{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YCTWFAE4BTXFKJMBJ5UIYGYGXC","short_pith_number":"pith:YCTWFAE4","canonical_record":{"source":{"id":"2506.04907","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:41:05Z","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"title_canon_sha256":"12cdfd1419e0db580bb57deb719963bde474154927a4fb24eb030608cd0bbc47","abstract_canon_sha256":"c50c3406b60dca328e66238f0b7e3ee5776437d7fe560fa95f0e40ee8870ff29"},"schema_version":"1.0"},"canonical_sha256":"c0a762809c0cee5525814f688c1b06b8a4f26610c93000ab1b100cf3f5c1c43b","source":{"kind":"arxiv","id":"2506.04907","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04907","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04907v4","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04907","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"pith_short_12","alias_value":"YCTWFAE4BTXF","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"pith_short_16","alias_value":"YCTWFAE4BTXFKJMB","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"pith_short_8","alias_value":"YCTWFAE4","created_at":"2026-07-05T11:20:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YCTWFAE4BTXFKJMBJ5UIYGYGXC","target":"record","payload":{"canonical_record":{"source":{"id":"2506.04907","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:41:05Z","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"title_canon_sha256":"12cdfd1419e0db580bb57deb719963bde474154927a4fb24eb030608cd0bbc47","abstract_canon_sha256":"c50c3406b60dca328e66238f0b7e3ee5776437d7fe560fa95f0e40ee8870ff29"},"schema_version":"1.0"},"canonical_sha256":"c0a762809c0cee5525814f688c1b06b8a4f26610c93000ab1b100cf3f5c1c43b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:12.996216Z","signature_b64":"OXjePd5jjk8XiO7Aan73wsx/yUVYiiaEPC5JS0WuAYv1ff6SdZwb3i8crDFP2CJeBXPJXnMol5qa1UrxRFDSAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c0a762809c0cee5525814f688c1b06b8a4f26610c93000ab1b100cf3f5c1c43b","last_reissued_at":"2026-07-05T11:20:12.995801Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:12.995801Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.04907","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":"2026-07-05T11:20:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0wR12xKua7cXdlBHC2G5GYQ+YyeAr+hVOxi787vECcNlzZY835Ue8l7+9rBSf1DAYZlR4/bqx65aW+ao2KSlAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T09:38:41.209573Z"},"content_sha256":"b81940ac6627c3da8a364e0ff70c8f16f3869918f69fd2ac54a18adf08762afc","schema_version":"1.0","event_id":"sha256:b81940ac6627c3da8a364e0ff70c8f16f3869918f69fd2ac54a18adf08762afc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YCTWFAE4BTXFKJMBJ5UIYGYGXC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Context Is Not Comprehension","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.IR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alex Pan, Mary-Anne Williams","submitted_at":"2025-06-05T11:41:05Z","abstract_excerpt":"The dominant way of judging Large Language Models (LLMs) has been to ask how well they can recall explicit facts from very long inputs. While today's best models achieve near perfect recall, this masks a harder skill: performing multi-step reasoning and tracking intermediate state that never appears verbatim. We introduce Verbose ListOps (VLO), a benchmark that embeds deterministic ListOps computations inside narrative camouflage and, crucially, allows step-level evaluation of every intermediate result. Experiments show that models which solve raw ListOps with approximately 100% accuracy colla"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04907","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/2506.04907/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":"2026-07-05T11:20:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fc31AIue5QzlRBFrrvZ0JFZ4LGVG69RXagWafcvAlVdiVSE03Ua8insEhGMlB2T31xAgfEwxQxiMsNSxtzy1Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T09:38:41.210497Z"},"content_sha256":"b554e1b4eefa8091f3de6057f6a1e768afd390528a701b576b815714b6885d9b","schema_version":"1.0","event_id":"sha256:b554e1b4eefa8091f3de6057f6a1e768afd390528a701b576b815714b6885d9b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YCTWFAE4BTXFKJMBJ5UIYGYGXC/bundle.json