{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QFI42BCPYQMM4427T7JDT5ECGP","short_pith_number":"pith:QFI42BCP","schema_version":"1.0","canonical_sha256":"8151cd044fc418ce735f9fd239f48233e70e76be19d737082d88e4bd2d5894c3","source":{"kind":"arxiv","id":"2303.02343","version":1},"attestation_state":"computed","paper":{"title":"What Is Missing in IRM Training and Evaluation? Challenges and Solutions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Kush Varshney, Mingyi Hong, Parikshit Ram, Pranay Sharma, Sijia Liu, Yihua Zhang","submitted_at":"2023-03-04T07:06:24Z","abstract_excerpt":"Invariant risk minimization (IRM) has received increasing attention as a way to acquire environment-agnostic data representations and predictions, and as a principled solution for preventing spurious correlations from being learned and for improving models' out-of-distribution generalization. Yet, recent works have found that the optimality of the originally-proposed IRM optimization (IRM) may be compromised in practice or could be impossible to achieve in some scenarios. Therefore, a series of advanced IRM algorithms have been developed that show practical improvement over IRM. In this work, "},"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":"2303.02343","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-03-04T07:06:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fea082742fa84910d6cd281bf95aa5738f416dfa2fea15367f47a1e10ff7cb3f","abstract_canon_sha256":"f75e8b192ba53836dd0f578cb850b01d2c97dd08b961e581511cd499b53ba9aa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:48:10.000723Z","signature_b64":"FSPGtpfYkCsj7lSMD6CKptrA7MsZw/EtrfrfptpfECa4YswaswPUIRZ4P+zxo2IbNUk6MD/D6basSt07FnyOBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8151cd044fc418ce735f9fd239f48233e70e76be19d737082d88e4bd2d5894c3","last_reissued_at":"2026-07-05T05:48:10.000385Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:48:10.000385Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"What Is Missing in IRM Training and Evaluation? Challenges and Solutions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Kush Varshney, Mingyi Hong, Parikshit Ram, Pranay Sharma, Sijia Liu, Yihua Zhang","submitted_at":"2023-03-04T07:06:24Z","abstract_excerpt":"Invariant risk minimization (IRM) has received increasing attention as a way to acquire environment-agnostic data representations and predictions, and as a principled solution for preventing spurious correlations from being learned and for improving models' out-of-distribution generalization. Yet, recent works have found that the optimality of the originally-proposed IRM optimization (IRM) may be compromised in practice or could be impossible to achieve in some scenarios. Therefore, a series of advanced IRM algorithms have been developed that show practical improvement over IRM. In this work, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.02343","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/2303.02343/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":"2303.02343","created_at":"2026-07-05T05:48:10.000441+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.02343v1","created_at":"2026-07-05T05:48:10.000441+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.02343","created_at":"2026-07-05T05:48:10.000441+00:00"},{"alias_kind":"pith_short_12","alias_value":"QFI42BCPYQMM","created_at":"2026-07-05T05:48:10.000441+00:00"},{"alias_kind":"pith_short_16","alias_value":"QFI42BCPYQMM4427","created_at":"2026-07-05T05:48:10.000441+00:00"},{"alias_kind":"pith_short_8","alias_value":"QFI42BCP","created_at":"2026-07-05T05:48:10.000441+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.16126","citing_title":"Robust Invariant Representation Learning by Distribution Extrapolation","ref_index":2024,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QFI42BCPYQMM4427T7JDT5ECGP","json":"https://pith.science/pith/QFI42BCPYQMM4427T7JDT5ECGP.json","graph_json":"https://pith.science/api/pith-number/QFI42BCPYQMM4427T7JDT5ECGP/graph.json","events_json":"https://pith.science/api/pith-number/QFI42BCPYQMM4427T7JDT5ECGP/events.json","paper":"https://pith.science/paper/QFI42BCP"},"agent_actions":{"view_html":"https://pith.science/pith/QFI42BCPYQMM4427T7JDT5ECGP","download_json":"https://pith.science/pith/QFI42BCPYQMM4427T7JDT5ECGP.json","view_paper":"https://pith.science/paper/QFI42BCP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.02343&json=true","fetch_graph":"https://pith.science/api/pith-number/QFI42BCPYQMM4427T7JDT5ECGP/graph.json","fetch_events":"https://pith.science/api/pith-number/QFI42BCPYQMM4427T7JDT5ECGP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QFI42BCPYQMM4427T7JDT5ECGP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QFI42BCPYQMM4427T7JDT5ECGP/action/storage_attestation","attest_author":"https://pith.science/pith/QFI42BCPYQMM4427T7JDT5ECGP/action/author_attestation","sign_citation":"https://pith.science/pith/QFI42BCPYQMM4427T7JDT5ECGP/action/citation_signature","submit_replication":"https://pith.science/pith/QFI42BCPYQMM4427T7JDT5ECGP/action/replication_record"}},"created_at":"2026-07-05T05:48:10.000441+00:00","updated_at":"2026-07-05T05:48:10.000441+00:00"}