{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:C7DXZZU4JU7EMS3BEQSXNLKTDI","short_pith_number":"pith:C7DXZZU4","canonical_record":{"source":{"id":"2512.07795","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-12-08T18:26:58Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"165d1bf1b3fa004c0bfcbf4c0c9bad239d4d43ededf458d51e54bf072fa54a85","abstract_canon_sha256":"2593b6b96b69c4a888bf8bf8e1270da48593abe399ec87af47aef2379eefa4de"},"schema_version":"1.0"},"canonical_sha256":"17c77ce69c4d3e464b61242576ad531a1dac74542229d70b63e4b3d5d8334291","source":{"kind":"arxiv","id":"2512.07795","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.07795","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"arxiv_version","alias_value":"2512.07795v2","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.07795","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"pith_short_12","alias_value":"C7DXZZU4JU7E","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"pith_short_16","alias_value":"C7DXZZU4JU7EMS3B","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"pith_short_8","alias_value":"C7DXZZU4","created_at":"2026-06-02T01:03:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:C7DXZZU4JU7EMS3BEQSXNLKTDI","target":"record","payload":{"canonical_record":{"source":{"id":"2512.07795","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-12-08T18:26:58Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"165d1bf1b3fa004c0bfcbf4c0c9bad239d4d43ededf458d51e54bf072fa54a85","abstract_canon_sha256":"2593b6b96b69c4a888bf8bf8e1270da48593abe399ec87af47aef2379eefa4de"},"schema_version":"1.0"},"canonical_sha256":"17c77ce69c4d3e464b61242576ad531a1dac74542229d70b63e4b3d5d8334291","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:39.371755Z","signature_b64":"na7Hrq6X6Hb0Vx/lWVzgJuBWXmd6Dbm/KVp1jw0TuvOFIrdHuYcW/1ca69h/UtIL7z216imb4RQ4bTNVP3+8Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"17c77ce69c4d3e464b61242576ad531a1dac74542229d70b63e4b3d5d8334291","last_reissued_at":"2026-06-02T01:03:39.371272Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:39.371272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2512.07795","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":"2026-06-02T01:03:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ytKXPYCAFr3EpArh6MXz1/5m20/6yeF8og/mSkvfnHEAzU0vk4C6hxkw4aicq2tg0ssnfbv7cX3clWcGl3CGAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T06:42:29.260828Z"},"content_sha256":"6ef3ea22e004c7e368ad87eaf3cd72ee12980e9a85790b7dee4aeb21dcd58473","schema_version":"1.0","event_id":"sha256:6ef3ea22e004c7e368ad87eaf3cd72ee12980e9a85790b7dee4aeb21dcd58473"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:C7DXZZU4JU7EMS3BEQSXNLKTDI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ReasonBENCH: Benchmarking the (In)Stability of LLM Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Akhil Arora, Dhairya Kuchhal, Har Ashish Arora, Lars Klein, Nearchos Potamitis, Vansh Ramani","submitted_at":"2025-12-08T18:26:58Z","abstract_excerpt":"Benchmark scores for LLM reasoning systems are reported as single numbers, yet the same model, strategy, and task can produce meaningfully different answers and costs across repeated executions, even under greedy decoding (T = 0). This variance is not a statistical nuisance: the highest-performing strategy wins only 77% of head-to-head runs against its nearest competitor, meaning a single observed score can silently misrank systems. We introduce ReasonBench, a benchmark suite recording 30 independent trials across 10 reasoning strategies, 12 models, and 6 tasks, treating quality and cost as di"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.07795","kind":"arxiv","version":2},"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/2512.07795/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":"2026-06-02T01:03:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yvw9sYUk3tDbHVrTnnhELINGzDkKaTE8YWmhQTAVaP5sL1pENzC5k+dYmFGhB1DZyJIfw4YOr3Qrzsm68gTjCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T06:42:29.261655Z"},"content_sha256":"1749e6b9b95556d9d22e3db7fc3f45c1fc54753f161d0a81babed8986344877f","schema_version":"1.0","event_id":"sha256:1749e6b9b95556d9d22e3db7fc3f45c1fc54753f161d0a81babed8986344877f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/C7DXZZU4JU7EMS3BEQSXNLKTDI/bundle.json","state_url":"https://pith.science