{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:WQUAFJNJB5G7TEIMYM76JH6AQ6","short_pith_number":"pith:WQUAFJNJ","canonical_record":{"source":{"id":"2607.27405","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T19:22:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"35a467ba9f9302e6f297d5929ab5ac31912de134720f9c371770782709f3bddf","abstract_canon_sha256":"e57cffc6600d92366b1dba894bd0d9f95b29fa98b9f3c7dcbdd94be79b4d1c56"},"schema_version":"1.0"},"canonical_sha256":"b42802a5a90f4df9910cc33fe49fc087a2c66c7cfe15b8fdcbdf5050c6ab7fe9","source":{"kind":"arxiv","id":"2607.27405","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.27405","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"arxiv_version","alias_value":"2607.27405v1","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27405","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"pith_short_12","alias_value":"WQUAFJNJB5G7","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"pith_short_16","alias_value":"WQUAFJNJB5G7TEIM","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"pith_short_8","alias_value":"WQUAFJNJ","created_at":"2026-07-31T00:11:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:WQUAFJNJB5G7TEIMYM76JH6AQ6","target":"record","payload":{"canonical_record":{"source":{"id":"2607.27405","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T19:22:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"35a467ba9f9302e6f297d5929ab5ac31912de134720f9c371770782709f3bddf","abstract_canon_sha256":"e57cffc6600d92366b1dba894bd0d9f95b29fa98b9f3c7dcbdd94be79b4d1c56"},"schema_version":"1.0"},"canonical_sha256":"b42802a5a90f4df9910cc33fe49fc087a2c66c7cfe15b8fdcbdf5050c6ab7fe9","receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b42802a5a90f4df9910cc33fe49fc087a2c66c7cfe15b8fdcbdf5050c6ab7fe9","last_reissued_at":"2026-07-31T00:11:42.478921Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-31T00:11:42.478921Z"},"source_kind":"arxiv","source_id":"2607.27405","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":"2026-07-31T00:11:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H7c4jI+0j8WwAs/DYA4VNWmSM214biVsru56ilC5Y1jzLyXlDPrbor7Eb3Pe99arBreAHDuCsuiAJGRngII0BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T16:43:33.765527Z"},"content_sha256":"50e796f7eb8ed1cf2a12542b0fa10383f6377192da5b123546ccc5849e3ceff0","schema_version":"1.0","event_id":"sha256:50e796f7eb8ed1cf2a12542b0fa10383f6377192da5b123546ccc5849e3ceff0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:WQUAFJNJB5G7TEIMYM76JH6AQ6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Benchmarking LLM Competence on Logical Inference over Probability Operators","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alvin Grissom II, Jack Greff, Nayera Hasan","submitted_at":"2026-07-29T19:22:09Z","abstract_excerpt":"Both expressions of uncertainty and inferences are ubiquitous in natural language, and valid inferences over natural-language expressions of uncertainty are necessary for not only everyday conversations but also for high-stakes domains such as medicine and law. While large language models are increasingly evaluated on logical reasoning tasks, disentangling principled, symbolic reasoning from clever surface-level pattern matching is fraught with difficulty. We introduce a benchmark for reasoning over probability operators--inference over sentences with gradable epistemic modals (e.g., probably,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27405","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/2607.27405/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":"2026-07-31T00:11:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"INT28LNR5a8xrRH5VEodP5pYkxE6tIKNLY+5hH5Cx1J2I4/UNBZ1ar9AmkwxumckjlaImcbL3RgEeQKi0mpjCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T16:43:33.766044Z"},"content_sha256":"c9c57ba1b6bfb9d26f45ef8f98ea297271a073fc9db679a959e201374d05c094","schema_version":"1.0","event_id":"sha256:c9c57ba1b6bfb9d26f45ef8f98ea297271a073fc9db679a959e201374d05c094"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WQUAFJNJB5G7TEIMYM76JH6AQ6/bundle.json","state_url":"https://pith.