{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:KJFWHFVITUL4AXL6M2L7PAJNQX","short_pith_number":"pith:KJFWHFVI","canonical_record":{"source":{"id":"2506.02058","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-01T15:32:44Z","cross_cats_sorted":["cs.IR","cs.LG","stat.AP","stat.ME"],"title_canon_sha256":"ed3e8aecf8212326dd0258e67b5fa10951d0341b49bb9d22e197943a450c5723","abstract_canon_sha256":"c3aa347aaccebbf5a3f370399a8047695fa825c880161833f0d9205d153e8981"},"schema_version":"1.0"},"canonical_sha256":"524b6396a89d17c05d7e6697f7812d85ced28be5a8fe2fd57e884494b8bc8dfd","source":{"kind":"arxiv","id":"2506.02058","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.02058","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"arxiv_version","alias_value":"2506.02058v1","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02058","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"pith_short_12","alias_value":"KJFWHFVITUL4","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"pith_short_16","alias_value":"KJFWHFVITUL4AXL6","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"pith_short_8","alias_value":"KJFWHFVI","created_at":"2026-07-05T11:14:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:KJFWHFVITUL4AXL6M2L7PAJNQX","target":"record","payload":{"canonical_record":{"source":{"id":"2506.02058","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-01T15:32:44Z","cross_cats_sorted":["cs.IR","cs.LG","stat.AP","stat.ME"],"title_canon_sha256":"ed3e8aecf8212326dd0258e67b5fa10951d0341b49bb9d22e197943a450c5723","abstract_canon_sha256":"c3aa347aaccebbf5a3f370399a8047695fa825c880161833f0d9205d153e8981"},"schema_version":"1.0"},"canonical_sha256":"524b6396a89d17c05d7e6697f7812d85ced28be5a8fe2fd57e884494b8bc8dfd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:31.588473Z","signature_b64":"QfOx9kcpsxipfVQl0PkD7J0lV+tdBRIkHJ7B2sIoU9LG1Gxbex30EC+Je2hLMwGTIVivXGUs+hKQY7K4phZvAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"524b6396a89d17c05d7e6697f7812d85ced28be5a8fe2fd57e884494b8bc8dfd","last_reissued_at":"2026-07-05T11:14:31.588029Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:31.588029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.02058","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":"2026-07-05T11:14:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vFKX/6xVvAdpEGZdoKKYIsVudqoiO6UPU9GQ5e5TGzQZA1xHE+itQZpAD7+n2BfJdqOlxvksEN+gKn6hIbjNAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-22T16:22:53.021909Z"},"content_sha256":"350ceadd70c9407447e9f8833af72d4cf1e63bdb1bc389927a8f125384c04659","schema_version":"1.0","event_id":"sha256:350ceadd70c9407447e9f8833af72d4cf1e63bdb1bc389927a8f125384c04659"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:KJFWHFVITUL4AXL6M2L7PAJNQX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating the Unseen Capabilities: How Many Theorems Do LLMs Know?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR","cs.LG","stat.AP","stat.ME"],"primary_cat":"cs.CL","authors_text":"Jiayi Xin, Qi Long, Weijie J. Su, Xiang Li","submitted_at":"2025-06-01T15:32:44Z","abstract_excerpt":"Accurate evaluation of large language models (LLMs) is crucial for understanding their capabilities and guiding their development. However, current evaluations often inconsistently reflect the actual capacities of these models. In this paper, we demonstrate that one of many contributing factors to this \\textit{evaluation crisis} is the oversight of unseen knowledge -- information encoded by LLMs but not directly observed or not yet observed during evaluations. We introduce KnowSum, a statistical framework designed to provide a more comprehensive assessment by quantifying the unseen knowledge f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02058","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/2506.02058/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":"2026-07-05T11:14:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q0ED8Yzr5P/N43vCruQXojYj7CQs2gv5/+yxmVCaoiCCJhOIHMpWhXo8AN2ukDp3amOe/6VxpF9Ua8eFxh38CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-22T16:22:53.022295Z"},"content_sha256":"4271a79456e90dd637b6a1955746d438eabe5cf61f91913bce542bccaa2a3249","schema_version":"1.0","event_id":"sha256:4271a79456e90dd637b6a1955746d438eabe5cf61f91913bce542bccaa2a3249"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KJFWHFVITUL4AXL6M2L7PAJNQX/bundle.json","state_url":"https://pith.science/pith