/pith/C7DXZZU4JU7EMS3BEQSXNLKTDI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/C7DXZZU4JU7EMS3BEQSXNLKTDI/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":"2026-06-10T06:42:29Z","links":{"resolver":"https://pith.science/pith/C7DXZZU4JU7EMS3BEQSXNLKTDI","bundle":"https://pith.science/pith/C7DXZZU4JU7EMS3BEQSXNLKTDI/bundle.json","state":"https://pith.science/pith/C7DXZZU4JU7EMS3BEQSXNLKTDI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/C7DXZZU4JU7EMS3BEQSXNLKTDI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:C7DXZZU4JU7EMS3BEQSXNLKTDI","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":"2593b6b96b69c4a888bf8bf8e1270da48593abe399ec87af47aef2379eefa4de","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-12-08T18:26:58Z","title_canon_sha256":"165d1bf1b3fa004c0bfcbf4c0c9bad239d4d43ededf458d51e54bf072fa54a85"},"schema_version":"1.0","source":{"id":"2512.07795","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.07795","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"arxiv_version","alias_value":"2512.07795v2","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.07795","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"pith_short_12","alias_value":"C7DXZZU4JU7E","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"pith_short_16","alias_value":"C7DXZZU4JU7EMS3B","created_at":"2026-06-02T01:03:39Z"},{"alias_kind":"pith_short_8","alias_value":"C7DXZZU4","created_at":"2026-06-02T01:03:39Z"}],"graph_snapshots":[{"event_id":"sha256:1749e6b9b95556d9d22e3db7fc3f45c1fc54753f161d0a81babed8986344877f","target":"graph","created_at":"2026-06-02T01:03:39Z","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/2512.07795/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Benchmark scores for LLM reasoning systems are reported as single numbers, yet the same model, strategy, and task can produce meaningfully different answers and costs across repeated executions, even under greedy decoding (T = 0). This variance is not a statistical nuisance: the highest-performing strategy wins only 77% of head-to-head runs against its nearest competitor, meaning a single observed score can silently misrank systems. We introduce ReasonBench, a benchmark suite recording 30 independent trials across 10 reasoning strategies, 12 models, and 6 tasks, treating quality and cost as di","authors_text":"Akhil Arora, Dhairya Kuchhal, Har Ashish Arora, Lars Klein, Nearchos Potamitis, Vansh Ramani","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-12-08T18:26:58Z","title":"ReasonBENCH: Benchmarking the (In)Stability of LLM Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.07795","kind":"arxiv","version":2},"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":"sha256:6ef3ea22e004c7e368ad87eaf3cd72ee12980e9a85790b7dee4aeb21dcd58473","target":"record","created_at":"2026-06-02T01:03:39Z","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":"2593b6b96b69c4a888bf8bf8e1270da48593abe399ec87af47aef2379eefa4de","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-12-08T18:26:58Z","title_canon_sha256":"165d1bf1b3fa004c0bfcbf4c0c9bad239d4d43ededf458d51e54bf072fa54a85"},"schema_version":"1.0","source":{"id":"2512.07795","kind":"arxiv","version":2}},"canonical_sha256":"17c77ce69c4d3e464b61242576ad531a1dac74542229d70b63e4b3d5d8334291","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"17c77ce69c4d3e464b61242576ad531a1dac74542229d70b63e4b3d5d8334291","first_computed_at":"2026-06-02T01:03:39.371272Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:39.371272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"na7Hrq6X6Hb0Vx/lWVzgJuBWXmd6Dbm/KVp1jw0TuvOFIrdHuYcW/1ca69h/UtIL7z216imb4RQ4bTNVP3+8Bg==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:39.371755Z","signed_message":"canonical_sha256_bytes"},"source_id":"2512.07795","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6ef3ea22e004c7e368ad87eaf3cd72ee12980e9a85790b7dee4aeb21dcd58473","sha256:1749e6b9b95556d9d22e3db7fc3f45c1fc54753f161d0a81babed8986344877f"],"state_sha256":"806adeb48edfb0b11553a0fbd36b463ac0a8cab0a7e8a58dac027667f7643120"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eym8w7oF268CClD6fyNzjYxfDA77tEsbu7+r7UBDPBsPUt9ZwIs6BFjSXDoisLHJT/vvitHAdLDAf4E1JhJICA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T06:42:29.265602Z","bundle_sha256":"c5d8a606f31ff8648fd2b6499c81c6471dfdf1aa5f2692578d86fbfb0a8101c1"}}