{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:X6TD6OMVUJJU2NFNM3TOG7HRI4","short_pith_number":"pith:X6TD6OMV","canonical_record":{"source":{"id":"2604.14441","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.SP","submitted_at":"2026-04-15T21:33:03Z","cross_cats_sorted":[],"title_canon_sha256":"9ced49be8bd76810066158751d779249c7dec77ebc375e53c72c373c79bfed9f","abstract_canon_sha256":"360c17bc47bc8b236be208f86d657b085b886f71623230c2b8eb584e0466a531"},"schema_version":"1.0"},"canonical_sha256":"bfa63f3995a2534d34ad66e6e37cf147102cda8ae2ef4aac1ea7d371702dca01","source":{"kind":"arxiv","id":"2604.14441","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.14441","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"arxiv_version","alias_value":"2604.14441v2","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.14441","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"pith_short_12","alias_value":"X6TD6OMVUJJU","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"pith_short_16","alias_value":"X6TD6OMVUJJU2NFN","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"pith_short_8","alias_value":"X6TD6OMV","created_at":"2026-06-09T02:08:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:X6TD6OMVUJJU2NFNM3TOG7HRI4","target":"record","payload":{"canonical_record":{"source":{"id":"2604.14441","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.SP","submitted_at":"2026-04-15T21:33:03Z","cross_cats_sorted":[],"title_canon_sha256":"9ced49be8bd76810066158751d779249c7dec77ebc375e53c72c373c79bfed9f","abstract_canon_sha256":"360c17bc47bc8b236be208f86d657b085b886f71623230c2b8eb584e0466a531"},"schema_version":"1.0"},"canonical_sha256":"bfa63f3995a2534d34ad66e6e37cf147102cda8ae2ef4aac1ea7d371702dca01","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:08:42.333856Z","signature_b64":"QJJ/UQS+I4m2lZFlCZXE+2vRJH/Rt9/dwNq92doDzcCVS9AMhbPFpBtiLUgmWKHBJ8kXx2RrOa2vaca3dhnIAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bfa63f3995a2534d34ad66e6e37cf147102cda8ae2ef4aac1ea7d371702dca01","last_reissued_at":"2026-06-09T02:08:42.332869Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:08:42.332869Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.14441","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-09T02:08:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rLqPU8P5zQdnqlH0wlqyoK1G4s/FA9HH+Jun0esAKTBXHh9BBn1Nj1hWCcR/qogj+Coq9NHRHcA0pO8hyEzVCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T00:47:36.136187Z"},"content_sha256":"36140d7ec0d0855ebdcdd9823ef93b68bbf1aedaf11b0bb0d4beb6b103ef1673","schema_version":"1.0","event_id":"sha256:36140d7ec0d0855ebdcdd9823ef93b68bbf1aedaf11b0bb0d4beb6b103ef1673"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:X6TD6OMVUJJU2NFNM3TOG7HRI4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Batch Effects In Brain Foundation Model Embeddings","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Foundation model embeddings from brain scans encode substantial batch effects that often dominate diagnosis signals.","cross_cats":[],"primary_cat":"eess.SP","authors_text":"Anand D. Sarwate, Bradley T. Baker, Sandeep Panta, Sergey Plis, Vince D. Calhoun, Ye Tao, Yu Wu","submitted_at":"2026-04-15T21:33:03Z","abstract_excerpt":"Foundation models show strong potential for large-scale, high-dimensional biomedical applications, yet their ability to capture relevant neurobiological characteristics remains underexplored. We systematically evaluate embeddings from two neuroimaging foundation models, BrainLM and SwiFT, across multi-site fMRI datasets using a comprehensive evaluation framework. Our results show that foundation model embeddings encode substantial batch-related variability, often dominating diagnosis-related information across heterogeneous datasets. We further investigate how harmonization, applied to reduce "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our results show that foundation model embeddings encode substantial batch-related variability, often dominating diagnosis-related information across heterogeneous datasets.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The evaluation framework and chosen metrics correctly isolate batch effects from biologically meaningful signals without the framework itself introducing or amplifying site-specific artifacts.