{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:XPMPOXGZJZYTJVRAQ7KV3RZVPX","short_pith_number":"pith:XPMPOXGZ","canonical_record":{"source":{"id":"2605.07598","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-08T11:16:40Z","cross_cats_sorted":[],"title_canon_sha256":"6b8225d65541a0c737c88378358b7d5ee437b5496515c0f6257d45f335279765","abstract_canon_sha256":"272cccac106c6881dc3d9c8908e9388f678cc3090dc80116987cd7eb941cc6f5"},"schema_version":"1.0"},"canonical_sha256":"bbd8f75cd94e7134d62087d55dc7357dd007b0819d29a41ead0cc815ec31cd54","source":{"kind":"arxiv","id":"2605.07598","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.07598","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"arxiv_version","alias_value":"2605.07598v2","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.07598","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"pith_short_12","alias_value":"XPMPOXGZJZYT","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"pith_short_16","alias_value":"XPMPOXGZJZYTJVRA","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"pith_short_8","alias_value":"XPMPOXGZ","created_at":"2026-05-22T01:04:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:XPMPOXGZJZYTJVRAQ7KV3RZVPX","target":"record","payload":{"canonical_record":{"source":{"id":"2605.07598","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-08T11:16:40Z","cross_cats_sorted":[],"title_canon_sha256":"6b8225d65541a0c737c88378358b7d5ee437b5496515c0f6257d45f335279765","abstract_canon_sha256":"272cccac106c6881dc3d9c8908e9388f678cc3090dc80116987cd7eb941cc6f5"},"schema_version":"1.0"},"canonical_sha256":"bbd8f75cd94e7134d62087d55dc7357dd007b0819d29a41ead0cc815ec31cd54","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:04:05.287217Z","signature_b64":"xuv699h869Zuz9ASxOaNQxoQXbdU/DknxKCztVcQWGe1FV7VKbdHwhZHhb3386bh6Rmwx2OP7y9pKuEoFAZrAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bbd8f75cd94e7134d62087d55dc7357dd007b0819d29a41ead0cc815ec31cd54","last_reissued_at":"2026-05-22T01:04:05.286712Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:04:05.286712Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.07598","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":"2026-05-22T01:04:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"se5ZhG2IZseqWuXoVcYkiJ5vogvyERs6sXmZrVibxk4Q33TCXYbvozpwNLdCoFEcg8gr8QjA5lfHq8k2X3BUBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T09:05:16.743429Z"},"content_sha256":"94c2926ef1b8e9023b87f33becd84fe9477fa79fd96ab7e45113e5a76bc438e5","schema_version":"1.0","event_id":"sha256:94c2926ef1b8e9023b87f33becd84fe9477fa79fd96ab7e45113e5a76bc438e5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:XPMPOXGZJZYTJVRAQ7KV3RZVPX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimal Recourse Summaries via Bi-Objective Decision Tree Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"SOGAR finds the complete Pareto front of recourse summaries by learning bi-objective decision trees that balance effectiveness against cost.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Athanasios Voulodimos, Giorgos Stamou, Ioannis Chatzis, Jason Liartis","submitted_at":"2026-05-08T11:16:40Z","abstract_excerpt":"Actionable Recourse provides individuals with actions they can take to change an unfavorable classifier outcome. While useful at the instance level, it is ill-suited for global auditing and bias detection, since aggregating local actions is costly and often inconsistent. Recourse Summaries address this limitation by partitioning the population and assigning one shared action per subgroup, enabling comparison across subgroups. Designing summaries involves a fundamental trade-off between recourse effectiveness and recourse cost, which existing methods do not adequately address. We introduce Summ"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"SOGAR enables post-hoc selection of the desired trade-off without retraining. Using shallow axis-parallel decision trees and sparse leaf actions, SOGAR produces stable, low-cost, and effective recourse summaries that outperform existing approaches across effectiveness and cost metrics.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That shallow axis-parallel decision trees combined with sparse leaf actions are sufficient to produce stable, globally effective recourse summaries without losing critical population structure or requiring more expressive models.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SOGAR learns Pareto-optimal recourse summaries by solving a bi-objective decision tree problem, yielding stable low-cost effective group actions that outperform prior methods on effectiveness and cost.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SOGAR finds the complete Pareto front of recourse summaries by learning bi-objective decision trees that balance effectiveness against cost.