{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FXQZLK6WS7FBBHJHRIDF3W42RQ","short_pith_number":"pith:FXQZLK6W","schema_version":"1.0","canonical_sha256":"2de195abd697ca109d278a065ddb9a8c3764ca3198b2b8e4b49223a6bed3515c","source":{"kind":"arxiv","id":"2502.07281","version":1},"attestation_state":"computed","paper":{"title":"Supervised Contrastive Block Disentanglement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ana Carolina Leote, Aviv Regev, Burkhard Hoeckendorf, David Richmond, Heming Yao, Jan-Christian Huetter, Ji Won Park, Kyunghyun Cho, Natasa Tagasovska, Paula Coelho, Romain Lopez, Stephen Ra, Takamasa Kudo, Taro Makino","submitted_at":"2025-02-11T05:55:27Z","abstract_excerpt":"Real-world datasets often combine data collected under different experimental conditions. This yields larger datasets, but also introduces spurious correlations that make it difficult to model the phenomena of interest. We address this by learning two embeddings to independently represent the phenomena of interest and the spurious correlations. The embedding representing the phenomena of interest is correlated with the target variable $y$, and is invariant to the environment variable $e$. In contrast, the embedding representing the spurious correlations is correlated with $e$. The invariance t"},"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":"2502.07281","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-11T05:55:27Z","cross_cats_sorted":[],"title_canon_sha256":"2cdef15aada5f6ecb366a44fc82b4cff5b13b914a9391c1aee27e3f783f3299c","abstract_canon_sha256":"ee37d1aaf13c5a0000f7ae7aa2e36dbda55785b956ac7a663b0837915b600af2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:12:35.287853Z","signature_b64":"+liSErNw9Vy48eC2+z6I+oQYqOeO+cDB/htkBW0jtdQpYj3kVnR1nQJHuyhfgmGnoSzhq0heSj8DyMItB4PYAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2de195abd697ca109d278a065ddb9a8c3764ca3198b2b8e4b49223a6bed3515c","last_reissued_at":"2026-07-05T10:12:35.287338Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:12:35.287338Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Supervised Contrastive Block Disentanglement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ana Carolina Leote, Aviv Regev, Burkhard Hoeckendorf, David Richmond, Heming Yao, Jan-Christian Huetter, Ji Won Park, Kyunghyun Cho, Natasa Tagasovska, Paula Coelho, Romain Lopez, Stephen Ra, Takamasa Kudo, Taro Makino","submitted_at":"2025-02-11T05:55:27Z","abstract_excerpt":"Real-world datasets often combine data collected under different experimental conditions. This yields larger datasets, but also introduces spurious correlations that make it difficult to model the phenomena of interest. We address this by learning two embeddings to independently represent the phenomena of interest and the spurious correlations. The embedding representing the phenomena of interest is correlated with the target variable $y$, and is invariant to the environment variable $e$. In contrast, the embedding representing the spurious correlations is correlated with $e$. The invariance t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.07281","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/2502.07281/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":"2502.07281","created_at":"2026-07-05T10:12:35.287402+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.07281v1","created_at":"2026-07-05T10:12:35.287402+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.07281","created_at":"2026-07-05T10:12:35.287402+00:00"},{"alias_kind":"pith_short_12","alias_value":"FXQZLK6WS7FB","created_at":"2026-07-05T10:12:35.287402+00:00"},{"alias_kind":"pith_short_16","alias_value":"FXQZLK6WS7FBBHJH","created_at":"2026-07-05T10:12:35.287402+00:00"},{"alias_kind":"pith_short_8","alias_value":"FXQZLK6W","created_at":"2026-07-05T10:12:35.287402+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20538","citing_title":"Multi-Task Bayesian In-Context Learning","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18576","citing_title":"scHelix: Asymmetric Dual-Stream Integration via Explicit Gene-Level Disentanglement","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FXQZLK6WS7FBBHJHRIDF3W42RQ","json":"https://pith.science/pith/FXQZLK6WS7FBBHJHRIDF3W42RQ.json","graph_json":"https://pith.science/api/pith-number/FXQZLK6WS7FBBHJHRIDF3W42RQ/graph.json","events_json":"https://pith.science/api/pith-number/FXQZLK6WS7FBBHJHRIDF3W42RQ/events.json","paper":"https://pith.science/paper/FXQZLK6W"},"agent_actions":{"view_html":"https://pith.science/pith/FXQZLK6WS7FBBHJHRIDF3W42RQ","download_json":"https://pith.science/pith/FXQZLK6WS7FBBHJHRIDF3W42RQ.json","view_paper":"https://pith.science/paper/FXQZLK6W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.07281&json=true","fetch_graph":"https://pith.science/api/pith-number/FXQZLK6WS7FBBHJHRIDF3W42RQ/graph.json","fetch_events":"https://pith.science/api/pith-number/FXQZLK6WS7FBBHJHRIDF3W42RQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FXQZLK6WS7FBBHJHRIDF3W42RQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FXQZLK6WS7FBBHJHRIDF3W42RQ/action/storage_attestation","attest_author":"https://pith.science/pith/FXQZLK6WS7FBBHJHRIDF3W42RQ/action/author_attestation","sign_citation":"https://pith.science/pith/FXQZLK6WS7FBBHJHRIDF3W42RQ/action/citation_signature","submit_replication":"https://pith.science/pith/FXQZLK6WS7FBBHJHRIDF3W42RQ/action/replication_record"}},"created_at":"2026-07-05T10:12:35.287402+00:00","updated_at":"2026-07-05T10:12:35.287402+00:00"}