{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:Q5TOTEGSXLK7KP55335M2BRXGT","short_pith_number":"pith:Q5TOTEGS","canonical_record":{"source":{"id":"2606.18203","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T17:34:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a399c4c4cb289cb6c511ee71b1b95bf3504fe8a010a01e9d890ae0b3c128cda7","abstract_canon_sha256":"bbab985425140ae83492af502ced2efd82860273092561dd6852847ad80e4926"},"schema_version":"1.0"},"canonical_sha256":"8766e990d2bad5f53fbddefacd063734e8375fdd92d745e00abf70899339c309","source":{"kind":"arxiv","id":"2606.18203","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.18203","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"arxiv_version","alias_value":"2606.18203v1","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.18203","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"pith_short_12","alias_value":"Q5TOTEGSXLK7","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"pith_short_16","alias_value":"Q5TOTEGSXLK7KP55","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"pith_short_8","alias_value":"Q5TOTEGS","created_at":"2026-06-19T16:10:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:Q5TOTEGSXLK7KP55335M2BRXGT","target":"record","payload":{"canonical_record":{"source":{"id":"2606.18203","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T17:34:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a399c4c4cb289cb6c511ee71b1b95bf3504fe8a010a01e9d890ae0b3c128cda7","abstract_canon_sha256":"bbab985425140ae83492af502ced2efd82860273092561dd6852847ad80e4926"},"schema_version":"1.0"},"canonical_sha256":"8766e990d2bad5f53fbddefacd063734e8375fdd92d745e00abf70899339c309","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:10:50.902770Z","signature_b64":"gTsbRqRgZS2+9v+T04flw6dM7iGuZx7tN2okj0FANA0/F5ppZaVdXTJbCodYwY40ixGSqSR1LOL4IKn8BEVFBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8766e990d2bad5f53fbddefacd063734e8375fdd92d745e00abf70899339c309","last_reissued_at":"2026-06-19T16:10:50.902430Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:10:50.902430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.18203","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-19T16:10:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NxL7Is3fpYRuJzUh5+Q6vxcKe+5LGkgOruQUu9l0Eem3lKou+vU7YJ4y4JSF1TSd5KkNvsPq6Ll7SoEQk0GXAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T03:40:43.762261Z"},"content_sha256":"cda09e78c8207a953482d42441705a2e36476dcf07c1ef2661c8369908eb7ce5","schema_version":"1.0","event_id":"sha256:cda09e78c8207a953482d42441705a2e36476dcf07c1ef2661c8369908eb7ce5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:Q5TOTEGSXLK7KP55335M2BRXGT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RubricsTree: Scalable and Evolving Open-Ended Evaluation of Personal Health Agents across Health Memory and Medical Skills","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"A. Ali Heydari, Ahmed A. Metwally, Ben Graef, Chloe Zhang, Daniel McDuff, Erik Schenck, Hamid Palangi, Lindsey Sunden, Mark Malhotra, Menglian Zhou, Philip S. Yu, Ray Luo, Salman Rahman, Shwetak Patel, Simon A. Lee, Weizhi Zhang, Yamin Li, Zechen Li, Zeinab Esmaeilpour","submitted_at":"2026-06-16T17:34:53Z","abstract_excerpt":"The LLM-empowered personal health agents with user health (sensor) metrics have offered a promising pathway to alleviate global disparities in healthcare access. However, large-scale clinical deployment remains constrained by an open-ended evaluation bottleneck: physician annotation is reliable but costly and unscalable, while LLM-as-a-judge evaluators are scalable but subjective, inconsistent, and sometimes clinically misaligned. We introduce RubricsTree, a scalable evaluation framework with an expert-aligned hierarchical taxonomy of over 100 atomic, clinically-verifiable Boolean rubrics, evo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.18203","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/2606.18203/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":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-19T16:10:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7Uxtn4zT9lekxURgDOQR24gMyoAr2fMSIeCWOgf31T1QatvU/igWt6eBc5+LSbTawGLRg+QEk+cd430JdvelAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T03:40:43.762646Z"},"content_sha256":"e0ee86466a10b6255416ce6d25c66b08ad72d57b3ee15100af649e1b076ccb71","schema_version":"1.0","event_id":"sha256:e0ee86466a10b6255416ce6d25c66b08ad72d57b3ee15100af649e1b076ccb71"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q5TOTEGSXLK7KP55335M2BRXGT/bundle.json","