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"252bf22fb41c0f00f60d3c5bd1fb6fa425bddd347b207bf43621abae27d29eb2"},"source":{"id":"2605.07598","kind":"arxiv","version":2},"verdict":{"id":"2b762831-dd9e-4410-b73c-3560dacdfd17","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-11T02:13:41.818356Z","strongest_claim":"SOGAR enables post-hoc selection of the desired trade-off without retraining. Using shallow axis-parallel decision trees and sparse leaf actions, SOGAR produces stable, low-cost, and effective recourse summaries that outperform existing approaches across effectiveness and cost metrics.","one_line_summary":"SOGAR learns Pareto-optimal recourse summaries by solving a bi-objective decision tree problem, yielding stable low-cost effective group actions that outperform prior methods on effectiveness and cost.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That shallow axis-parallel decision trees combined with sparse leaf actions are sufficient to produce stable, globally effective recourse summaries without losing critical population structure or requiring more expressive models.","pith_extraction_headline":"SOGAR finds the complete Pareto front of recourse summaries by learning bi-objective decision trees that balance effectiveness against cost."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.07598/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T10:42:02.812861Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-20T05:38:38.525240Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T16:01:19.010675Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T11:40:51.106023Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"9bf8a8bdf8ba71a0341b7eb8291d0a803a60d5c094ef22feb104c9b3852d28e2"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"fb5254418f41f69e6ac5e4fa87e7e72d9b05713ac00322db69861d6900aa25c2"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"2b762831-dd9e-4410-b73c-3560dacdfd17"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:04:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XpkJNc+dPny4z7TmlQFanjY7zLz531c6eWKDlQOS2UIRWVRxwof3F/48bIsUEb2E5Vt4O19SlZEbVKB9EOSNBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T09:05:16.744058Z"},"content_sha256":"7903e2c24f47f0cabe571e083fa7eca44110b8b1a360210c64ae674eec0ae10a","schema_version":"1.0","event_id":"sha256:7903e2c24f47f0cabe571e083fa7eca44110b8b1a360210c64ae674eec0ae10a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XPMPOXGZJZYTJVRAQ7KV3RZVPX/bundle.json","state_url":"https://pith.science/pith/XPMPOXGZJZYTJVRAQ7KV3RZVPX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XPMPOXGZJZYTJVRAQ7KV3RZVPX/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":"2026-06-11T09:05:16Z","links":{"resolver":"https://pith.science/pith/XPMPOXGZJZYTJVRAQ7KV3RZVPX","bundle":"https://pith.science/pith/XPMPOXGZJZYTJVRAQ7KV3RZVPX/bundle.json","state":"https://pith.science/pith/XPMPOXGZJZYTJVRAQ7KV3RZVPX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XPMPOXGZJZYTJVRAQ7KV3RZVPX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XPMPOXGZJZYTJVRAQ7KV3RZVPX","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":"272cccac106c6881dc3d9c8908e9388f678cc3090dc80116987cd7eb941cc6f5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-08T11:16:40Z","title_canon_sha256":"6b8225d65541a0c737c88378358b7d5ee437b5496515c0f6257d45f335279765"},"schema_version":"1.0","source":{"id":"2605.07598","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.07598","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"arxiv_version","alias_value":"2605.07598v2","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.07598","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"pith_short_12","alias_value":"XPMPOXGZJZYT","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"pith_short_16","alias_value":"XPMPOXGZJZYTJVRA","created_at":"2026-05-22T01:04:05Z"},{"alias_kind":"pith_short_8","alias_value":"XPMPOXGZ","created_at":"2026-05-22T01:04:05Z"}],"graph_snapshots":[{"event_id":"sha256:7903e2c24f47f0cabe571e083fa7eca44110b8b1a360210c64ae674eec0ae10a","target":"graph","created_at":"2026-05-22T01:04:05Z","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":"SOGAR enables post-hoc selection of the desired trade-off without retraining. Using shallow axis-parallel decision trees and sparse leaf actions, SOGAR produces stable, low-cost, and effective recourse summaries that outperform existing approaches across effectiveness and cost metrics."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That shallow axis-parallel decision trees combined with sparse leaf actions are sufficient to produce stable, globally effective recourse summaries without losing critical population structure or requiring more expressive models."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SOGAR learns Pareto-optimal recourse summaries by solving a bi-objective decision tree problem, yielding stable low-cost effective group actions that outperform prior methods on effectiveness and cost."