{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:T6U3B4K35QKB5ZEONYQQ2E2RBD","short_pith_number":"pith:T6U3B4K3","schema_version":"1.0","canonical_sha256":"9fa9b0f15bec141ee48e6e210d135108ef64e0888085e4634b4d2952e965bda2","source":{"kind":"arxiv","id":"2310.09336","version":5},"attestation_state":"computed","paper":{"title":"Compositional Abilities Emerge Multiplicatively: Exploring Diffusion Models on a Synthetic Task","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ekdeep Singh Lubana, Hidenori Tanaka, Maya Okawa, Robert P. Dick","submitted_at":"2023-10-13T18:00:59Z","abstract_excerpt":"Modern generative models exhibit unprecedented capabilities to generate extremely realistic data. However, given the inherent compositionality of the real world, reliable use of these models in practical applications requires that they exhibit the capability to compose a novel set of concepts to generate outputs not seen in the training data set. Prior work demonstrates that recent diffusion models do exhibit intriguing compositional generalization abilities, but also fail unpredictably. Motivated by this, we perform a controlled study for understanding compositional generalization in conditio"},"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":"2310.09336","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-13T18:00:59Z","cross_cats_sorted":[],"title_canon_sha256":"5910503fb5226184ed22217cf1274ba77cab6ad0e77bf9a47ddf033563d9916d","abstract_canon_sha256":"96a24a07df5dfe3a72eb65794a2f5c0115648b2727904e289e1d687bfcf6a42f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:43:27.234602Z","signature_b64":"afkmT/OfG/w1FN2Tyf6nWUMXlGpfSFf3aIGt3Kt8fptqwKy4E17D1NJX4n9D11dvrrf8Wg4s7NPJnW4k+SEuDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9fa9b0f15bec141ee48e6e210d135108ef64e0888085e4634b4d2952e965bda2","last_reissued_at":"2026-07-05T11:43:27.234115Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:43:27.234115Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Compositional Abilities Emerge Multiplicatively: Exploring Diffusion Models on a Synthetic Task","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ekdeep Singh Lubana, Hidenori Tanaka, Maya Okawa, Robert P. Dick","submitted_at":"2023-10-13T18:00:59Z","abstract_excerpt":"Modern generative models exhibit unprecedented capabilities to generate extremely realistic data. However, given the inherent compositionality of the real world, reliable use of these models in practical applications requires that they exhibit the capability to compose a novel set of concepts to generate outputs not seen in the training data set. Prior work demonstrates that recent diffusion models do exhibit intriguing compositional generalization abilities, but also fail unpredictably. Motivated by this, we perform a controlled study for understanding compositional generalization in conditio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.09336","kind":"arxiv","version":5},"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/2310.09336/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":"2310.09336","created_at":"2026-07-05T11:43:27.234171+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.09336v5","created_at":"2026-07-05T11:43:27.234171+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.09336","created_at":"2026-07-05T11:43:27.234171+00:00"},{"alias_kind":"pith_short_12","alias_value":"T6U3B4K35QKB","created_at":"2026-07-05T11:43:27.234171+00:00"},{"alias_kind":"pith_short_16","alias_value":"T6U3B4K35QKB5ZEO","created_at":"2026-07-05T11:43:27.234171+00:00"},{"alias_kind":"pith_short_8","alias_value":"T6U3B4K3","created_at":"2026-07-05T11:43:27.234171+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08127","citing_title":"Understanding and Mitigating the Video-Action Generalization Gap via Temporal Ratio","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2605.27696","citing_title":"Structure over Pixels: Learning Variable-Length Visual Programs","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2602.04759","citing_title":"How to Stop Playing Whack-a-Mole: Mapping the Ecosystem of Technologies Facilitating AI-Generated Non-Consensual Intimate Images","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20299","citing_title":"Mechanisms of Misgeneralization in Physical Sequence Modeling","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10019","citing_title":"The two clocks and the innovation window: When and how generative models learn rules","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17673","citing_title":"Grokking of Diffusion Models: Case Study on Modular Addition","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T6U3B4K35QKB5ZEONYQQ2E2RBD","json":"https://pith.science/pith/T6U3B4K35QKB5ZEONYQQ2E2RBD.json","graph_json":"https://pith.science/api/pith-number/T6U3B4K35QKB5ZEONYQQ2E2RBD/graph.json","events_json":"https://pith.science/api/pith-number/T6U3B4K35QKB5ZEONYQQ2E2RBD/events.json","paper":"https://pith.science/paper/T6U3B4K3"},"agent_actions":{"view_html":"https://pith.science/pith/T6U3B4K35QKB5ZEONYQQ2E2RBD","download_json":"https://pith.science/pith/T6U3B4K35QKB5ZEONYQQ2E2RBD.json","view_paper":"https://pith.science/paper/T6U3B4K3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.09336&json=true","fetch_graph":"https://pith.science/api/pith-number/T6U3B4K35QKB5ZEONYQQ2E2RBD/graph.json","fetch_events":"https://pith.science/api/pith-number/T6U3B4K35QKB5ZEONYQQ2E2RBD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T6U3B4K35QKB5ZEONYQQ2E2RBD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T6U3B4K35QKB5ZEONYQQ2E2RBD/action/storage_attestation","attest_author":"https://pith.science/pith/T6U3B4K35QKB5ZEONYQQ2E2RBD/action/author_attestation","sign_citation":"https://pith.science/pith/T6U3B4K35QKB5ZEONYQQ2E2RBD/action/citation_signature","submit_replication":"https://pith.science/pith/T6U3B4K35QKB5ZEONYQQ2E2RBD/action/replication_record"}},"created_at":"2026-07-05T11:43:27.234171+00:00","updated_at":"2026-07-05T11:43:27.234171+00:00"}