state_url":"https://pith.science/pith/Q5TOTEGSXLK7KP55335M2BRXGT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q5TOTEGSXLK7KP55335M2BRXGT/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":"2026-06-30T03:40:43Z","links":{"resolver":"https://pith.science/pith/Q5TOTEGSXLK7KP55335M2BRXGT","bundle":"https://pith.science/pith/Q5TOTEGSXLK7KP55335M2BRXGT/bundle.json","state":"https://pith.science/pith/Q5TOTEGSXLK7KP55335M2BRXGT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q5TOTEGSXLK7KP55335M2BRXGT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:Q5TOTEGSXLK7KP55335M2BRXGT","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":"bbab985425140ae83492af502ced2efd82860273092561dd6852847ad80e4926","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T17:34:53Z","title_canon_sha256":"a399c4c4cb289cb6c511ee71b1b95bf3504fe8a010a01e9d890ae0b3c128cda7"},"schema_version":"1.0","source":{"id":"2606.18203","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.18203","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"arxiv_version","alias_value":"2606.18203v1","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.18203","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"pith_short_12","alias_value":"Q5TOTEGSXLK7","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"pith_short_16","alias_value":"Q5TOTEGSXLK7KP55","created_at":"2026-06-19T16:10:50Z"},{"alias_kind":"pith_short_8","alias_value":"Q5TOTEGS","created_at":"2026-06-19T16:10:50Z"}],"graph_snapshots":[{"event_id":"sha256:e0ee86466a10b6255416ce6d25c66b08ad72d57b3ee15100af649e1b076ccb71","target":"graph","created_at":"2026-06-19T16:10:50Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.18203/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The LLM-empowered personal health agents with user health (sensor) metrics have offered a promising pathway to alleviate global disparities in healthcare access. However, large-scale clinical deployment remains constrained by an open-ended evaluation bottleneck: physician annotation is reliable but costly and unscalable, while LLM-as-a-judge evaluators are scalable but subjective, inconsistent, and sometimes clinically misaligned. We introduce RubricsTree, a scalable evaluation framework with an expert-aligned hierarchical taxonomy of over 100 atomic, clinically-verifiable Boolean rubrics, evo","authors_text":"A. Ali Heydari, Ahmed A. Metwally, Ben Graef, Chloe Zhang, Daniel McDuff, Erik Schenck, Hamid Palangi, Lindsey Sunden, Mark Malhotra, Menglian Zhou, Philip S. Yu, Ray Luo, Salman Rahman, Shwetak Patel, Simon A. Lee, Weizhi Zhang, Yamin Li, Zechen Li, Zeinab Esmaeilpour","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T17:34:53Z","title":"RubricsTree: Scalable and Evolving Open-Ended Evaluation of Personal Health Agents across Health Memory and Medical Skills"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.18203","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:cda09e78c8207a953482d42441705a2e36476dcf07c1ef2661c8369908eb7ce5","target":"record","created_at":"2026-06-19T16:10:50Z","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":"bbab985425140ae83492af502ced2efd82860273092561dd6852847ad80e4926","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T17:34:53Z","title_canon_sha256":"a399c4c4cb289cb6c511ee71b1b95bf3504fe8a010a01e9d890ae0b3c128cda7"},"schema_version":"1.0","source":{"id":"2606.18203","kind":"arxiv","version":1}},"canonical_sha256":"8766e990d2bad5f53fbddefacd063734e8375fdd92d745e00abf70899339c309","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8766e990d2bad5f53fbddefacd063734e8375fdd92d745e00abf70899339c309","first_computed_at":"2026-06-19T16:10:50.902430Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:10:50.902430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gTsbRqRgZS2+9v+T04flw6dM7iGuZx7tN2okj0FANA0/F5ppZaVdXTJbCodYwY40ixGSqSR1LOL4IKn8BEVFBg==","signature_status":"signed_v1","signed_at":"2026-06-19T16:10:50.902770Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.18203","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cda09e78c8207a953482d42441705a2e36476dcf07c1ef2661c8369908eb7ce5","sha256:e0ee86466a10b6255416ce6d25c66b08ad72d57b3ee15100af649e1b076ccb71"],"state_sha256":"5a573b7f058845d2fd4ba16f6e4139f2cc300d0c15c7bbfc9c24b50b60ddea0f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zugzT1VNbuvx5UicmeEWABqlHGRcmk5VJln3XGkLX5axpN+gXwwDkN+v/Bxb9EB/Fm391q8MTYRxsOxerQyHDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T03:40:43.765900Z","bundle_sha256":"c8c57b5a607559c393dbcce00bf4eb8aad48ebddb22e1009fe0284b270c170f4"}}