{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XJ5LJAZE3UDF6HMGLEDQCHZCRP","short_pith_number":"pith:XJ5LJAZE","canonical_record":{"source":{"id":"2505.14107","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-20T09:14:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b99ad760c0f955d71df028a691400db6816877b6d2b7b6682f807e5955798b94","abstract_canon_sha256":"677204a5dfc40893af5b10ddff72f6ab57d3043198c8caa6fa9654c09a7c763c"},"schema_version":"1.0"},"canonical_sha256":"ba7ab48324dd065f1d865907011f228bf21acab4181b36537e961394035c34ab","source":{"kind":"arxiv","id":"2505.14107","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.14107","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"arxiv_version","alias_value":"2505.14107v4","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14107","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"pith_short_12","alias_value":"XJ5LJAZE3UDF","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"pith_short_16","alias_value":"XJ5LJAZE3UDF6HMG","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"pith_short_8","alias_value":"XJ5LJAZE","created_at":"2026-07-05T11:11:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XJ5LJAZE3UDF6HMGLEDQCHZCRP","target":"record","payload":{"canonical_record":{"source":{"id":"2505.14107","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-20T09:14:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b99ad760c0f955d71df028a691400db6816877b6d2b7b6682f807e5955798b94","abstract_canon_sha256":"677204a5dfc40893af5b10ddff72f6ab57d3043198c8caa6fa9654c09a7c763c"},"schema_version":"1.0"},"canonical_sha256":"ba7ab48324dd065f1d865907011f228bf21acab4181b36537e961394035c34ab","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:45.773231Z","signature_b64":"mtPCsXP/GYn/NEJGYcHrH0iv5u5FQduslDm2NJXdWQuyNmsusz2XdcmWxnDKVy+nVpVL5apTGUFBoAp/R2oWAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ba7ab48324dd065f1d865907011f228bf21acab4181b36537e961394035c34ab","last_reissued_at":"2026-07-05T11:11:45.772747Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:45.772747Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.14107","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:11:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jJL/ENoQN7y8j8MLs1xFCce/CDoE/JWdQGiUwwEf+oO8SAgJGbZXUV1lDh3yjtHtBCxlT1EuNaUxL3OMgX0pCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:09:26.995752Z"},"content_sha256":"a25b325a59c6072de8a4dbfc4d7c34bf46a846f7903a3806aeb40ae68917467d","schema_version":"1.0","event_id":"sha256:a25b325a59c6072de8a4dbfc4d7c34bf46a846f7903a3806aeb40ae68917467d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XJ5LJAZE3UDF6HMGLEDQCHZCRP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DiagnosisArena: Benchmarking Diagnostic Reasoning for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jiaji Liu, Linjie Mu, Pengfei Liu, Shaoting Zhang, Wei Nie, Xiaofan Zhang, Yakun Zhu, Yutong Huang, Zhongzhen Huang","submitted_at":"2025-05-20T09:14:53Z","abstract_excerpt":"The emergence of groundbreaking large language models capable of performing complex reasoning tasks holds significant promise for addressing various scientific challenges, including those arising in complex clinical scenarios. To enable their safe and effective deployment in real-world healthcare settings, it is urgently necessary to benchmark the diagnostic capabilities of current models systematically. Given the limitations of existing medical benchmarks in evaluating advanced diagnostic reasoning, we present DiagnosisArena, a comprehensive and challenging benchmark designed to rigorously as"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14107","kind":"arxiv","version":4},"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/2505.14107/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":"2026-07-05T11:11:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZxWhoQpwspBUJrxrpJmIGEn8JAfVb2HNKNwxc+MZSMSPCvcXUKDJhnAx0oxroToZFRaPF59a/DwTOSvqthNiCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:09:26.996249Z"},"content_sha256":"c27ecbd666230f9ff949301c3e1438a6042c282308019b143ef107f54085b5f9","schema_version":"1.0","event_id":"sha256:c27ecbd666230f9ff949301c3e1438a6042c282308019b143ef107f54085b5f9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XJ5LJAZE3UDF6HMGLEDQCHZCRP/bundle.json","state_url":"https://pith.science