","state_url":"https://pith.science/pith/YCTWFAE4BTXFKJMBJ5UIYGYGXC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YCTWFAE4BTXFKJMBJ5UIYGYGXC/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":"2026-08-08T09:38:41Z","links":{"resolver":"https://pith.science/pith/YCTWFAE4BTXFKJMBJ5UIYGYGXC","bundle":"https://pith.science/pith/YCTWFAE4BTXFKJMBJ5UIYGYGXC/bundle.json","state":"https://pith.science/pith/YCTWFAE4BTXFKJMBJ5UIYGYGXC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YCTWFAE4BTXFKJMBJ5UIYGYGXC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YCTWFAE4BTXFKJMBJ5UIYGYGXC","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":"c50c3406b60dca328e66238f0b7e3ee5776437d7fe560fa95f0e40ee8870ff29","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:41:05Z","title_canon_sha256":"12cdfd1419e0db580bb57deb719963bde474154927a4fb24eb030608cd0bbc47"},"schema_version":"1.0","source":{"id":"2506.04907","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04907","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04907v4","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04907","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"pith_short_12","alias_value":"YCTWFAE4BTXF","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"pith_short_16","alias_value":"YCTWFAE4BTXFKJMB","created_at":"2026-07-05T11:20:12Z"},{"alias_kind":"pith_short_8","alias_value":"YCTWFAE4","created_at":"2026-07-05T11:20:12Z"}],"graph_snapshots":[{"event_id":"sha256:b554e1b4eefa8091f3de6057f6a1e768afd390528a701b576b815714b6885d9b","target":"graph","created_at":"2026-07-05T11:20:12Z","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/2506.04907/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The dominant way of judging Large Language Models (LLMs) has been to ask how well they can recall explicit facts from very long inputs. While today's best models achieve near perfect recall, this masks a harder skill: performing multi-step reasoning and tracking intermediate state that never appears verbatim. We introduce Verbose ListOps (VLO), a benchmark that embeds deterministic ListOps computations inside narrative camouflage and, crucially, allows step-level evaluation of every intermediate result. Experiments show that models which solve raw ListOps with approximately 100% accuracy colla","authors_text":"Alex Pan, Mary-Anne Williams","cross_cats":["cs.AI","cs.IR","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:41:05Z","title":"Context Is Not Comprehension"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04907","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:b81940ac6627c3da8a364e0ff70c8f16f3869918f69fd2ac54a18adf08762afc","target":"record","created_at":"2026-07-05T11:20:12Z","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":"c50c3406b60dca328e66238f0b7e3ee5776437d7fe560fa95f0e40ee8870ff29","cross_cats_sorted":["cs.AI","cs.IR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:41:05Z","title_canon_sha256":"12cdfd1419e0db580bb57deb719963bde474154927a4fb24eb030608cd0bbc47"},"schema_version":"1.0","source":{"id":"2506.04907","kind":"arxiv","version":4}},"canonical_sha256":"c0a762809c0cee5525814f688c1b06b8a4f26610c93000ab1b100cf3f5c1c43b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c0a762809c0cee5525814f688c1b06b8a4f26610c93000ab1b100cf3f5c1c43b","first_computed_at":"2026-07-05T11:20:12.995801Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:20:12.995801Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OXjePd5jjk8XiO7Aan73wsx/yUVYiiaEPC5JS0WuAYv1ff6SdZwb3i8crDFP2CJeBXPJXnMol5qa1UrxRFDSAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:20:12.996216Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.04907","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b81940ac6627c3da8a364e0ff70c8f16f3869918f69fd2ac54a18adf08762afc","sha256:b554e1b4eefa8091f3de6057f6a1e768afd390528a701b576b815714b6885d9b"],"state_sha256":"f1fb61711904bdbce89de60cc7a1221cc4b5e3e552ac0f3b96a2431d7db4331d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"r4gGe7tURbRJ+hsBrKjjJfuTtDTgobFZC2TFbkrM6vkm/+6V9HMnpWV7VhO0vKgrGKKyOH+5Ushhdgz9tSz/BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T09:38:41.215708Z","bundle_sha256":"c4f4f142ee360d80d007beaa9f976380003f7cd1f74b873f682291ddeef61bb7"}}