integrating Relational and Sensory Information in Transformer Architectures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Awni Altabaa, John Lafferty","submitted_at":"2024-05-26T23:52:51Z","abstract_excerpt":"Relational reasoning is a central component of generally intelligent systems, enabling robust and data-efficient inductive generalization. Recent empirical evidence shows that many existing neural architectures, including Transformers, struggle with tasks requiring relational reasoning. In this work, we distinguish between two types of information: sensory information about the properties of individual objects, and relational information about the relationships between objects. While neural attention provides a powerful mechanism for controlling the flow of sensory information between objects,"},"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":"2405.16727","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-26T23:52:51Z","cross_cats_sorted":[],"title_canon_sha256":"82f6ac23c5b5811acc1a7f3c3f7a95874a20ba8b3f2499fc4bcc78e3dcb32257","abstract_canon_sha256":"8b26af159801c9e7cbb5bd5984970044b0dab8d6ecdb3f58cbc165addf331367"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:24:04.312336Z","signature_b64":"NkdEtNMfb686WzFYs5nQe9CTSag+n3Lgpi9bau0jETZx34qgwp+IsL5Y6K0+hBKx3L79M/sBolp8h7l5gmeGBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3b4abdc5e9f1cab630c6deed163a4e8f22a72af1acd247cfc879b35aff6f9946","last_reissued_at":"2026-07-05T11:24:04.311759Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:24:04.311759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Disentangling and Integrating Relational and Sensory Information in Transformer Architectures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Awni Altabaa, John Lafferty","submitted_at":"2024-05-26T23:52:51Z","abstract_excerpt":"Relational reasoning is a central component of generally intelligent systems, enabling robust and data-efficient inductive generalization. Recent empirical evidence shows that many existing neural architectures, including Transformers, struggle with tasks requiring relational reasoning. In this work, we distinguish between two types of information: sensory information about the properties of individual objects, and relational information about the relationships between objects. While neural attention provides a powerful mechanism for controlling the flow of sensory information between objects,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.16727","kind":"arxiv","version":3},"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/2405.16727/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":"2405.16727","created_at":"2026-07-05T11:24:04.311820+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.16727v3","created_at":"2026-07-05T11:24:04.311820+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.16727","created_at":"2026-07-05T11:24:04.311820+00:00"},{"alias_kind":"pith_short_12","alias_value":"HNFL3RPJ6HFL","created_at":"2026-07-05T11:24:04.311820+00:00"},{"alias_kind":"pith_short_16","alias_value":"HNFL3RPJ6HFLMMGG","created_at":"2026-07-05T11:24:04.311820+00:00"},{"alias_kind":"pith_short_8","alias_value":"HNFL3RPJ","created_at":"2026-07-05T11:24:04.311820+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.05495","citing_title":"Shortcut Solutions Learned by Transformers Impair Continual Compositional Reasoning","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HNFL3RPJ6HFLMMGG33WRMOSOR4","json":"https://pith.science/pith/HNFL3RPJ6HFLMMGG33WRMOSOR4.json","graph_json":"https://pith.science/api/pith-number/HNFL3RPJ6HFLMMGG33WRMOSOR4/graph.json","events_json":"https://pith.science/api/pith-number/HNFL3RPJ6HFLMMGG33WRMOSOR4/events.json","paper":"https://pith.science/paper/HNFL3RPJ"},"agent_actions":{"view_html":"https://pith.science/pith/HNFL3RPJ6HFLMMGG33WRMOSOR4","download_json":"https://pith.science/pith/HNFL3RPJ6HFLMMGG33WRMOSOR4.json","view_paper":"https://pith.science/paper/HNFL3RPJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.16727&json=true","fetch_graph":"https://pith.science/api/pith-number/HNFL3RPJ6HFLMMGG33WRMOSOR4/graph.json","fetch_events":"https://pith.science/api/pith-number/HNFL3RPJ6HFLMMGG33WRMOSOR4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HNFL3RPJ6HFLMMGG33WRMOSOR4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HNFL3RPJ6HFLMMGG33WRMOSOR4/action/storage_attestation","attest_author":"https://pith.science/pith/HNFL3RPJ6HFLMMGG33WRMOSOR4/action/author_attestation","sign_citation":"https://pith.science/pith/HNFL3RPJ6HFLMMGG33WRMOSOR4/action/citation_signature","submit_replication":"https://pith.science/pith/HNFL3RPJ6HFLMMGG33WRMOSOR4/action/replication_record"}},"created_at":"2026-07-05T11:24:04.311820+00:00","updated_at":"2026-07-05T11:24:04.311820+00:00"}