science/pith/WQUAFJNJB5G7TEIMYM76JH6AQ6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WQUAFJNJB5G7TEIMYM76JH6AQ6/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":"2026-08-06T16:43:33Z","links":{"resolver":"https://pith.science/pith/WQUAFJNJB5G7TEIMYM76JH6AQ6","bundle":"https://pith.science/pith/WQUAFJNJB5G7TEIMYM76JH6AQ6/bundle.json","state":"https://pith.science/pith/WQUAFJNJB5G7TEIMYM76JH6AQ6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WQUAFJNJB5G7TEIMYM76JH6AQ6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:WQUAFJNJB5G7TEIMYM76JH6AQ6","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":"e57cffc6600d92366b1dba894bd0d9f95b29fa98b9f3c7dcbdd94be79b4d1c56","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T19:22:09Z","title_canon_sha256":"35a467ba9f9302e6f297d5929ab5ac31912de134720f9c371770782709f3bddf"},"schema_version":"1.0","source":{"id":"2607.27405","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.27405","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"arxiv_version","alias_value":"2607.27405v1","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27405","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"pith_short_12","alias_value":"WQUAFJNJB5G7","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"pith_short_16","alias_value":"WQUAFJNJB5G7TEIM","created_at":"2026-07-31T00:11:42Z"},{"alias_kind":"pith_short_8","alias_value":"WQUAFJNJ","created_at":"2026-07-31T00:11:42Z"}],"graph_snapshots":[{"event_id":"sha256:c9c57ba1b6bfb9d26f45ef8f98ea297271a073fc9db679a959e201374d05c094","target":"graph","created_at":"2026-07-31T00:11:42Z","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/2607.27405/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Both expressions of uncertainty and inferences are ubiquitous in natural language, and valid inferences over natural-language expressions of uncertainty are necessary for not only everyday conversations but also for high-stakes domains such as medicine and law. While large language models are increasingly evaluated on logical reasoning tasks, disentangling principled, symbolic reasoning from clever surface-level pattern matching is fraught with difficulty. We introduce a benchmark for reasoning over probability operators--inference over sentences with gradable epistemic modals (e.g., probably,","authors_text":"Alvin Grissom II, Jack Greff, Nayera Hasan","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T19:22:09Z","title":"Benchmarking LLM Competence on Logical Inference over Probability Operators"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27405","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":"sha256:50e796f7eb8ed1cf2a12542b0fa10383f6377192da5b123546ccc5849e3ceff0","target":"record","created_at":"2026-07-31T00:11:42Z","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":"e57cffc6600d92366b1dba894bd0d9f95b29fa98b9f3c7dcbdd94be79b4d1c56","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T19:22:09Z","title_canon_sha256":"35a467ba9f9302e6f297d5929ab5ac31912de134720f9c371770782709f3bddf"},"schema_version":"1.0","source":{"id":"2607.27405","kind":"arxiv","version":1}},"canonical_sha256":"b42802a5a90f4df9910cc33fe49fc087a2c66c7cfe15b8fdcbdf5050c6ab7fe9","receipt":{"builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b42802a5a90f4df9910cc33fe49fc087a2c66c7cfe15b8fdcbdf5050c6ab7fe9","first_computed_at":"2026-07-31T00:11:42.478921Z","kind":"pith_receipt","last_reissued_at":"2026-07-31T00:11:42.478921Z","receipt_version":"0.3","signature_status":"unsigned_v0"},"source_id":"2607.27405","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:50e796f7eb8ed1cf2a12542b0fa10383f6377192da5b123546ccc5849e3ceff0","sha256:c9c57ba1b6bfb9d26f45ef8f98ea297271a073fc9db679a959e201374d05c094"],"state_sha256":"85e48c5b349de950eae372eb5c965e19f6948937dc794664c2c920702c13bda7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nDdFCJnqLXCNLu2nG7HGinKWOhuBhvyQ5+6tS0YHszzOEkp89yp5VbXOSuqzsxJVmHIZmfqGyGjz+WhyqQLiCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T16:43:33.769629Z","bundle_sha256":"d46c32f79aa0590e94ea22add5334731103fa16ff7218b54093aa4577a42a649"}}