/pith/XJ5LJAZE3UDF6HMGLEDQCHZCRP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XJ5LJAZE3UDF6HMGLEDQCHZCRP/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":"2026-08-08T22:09:26Z","links":{"resolver":"https://pith.science/pith/XJ5LJAZE3UDF6HMGLEDQCHZCRP","bundle":"https://pith.science/pith/XJ5LJAZE3UDF6HMGLEDQCHZCRP/bundle.json","state":"https://pith.science/pith/XJ5LJAZE3UDF6HMGLEDQCHZCRP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XJ5LJAZE3UDF6HMGLEDQCHZCRP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XJ5LJAZE3UDF6HMGLEDQCHZCRP","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":"677204a5dfc40893af5b10ddff72f6ab57d3043198c8caa6fa9654c09a7c763c","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-20T09:14:53Z","title_canon_sha256":"b99ad760c0f955d71df028a691400db6816877b6d2b7b6682f807e5955798b94"},"schema_version":"1.0","source":{"id":"2505.14107","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.14107","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"arxiv_version","alias_value":"2505.14107v4","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14107","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"pith_short_12","alias_value":"XJ5LJAZE3UDF","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"pith_short_16","alias_value":"XJ5LJAZE3UDF6HMG","created_at":"2026-07-05T11:11:45Z"},{"alias_kind":"pith_short_8","alias_value":"XJ5LJAZE","created_at":"2026-07-05T11:11:45Z"}],"graph_snapshots":[{"event_id":"sha256:c27ecbd666230f9ff949301c3e1438a6042c282308019b143ef107f54085b5f9","target":"graph","created_at":"2026-07-05T11:11:45Z","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/2505.14107/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The emergence of groundbreaking large language models capable of performing complex reasoning tasks holds significant promise for addressing various scientific challenges, including those arising in complex clinical scenarios. To enable their safe and effective deployment in real-world healthcare settings, it is urgently necessary to benchmark the diagnostic capabilities of current models systematically. Given the limitations of existing medical benchmarks in evaluating advanced diagnostic reasoning, we present DiagnosisArena, a comprehensive and challenging benchmark designed to rigorously as","authors_text":"Jiaji Liu, Linjie Mu, Pengfei Liu, Shaoting Zhang, Wei Nie, Xiaofan Zhang, Yakun Zhu, Yutong Huang, Zhongzhen Huang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-20T09:14:53Z","title":"DiagnosisArena: Benchmarking Diagnostic Reasoning for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14107","kind":"arxiv","version":4},"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:a25b325a59c6072de8a4dbfc4d7c34bf46a846f7903a3806aeb40ae68917467d","target":"record","created_at":"2026-07-05T11:11:45Z","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":"677204a5dfc40893af5b10ddff72f6ab57d3043198c8caa6fa9654c09a7c763c","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-20T09:14:53Z","title_canon_sha256":"b99ad760c0f955d71df028a691400db6816877b6d2b7b6682f807e5955798b94"},"schema_version":"1.0","source":{"id":"2505.14107","kind":"arxiv","version":4}},"canonical_sha256":"ba7ab48324dd065f1d865907011f228bf21acab4181b36537e961394035c34ab","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ba7ab48324dd065f1d865907011f228bf21acab4181b36537e961394035c34ab","first_computed_at":"2026-07-05T11:11:45.772747Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:11:45.772747Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mtPCsXP/GYn/NEJGYcHrH0iv5u5FQduslDm2NJXdWQuyNmsusz2XdcmWxnDKVy+nVpVL5apTGUFBoAp/R2oWAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:11:45.773231Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.14107","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a25b325a59c6072de8a4dbfc4d7c34bf46a846f7903a3806aeb40ae68917467d","sha256:c27ecbd666230f9ff949301c3e1438a6042c282308019b143ef107f54085b5f9"],"state_sha256":"e459bce80017af01afba447939f813aa8ab9f1ce9114627fd83968e5d57a7b99"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6eKvEocuYRMJt1PgQHGmLfW6sczVHkP+saNPqub42YUjICfTfrue4oyRLhkJVXuh9DqmPyz3eVszBjm4yhENDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T22:09:27.000053Z","bundle_sha256":"d2722b9edce4906a0e53e946379c20d319a832a3a67b53c888de2e8ca4a5bad8"}}