{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:SILNRJJ3NOSWX25VJSZ62LQAQB","short_pith_number":"pith:SILNRJJ3","canonical_record":{"source":{"id":"2412.01020","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-12-02T00:38:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ba3ab04c76333a927c6379dd54a4087d81ed0ee8399e3b5bda87ef05d8e3f012","abstract_canon_sha256":"954c3ab597a013884bed3ade41a61f9189166854456b327d9d6ecb038bd84419"},"schema_version":"1.0"},"canonical_sha256":"9216d8a53b6ba56bebb54cb3ed2e008075cca81372417ffad93330aa0900aa13","source":{"kind":"arxiv","id":"2412.01020","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.01020","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"arxiv_version","alias_value":"2412.01020v1","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01020","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"pith_short_12","alias_value":"SILNRJJ3NOSW","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"pith_short_16","alias_value":"SILNRJJ3NOSWX25V","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"pith_short_8","alias_value":"SILNRJJ3","created_at":"2026-07-05T09:42:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:SILNRJJ3NOSWX25VJSZ62LQAQB","target":"record","payload":{"canonical_record":{"source":{"id":"2412.01020","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-12-02T00:38:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ba3ab04c76333a927c6379dd54a4087d81ed0ee8399e3b5bda87ef05d8e3f012","abstract_canon_sha256":"954c3ab597a013884bed3ade41a61f9189166854456b327d9d6ecb038bd84419"},"schema_version":"1.0"},"canonical_sha256":"9216d8a53b6ba56bebb54cb3ed2e008075cca81372417ffad93330aa0900aa13","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:42:52.217370Z","signature_b64":"rpcxUNLInaiFXn2fS9j+I6jtQrLGi+HXa4zQOHBXVFv6aXXLk2qFmKoQ4TeAq/FgSBHBb/fVDce9FVpZaPNXAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9216d8a53b6ba56bebb54cb3ed2e008075cca81372417ffad93330aa0900aa13","last_reissued_at":"2026-07-05T09:42:52.216850Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:42:52.216850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.01020","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":"2026-07-05T09:42:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dtvhG80Z23zCLfOmAdLEi4ft5Uzc6AI2V6C4oMEFkxzgI8s2G244OF0hAHOfvnesfk7xYzCyXW2f4+IgmZh+AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T15:57:31.093936Z"},"content_sha256":"bc75e7f1a457e0d54407f33cd9e394cf0f0c0b85f69c1bc624a8e7ea73f1a1fc","schema_version":"1.0","event_id":"sha256:bc75e7f1a457e0d54407f33cd9e394cf0f0c0b85f69c1bc624a8e7ea73f1a1fc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:SILNRJJ3NOSWX25VJSZ62LQAQB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AI Benchmarks and Datasets for LLM Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Todor Ivanov, Valeri Penchev","submitted_at":"2024-12-02T00:38:57Z","abstract_excerpt":"LLMs demand significant computational resources for both pre-training and fine-tuning, requiring distributed computing capabilities due to their large model sizes \\cite{sastry2024computing}. Their complex architecture poses challenges throughout the entire AI lifecycle, from data collection to deployment and monitoring \\cite{OECD_AIlifecycle}. Addressing critical AI system challenges, such as explainability, corrigibility, interpretability, and hallucination, necessitates a systematic methodology and rigorous benchmarking \\cite{guldimann2024complai}. To effectively improve AI systems, we must "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01020","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/2412.01020/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":"2026-07-05T09:42:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9OWI3L+QBGHV+PnzaiMkLfsOKl80TKHlFJhB+s0HtzCb/ZOCX6K9QbGkQ5uWw2JgaKegQ/0Njm2l0eAGmSCGDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T15:57:31.094759Z"},"content_sha256":"8cdd7ebdd65bf8d9ce5fcff21944f45e21962616b7ca35aa1577e1b6e1a0eb89","schema_version":"1.0","event_id":"sha256:8cdd7ebdd65bf8d9ce5fcff21944f45e21962616b7ca35aa1577e1b6e1a0eb89"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SILNRJJ3NOSWX25VJSZ62LQAQB/bundle.json","state_url":"https://