/KJFWHFVITUL4AXL6M2L7PAJNQX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KJFWHFVITUL4AXL6M2L7PAJNQX/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":"2026-07-22T16:22:53Z","links":{"resolver":"https://pith.science/pith/KJFWHFVITUL4AXL6M2L7PAJNQX","bundle":"https://pith.science/pith/KJFWHFVITUL4AXL6M2L7PAJNQX/bundle.json","state":"https://pith.science/pith/KJFWHFVITUL4AXL6M2L7PAJNQX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KJFWHFVITUL4AXL6M2L7PAJNQX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:KJFWHFVITUL4AXL6M2L7PAJNQX","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":"c3aa347aaccebbf5a3f370399a8047695fa825c880161833f0d9205d153e8981","cross_cats_sorted":["cs.IR","cs.LG","stat.AP","stat.ME"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-01T15:32:44Z","title_canon_sha256":"ed3e8aecf8212326dd0258e67b5fa10951d0341b49bb9d22e197943a450c5723"},"schema_version":"1.0","source":{"id":"2506.02058","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.02058","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"arxiv_version","alias_value":"2506.02058v1","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02058","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"pith_short_12","alias_value":"KJFWHFVITUL4","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"pith_short_16","alias_value":"KJFWHFVITUL4AXL6","created_at":"2026-07-05T11:14:31Z"},{"alias_kind":"pith_short_8","alias_value":"KJFWHFVI","created_at":"2026-07-05T11:14:31Z"}],"graph_snapshots":[{"event_id":"sha256:4271a79456e90dd637b6a1955746d438eabe5cf61f91913bce542bccaa2a3249","target":"graph","created_at":"2026-07-05T11:14:31Z","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/2506.02058/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Accurate evaluation of large language models (LLMs) is crucial for understanding their capabilities and guiding their development. However, current evaluations often inconsistently reflect the actual capacities of these models. In this paper, we demonstrate that one of many contributing factors to this \\textit{evaluation crisis} is the oversight of unseen knowledge -- information encoded by LLMs but not directly observed or not yet observed during evaluations. We introduce KnowSum, a statistical framework designed to provide a more comprehensive assessment by quantifying the unseen knowledge f","authors_text":"Jiayi Xin, Qi Long, Weijie J. Su, Xiang Li","cross_cats":["cs.IR","cs.LG","stat.AP","stat.ME"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-01T15:32:44Z","title":"Evaluating the Unseen Capabilities: How Many Theorems Do LLMs Know?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02058","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":"sha256:350ceadd70c9407447e9f8833af72d4cf1e63bdb1bc389927a8f125384c04659","target":"record","created_at":"2026-07-05T11:14:31Z","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":"c3aa347aaccebbf5a3f370399a8047695fa825c880161833f0d9205d153e8981","cross_cats_sorted":["cs.IR","cs.LG","stat.AP","stat.ME"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-01T15:32:44Z","title_canon_sha256":"ed3e8aecf8212326dd0258e67b5fa10951d0341b49bb9d22e197943a450c5723"},"schema_version":"1.0","source":{"id":"2506.02058","kind":"arxiv","version":1}},"canonical_sha256":"524b6396a89d17c05d7e6697f7812d85ced28be5a8fe2fd57e884494b8bc8dfd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"524b6396a89d17c05d7e6697f7812d85ced28be5a8fe2fd57e884494b8bc8dfd","first_computed_at":"2026-07-05T11:14:31.588029Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:31.588029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QfOx9kcpsxipfVQl0PkD7J0lV+tdBRIkHJ7B2sIoU9LG1Gxbex30EC+Je2hLMwGTIVivXGUs+hKQY7K4phZvAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:31.588473Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.02058","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:350ceadd70c9407447e9f8833af72d4cf1e63bdb1bc389927a8f125384c04659","sha256:4271a79456e90dd637b6a1955746d438eabe5cf61f91913bce542bccaa2a3249"],"state_sha256":"0b51a1dc6736a919eee565c57e8a0e3ffe207243736001beba2e720dae504e22"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iiAsWu2DxCIQLthe+I+1V45KAcey7FOUOiU6ROMGKSYFoHFXhcNPM8PisoW/tpiuMZqaEyvZVdU6DQZabaE3BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-22T16:22:53.024495Z","bundle_sha256":"368e0caffab34db63aac5523e46766497e29a2a8b87872d0b7527724c5c744e2"}}