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Embeddings from BrainLM and SwiFT foundation models encode substantial batch-related variability that frequently overshadows diagnosis signals in multi-site fMRI datasets.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Foundation model embeddings from brain scans encode substantial batch effects that often dominate diagnosis signals.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"2dc76b545d9faa1a6207ad25bfcd16b6d546054297f89317d671f507e6307b6b"},"source":{"id":"2604.14441","kind":"arxiv","version":2},"verdict":{"id":"339294e9-6495-4bd0-adb6-65b02ea50e9b","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T12:10:42.587382Z","strongest_claim":"Our results show that foundation model embeddings encode substantial batch-related variability, often dominating diagnosis-related information across heterogeneous datasets.","one_line_summary":"Embeddings from BrainLM and SwiFT foundation models encode substantial batch-related variability that frequently overshadows diagnosis signals in multi-site fMRI datasets.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The evaluation framework and chosen metrics correctly isolate batch effects from biologically meaningful signals without the framework itself introducing or amplifying site-specific artifacts.","pith_extraction_headline":"Foundation model embeddings from brain scans encode substantial batch effects that often dominate diagnosis signals."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.14441/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"},"verdict_id":"339294e9-6495-4bd0-adb6-65b02ea50e9b"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-09T02:08:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EbFKvWdz4h0ViEnHm9FrBEVwahxe0vOX7VY8nG2c3tLZYN/ClPQMKlyqiHAw4MpaBDgXazoCcrtuvc83Yy7fDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T00:47:36.136809Z"},"content_sha256":"649fb7fa0af87f18e589a386de29444ed0fe8d279e0c31b8f5d4aa3f63e9f8a0","schema_version":"1.0","event_id":"sha256:649fb7fa0af87f18e589a386de29444ed0fe8d279e0c31b8f5d4aa3f63e9f8a0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/X6TD6OMVUJJU2NFNM3TOG7HRI4/bundle.json","state_url":"https://pith.science/pith/X6TD6OMVUJJU2NFNM3TOG7HRI4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/X6TD6OMVUJJU2NFNM3TOG7HRI4/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T00:47:36Z","links":{"resolver":"https://pith.science/pith/X6TD6OMVUJJU2NFNM3TOG7HRI4","bundle":"https://pith.science/pith/X6TD6OMVUJJU2NFNM3TOG7HRI4/bundle.json","state":"https://pith.science/pith/X6TD6OMVUJJU2NFNM3TOG7HRI4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/X6TD6OMVUJJU2NFNM3TOG7HRI4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:X6TD6OMVUJJU2NFNM3TOG7HRI4","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"360c17bc47bc8b236be208f86d657b085b886f71623230c2b8eb584e0466a531","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.SP","submitted_at":"2026-04-15T21:33:03Z","title_canon_sha256":"9ced49be8bd76810066158751d779249c7dec77ebc375e53c72c373c79bfed9f"},"schema_version":"1.0","source":{"id":"2604.14441","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.14441","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"arxiv_version","alias_value":"2604.14441v2","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.14441","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"pith_short_12","alias_value":"X6TD6OMVUJJU","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"pith_short_16","alias_value":"X6TD6OMVUJJU2NFN","created_at":"2026-06-09T02:08:42Z"},{"alias_kind":"pith_short_8","alias_value":"X6TD6OMV","created_at":"2026-06-09T02:08:42Z"}],"graph_snapshots":[{"event_id":"sha256:649fb7fa0af87f18e589a386de29444ed0fe8d279e0c31b8f5d4aa3f63e9f8a0","target":"graph","created_at":"2026-06-09T02:08:42Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Our results show that foundation model embeddings encode substantial batch-related variability, often dominating diagnosis-related information across heterogeneous datasets."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The evaluation framework and chosen metrics correctly isolate batch effects from biologically meaningful signals without the framework itself introducing or amplifying site-specific artifacts."