pith.science/pith/SILNRJJ3NOSWX25VJSZ62LQAQB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SILNRJJ3NOSWX25VJSZ62LQAQB/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":"2026-08-13T15:57:31Z","links":{"resolver":"https://pith.science/pith/SILNRJJ3NOSWX25VJSZ62LQAQB","bundle":"https://pith.science/pith/SILNRJJ3NOSWX25VJSZ62LQAQB/bundle.json","state":"https://pith.science/pith/SILNRJJ3NOSWX25VJSZ62LQAQB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SILNRJJ3NOSWX25VJSZ62LQAQB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SILNRJJ3NOSWX25VJSZ62LQAQB","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":"954c3ab597a013884bed3ade41a61f9189166854456b327d9d6ecb038bd84419","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-12-02T00:38:57Z","title_canon_sha256":"ba3ab04c76333a927c6379dd54a4087d81ed0ee8399e3b5bda87ef05d8e3f012"},"schema_version":"1.0","source":{"id":"2412.01020","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.01020","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"arxiv_version","alias_value":"2412.01020v1","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01020","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"pith_short_12","alias_value":"SILNRJJ3NOSW","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"pith_short_16","alias_value":"SILNRJJ3NOSWX25V","created_at":"2026-07-05T09:42:52Z"},{"alias_kind":"pith_short_8","alias_value":"SILNRJJ3","created_at":"2026-07-05T09:42:52Z"}],"graph_snapshots":[{"event_id":"sha256:8cdd7ebdd65bf8d9ce5fcff21944f45e21962616b7ca35aa1577e1b6e1a0eb89","target":"graph","created_at":"2026-07-05T09:42:52Z","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/2412.01020/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLMs demand significant computational resources for both pre-training and fine-tuning, requiring distributed computing capabilities due to their large model sizes \\cite{sastry2024computing}. Their complex architecture poses challenges throughout the entire AI lifecycle, from data collection to deployment and monitoring \\cite{OECD_AIlifecycle}. Addressing critical AI system challenges, such as explainability, corrigibility, interpretability, and hallucination, necessitates a systematic methodology and rigorous benchmarking \\cite{guldimann2024complai}. To effectively improve AI systems, we must ","authors_text":"Todor Ivanov, Valeri Penchev","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-12-02T00:38:57Z","title":"AI Benchmarks and Datasets for LLM Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01020","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:bc75e7f1a457e0d54407f33cd9e394cf0f0c0b85f69c1bc624a8e7ea73f1a1fc","target":"record","created_at":"2026-07-05T09:42:52Z","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":"954c3ab597a013884bed3ade41a61f9189166854456b327d9d6ecb038bd84419","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-12-02T00:38:57Z","title_canon_sha256":"ba3ab04c76333a927c6379dd54a4087d81ed0ee8399e3b5bda87ef05d8e3f012"},"schema_version":"1.0","source":{"id":"2412.01020","kind":"arxiv","version":1}},"canonical_sha256":"9216d8a53b6ba56bebb54cb3ed2e008075cca81372417ffad93330aa0900aa13","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9216d8a53b6ba56bebb54cb3ed2e008075cca81372417ffad93330aa0900aa13","first_computed_at":"2026-07-05T09:42:52.216850Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:42:52.216850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rpcxUNLInaiFXn2fS9j+I6jtQrLGi+HXa4zQOHBXVFv6aXXLk2qFmKoQ4TeAq/FgSBHBb/fVDce9FVpZaPNXAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:42:52.217370Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.01020","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bc75e7f1a457e0d54407f33cd9e394cf0f0c0b85f69c1bc624a8e7ea73f1a1fc","sha256:8cdd7ebdd65bf8d9ce5fcff21944f45e21962616b7ca35aa1577e1b6e1a0eb89"],"state_sha256":"95fdedd7ddab190374b13edbf0b046062948804a3d811fc0ae0c4ba6eee7a09c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"87s+bthX8huaz1MveCr7QlqckWUwBP7wnLF+VMJo1opIiBJXDgWcRVcB/4VVVTlGPp/5O845iZtulEefNvZMCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T15:57:31.103405Z","bundle_sha256":"06f1c1f7df3a853e23a8bd8b6d46a89acc45281446b439ed8266d6321527a2d8"}}