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"SOGAR finds the complete Pareto front of recourse summaries by learning bi-objective decision trees that balance effectiveness against cost."}],"snapshot_sha256":"252bf22fb41c0f00f60d3c5bd1fb6fa425bddd347b207bf43621abae27d29eb2"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"fb5254418f41f69e6ac5e4fa87e7e72d9b05713ac00322db69861d6900aa25c2"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T10:42:02.812861Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T05:38:38.525240Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T16:01:19.010675Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T11:40:51.106023Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.07598/integrity.json","findings":[],"snapshot_sha256":"9bf8a8bdf8ba71a0341b7eb8291d0a803a60d5c094ef22feb104c9b3852d28e2","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Actionable Recourse provides individuals with actions they can take to change an unfavorable classifier outcome. While useful at the instance level, it is ill-suited for global auditing and bias detection, since aggregating local actions is costly and often inconsistent. Recourse Summaries address this limitation by partitioning the population and assigning one shared action per subgroup, enabling comparison across subgroups. Designing summaries involves a fundamental trade-off between recourse effectiveness and recourse cost, which existing methods do not adequately address. We introduce Summ","authors_text":"Athanasios Voulodimos, Giorgos Stamou, Ioannis Chatzis, Jason Liartis","cross_cats":[],"headline":"SOGAR finds the complete Pareto front of recourse summaries by learning bi-objective decision trees that balance effectiveness against cost.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-08T11:16:40Z","title":"Optimal Recourse Summaries via Bi-Objective Decision Tree Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.07598","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-11T02:13:41.818356Z","id":"2b762831-dd9e-4410-b73c-3560dacdfd17","model_set":{"reader":"grok-4.3"},"one_line_summary":"SOGAR learns Pareto-optimal recourse summaries by solving a bi-objective decision tree problem, yielding stable low-cost effective group actions that outperform prior methods on effectiveness and cost.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"SOGAR finds the complete Pareto front of recourse summaries by learning bi-objective decision trees that balance effectiveness against cost.","strongest_claim":"SOGAR enables post-hoc selection of the desired trade-off without retraining. Using shallow axis-parallel decision trees and sparse leaf actions, SOGAR produces stable, low-cost, and effective recourse summaries that outperform existing approaches across effectiveness and cost metrics.","weakest_assumption":"That shallow axis-parallel decision trees combined with sparse leaf actions are sufficient to produce stable, globally effective recourse summaries without losing critical population structure or requiring more expressive models."}},"verdict_id":"2b762831-dd9e-4410-b73c-3560dacdfd17"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:94c2926ef1b8e9023b87f33becd84fe9477fa79fd96ab7e45113e5a76bc438e5","target":"record","created_at":"2026-05-22T01:04:05Z","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":"272cccac106c6881dc3d9c8908e9388f678cc3090dc80116987cd7eb941cc6f5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-08T11:16:40Z","title_canon_sha256":"6b8225d65541a0c737c88378358b7d5ee437b5496515c0f6257d45f335279765"},"schema_version":"1.0","source":{"id":"2605.07598","kind":"arxiv","version":2}},"canonical_sha256":"bbd8f75cd94e7134d62087d55dc7357dd007b0819d29a41ead0cc815ec31cd54","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bbd8f75cd94e7134d62087d55dc7357dd007b0819d29a41ead0cc815ec31cd54","first_computed_at":"2026-05-22T01:04:05.286712Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:04:05.286712Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xuv699h869Zuz9ASxOaNQxoQXbdU/DknxKCztVcQWGe1FV7VKbdHwhZHhb3386bh6Rmwx2OP7y9pKuEoFAZrAg==","signature_status":"signed_v1","signed_at":"2026-05-22T01:04:05.287217Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.07598","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:94c2926ef1b8e9023b87f33becd84fe9477fa79fd96ab7e45113e5a76bc438e5","sha256:7903e2c24f47f0cabe571e083fa7eca44110b8b1a360210c64ae674eec0ae10a"],"state_sha256":"67466d380f1ca2c48d386f7b20810bfcf12a43fdf5170fd7a093fac98ed3bbee"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1C15vyvtHtKXY+pqryetl0G/ZyR17fYOHDGPVneQie4OEb6I4Y6lL6cgGgEJqolWQIyDkoOwr08AJ1MJV3yjCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-11T09:05:16.747141Z","bundle_sha256":"8c2ce682f56297824decadcf3c0375368a658cb2daecdd540649aa422ccaaf58"}}