{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LON2GVRIXAQ44S7QH3TN32ER2I","short_pith_number":"pith:LON2GVRI","canonical_record":{"source":{"id":"2506.14927","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-17T19:14:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d8325cc707cfb10761da8386cdb37fc4c64c46c620ceab7ca37a7be3489a03dc","abstract_canon_sha256":"540727a0b97031d73821c64815921d3ae1d7f29d1843354359123f584b7d08bb"},"schema_version":"1.0"},"canonical_sha256":"5b9ba35628b821ce4bf03ee6dde891d22b0b81080329ff82daa2b56a499b07d4","source":{"kind":"arxiv","id":"2506.14927","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.14927","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"arxiv_version","alias_value":"2506.14927v1","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.14927","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"pith_short_12","alias_value":"LON2GVRIXAQ4","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"pith_short_16","alias_value":"LON2GVRIXAQ44S7Q","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"pith_short_8","alias_value":"LON2GVRI","created_at":"2026-07-05T11:23:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LON2GVRIXAQ44S7QH3TN32ER2I","target":"record","payload":{"canonical_record":{"source":{"id":"2506.14927","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-17T19:14:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d8325cc707cfb10761da8386cdb37fc4c64c46c620ceab7ca37a7be3489a03dc","abstract_canon_sha256":"540727a0b97031d73821c64815921d3ae1d7f29d1843354359123f584b7d08bb"},"schema_version":"1.0"},"canonical_sha256":"5b9ba35628b821ce4bf03ee6dde891d22b0b81080329ff82daa2b56a499b07d4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:23:34.208248Z","signature_b64":"rHY5yLhpap+Xi2O9c8s46rCyIhKEbjnFBRJAIuphb5Z8xpwFBvEbjx3PwQK5KIU/Yl6rXUk80gwB+x23iDPEBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b9ba35628b821ce4bf03ee6dde891d22b0b81080329ff82daa2b56a499b07d4","last_reissued_at":"2026-07-05T11:23:34.207759Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:23:34.207759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.14927","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":"2026-07-05T11:23:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9ArvfMyp9nPx/Zl5KWZiF73nZqEmdYDZeQZXj6cIZmgIlfldoMGcG/Aj3If7lA957s+EvTTGklkTRw+wDuG/AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:42:14.121530Z"},"content_sha256":"bede7de861897061dbe82bd13caad85df5a1636e0a6819118b75727c771faabb","schema_version":"1.0","event_id":"sha256:bede7de861897061dbe82bd13caad85df5a1636e0a6819118b75727c771faabb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LON2GVRIXAQ44S7QH3TN32ER2I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MDBench: A Synthetic Multi-Document Reasoning Benchmark Generated with Knowledge Guidance","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ali Payani, Joseph J. Peper, Lu Wang, Wenzhao Qiu","submitted_at":"2025-06-17T19:14:30Z","abstract_excerpt":"Natural language processing evaluation has made significant progress, largely driven by the proliferation of powerful large language mod-els (LLMs). New evaluation benchmarks are of increasing priority as the reasoning capabilities of LLMs are expanding at a rapid pace. In particular, while multi-document (MD) reasoning is an area of extreme relevance given LLM capabilities in handling longer-context inputs, few benchmarks exist to rigorously examine model behavior in this setting. Moreover, the multi-document setting is historically challenging for benchmark creation due to the expensive cost"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.14927","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/2506.14927/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":"2026-07-05T11:23:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dxrggBZeQxT1HpOQr8zM5V3IWn/3YT6o1xccVXGXmdUOfRBkXJYZnicpeZ0Hj6Bp7yR2m6XeWlvI+bEnynlbDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:42:14.122086Z"},"content_sha256":"8a1ab796063f6c88339c5727e504828882b6211088b7340848994fea40d24bdc","schema_version":"1.0","event_id":"sha256:8a1ab796063f6c88339c5727e504828882b6211088b7340848994fea40d24bdc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LON2GVRIXAQ44S7QH3TN32ER2I/bundle.json","state_url":"https://