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Embeddings from BrainLM and SwiFT foundation models encode substantial batch-related variability that frequently overshadows diagnosis signals in multi-site fMRI datasets."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Foundation model embeddings from brain scans encode substantial batch effects that often dominate diagnosis signals."}],"snapshot_sha256":"2dc76b545d9faa1a6207ad25bfcd16b6d546054297f89317d671f507e6307b6b"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.14441/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Foundation models show strong potential for large-scale, high-dimensional biomedical applications, yet their ability to capture relevant neurobiological characteristics remains underexplored. We systematically evaluate embeddings from two neuroimaging foundation models, BrainLM and SwiFT, across multi-site fMRI datasets using a comprehensive evaluation framework. Our results show that foundation model embeddings encode substantial batch-related variability, often dominating diagnosis-related information across heterogeneous datasets. We further investigate how harmonization, applied to reduce ","authors_text":"Anand D. Sarwate, Bradley T. Baker, Sandeep Panta, Sergey Plis, Vince D. Calhoun, Ye Tao, Yu Wu","cross_cats":[],"headline":"Foundation model embeddings from brain scans encode substantial batch effects that often dominate diagnosis signals.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.SP","submitted_at":"2026-04-15T21:33:03Z","title":"Batch Effects In Brain Foundation Model Embeddings"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.14441","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T12:10:42.587382Z","id":"339294e9-6495-4bd0-adb6-65b02ea50e9b","model_set":{"reader":"grok-4.3"},"one_line_summary":"Embeddings from BrainLM and SwiFT foundation models encode substantial batch-related variability that frequently overshadows diagnosis signals in multi-site fMRI datasets.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Foundation model embeddings from brain scans encode substantial batch effects that often dominate diagnosis signals.","strongest_claim":"Our results show that foundation model embeddings encode substantial batch-related variability, often dominating diagnosis-related information across heterogeneous datasets.","weakest_assumption":"The evaluation framework and chosen metrics correctly isolate batch effects from biologically meaningful signals without the framework itself introducing or amplifying site-specific artifacts."}},"verdict_id":"339294e9-6495-4bd0-adb6-65b02ea50e9b"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:36140d7ec0d0855ebdcdd9823ef93b68bbf1aedaf11b0bb0d4beb6b103ef1673","target":"record","created_at":"2026-06-09T02:08:42Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"360c17bc47bc8b236be208f86d657b085b886f71623230c2b8eb584e0466a531","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.SP","submitted_at":"2026-04-15T21:33:03Z","title_canon_sha256":"9ced49be8bd76810066158751d779249c7dec77ebc375e53c72c373c79bfed9f"},"schema_version":"1.0","source":{"id":"2604.14441","kind":"arxiv","version":2}},"canonical_sha256":"bfa63f3995a2534d34ad66e6e37cf147102cda8ae2ef4aac1ea7d371702dca01","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bfa63f3995a2534d34ad66e6e37cf147102cda8ae2ef4aac1ea7d371702dca01","first_computed_at":"2026-06-09T02:08:42.332869Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T02:08:42.332869Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QJJ/UQS+I4m2lZFlCZXE+2vRJH/Rt9/dwNq92doDzcCVS9AMhbPFpBtiLUgmWKHBJ8kXx2RrOa2vaca3dhnIAA==","signature_status":"signed_v1","signed_at":"2026-06-09T02:08:42.333856Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.14441","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:36140d7ec0d0855ebdcdd9823ef93b68bbf1aedaf11b0bb0d4beb6b103ef1673","sha256:649fb7fa0af87f18e589a386de29444ed0fe8d279e0c31b8f5d4aa3f63e9f8a0"],"state_sha256":"e0e30a432f0bc1c5daad4f2dd3c76cd9f2d153d13877b3df9c0db09961fac25f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0/LLXxar1GZhLJcfaRSX+0DSN/1EeiZc2NOCIkcHFk6SQuVu47GE247uRo4yU0yF7MOH6nbcpgf8jydhT9FZCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T00:47:36.140759Z","bundle_sha256":"513923dec1127d7641b49d3428ef797c7ab7f33e14c6378b655bb29df468a6d6"}}