pith.science/pith/LON2GVRIXAQ44S7QH3TN32ER2I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LON2GVRIXAQ44S7QH3TN32ER2I/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":"2026-08-03T21:42:14Z","links":{"resolver":"https://pith.science/pith/LON2GVRIXAQ44S7QH3TN32ER2I","bundle":"https://pith.science/pith/LON2GVRIXAQ44S7QH3TN32ER2I/bundle.json","state":"https://pith.science/pith/LON2GVRIXAQ44S7QH3TN32ER2I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LON2GVRIXAQ44S7QH3TN32ER2I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LON2GVRIXAQ44S7QH3TN32ER2I","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":"540727a0b97031d73821c64815921d3ae1d7f29d1843354359123f584b7d08bb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-17T19:14:30Z","title_canon_sha256":"d8325cc707cfb10761da8386cdb37fc4c64c46c620ceab7ca37a7be3489a03dc"},"schema_version":"1.0","source":{"id":"2506.14927","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.14927","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"arxiv_version","alias_value":"2506.14927v1","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.14927","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"pith_short_12","alias_value":"LON2GVRIXAQ4","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"pith_short_16","alias_value":"LON2GVRIXAQ44S7Q","created_at":"2026-07-05T11:23:34Z"},{"alias_kind":"pith_short_8","alias_value":"LON2GVRI","created_at":"2026-07-05T11:23:34Z"}],"graph_snapshots":[{"event_id":"sha256:8a1ab796063f6c88339c5727e504828882b6211088b7340848994fea40d24bdc","target":"graph","created_at":"2026-07-05T11:23:34Z","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/2506.14927/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Natural language processing evaluation has made significant progress, largely driven by the proliferation of powerful large language mod-els (LLMs). New evaluation benchmarks are of increasing priority as the reasoning capabilities of LLMs are expanding at a rapid pace. In particular, while multi-document (MD) reasoning is an area of extreme relevance given LLM capabilities in handling longer-context inputs, few benchmarks exist to rigorously examine model behavior in this setting. Moreover, the multi-document setting is historically challenging for benchmark creation due to the expensive cost","authors_text":"Ali Payani, Joseph J. Peper, Lu Wang, Wenzhao Qiu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-17T19:14:30Z","title":"MDBench: A Synthetic Multi-Document Reasoning Benchmark Generated with Knowledge Guidance"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.14927","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:bede7de861897061dbe82bd13caad85df5a1636e0a6819118b75727c771faabb","target":"record","created_at":"2026-07-05T11:23:34Z","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":"540727a0b97031d73821c64815921d3ae1d7f29d1843354359123f584b7d08bb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-17T19:14:30Z","title_canon_sha256":"d8325cc707cfb10761da8386cdb37fc4c64c46c620ceab7ca37a7be3489a03dc"},"schema_version":"1.0","source":{"id":"2506.14927","kind":"arxiv","version":1}},"canonical_sha256":"5b9ba35628b821ce4bf03ee6dde891d22b0b81080329ff82daa2b56a499b07d4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5b9ba35628b821ce4bf03ee6dde891d22b0b81080329ff82daa2b56a499b07d4","first_computed_at":"2026-07-05T11:23:34.207759Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:23:34.207759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rHY5yLhpap+Xi2O9c8s46rCyIhKEbjnFBRJAIuphb5Z8xpwFBvEbjx3PwQK5KIU/Yl6rXUk80gwB+x23iDPEBA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:23:34.208248Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.14927","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bede7de861897061dbe82bd13caad85df5a1636e0a6819118b75727c771faabb","sha256:8a1ab796063f6c88339c5727e504828882b6211088b7340848994fea40d24bdc"],"state_sha256":"b59db58ce4bf9681d3b9a12d5aa4c86078cd7a9336b7b7072e6355bda7c232ce"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YsG6xLaw8Q/SDMVjey1Ayrg9BLnpWVbVeEdqzYxBPEn4COlAo4xhARRjPLmwywTpZCwMhrPUbeig40qD+RzhAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T21:42:14.126072Z","bundle_sha256":"2b34872fc39256e86d8c2ebbc2ea35edab89068f2b983db34751d9b1d9659c01"}}