{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YAJTQIQK24N5AD62KZXRFGOUY7","short_pith_number":"pith:YAJTQIQK","canonical_record":{"source":{"id":"2503.00808","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-02T09:21:28Z","cross_cats_sorted":[],"title_canon_sha256":"a33da15dfb1a3e981954d171b034ef0f5fc5c99aa47b9bc0b1e3fc88944c4b11","abstract_canon_sha256":"c91a8a0ee602f47417fab02625160df0d98fe081d045d4419d3293e5536eb12a"},"schema_version":"1.0"},"canonical_sha256":"c01338220ad71bd00fda566f1299d4c7e02c5106d2e4686b8d32ff2bd15df251","source":{"kind":"arxiv","id":"2503.00808","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.00808","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"arxiv_version","alias_value":"2503.00808v4","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.00808","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"pith_short_12","alias_value":"YAJTQIQK24N5","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"pith_short_16","alias_value":"YAJTQIQK24N5AD62","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"pith_short_8","alias_value":"YAJTQIQK","created_at":"2026-07-05T11:47:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YAJTQIQK24N5AD62KZXRFGOUY7","target":"record","payload":{"canonical_record":{"source":{"id":"2503.00808","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-02T09:21:28Z","cross_cats_sorted":[],"title_canon_sha256":"a33da15dfb1a3e981954d171b034ef0f5fc5c99aa47b9bc0b1e3fc88944c4b11","abstract_canon_sha256":"c91a8a0ee602f47417fab02625160df0d98fe081d045d4419d3293e5536eb12a"},"schema_version":"1.0"},"canonical_sha256":"c01338220ad71bd00fda566f1299d4c7e02c5106d2e4686b8d32ff2bd15df251","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:47:19.531099Z","signature_b64":"lQSyy2BOcJqGGWIoNdUJRJsgJbqGO9KVmDm25mE9cZzGGWsXrKMCqgmU16KU8dPGGEKIAADXgVP9xhkvgvZaCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c01338220ad71bd00fda566f1299d4c7e02c5106d2e4686b8d32ff2bd15df251","last_reissued_at":"2026-07-05T11:47:19.530638Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:47:19.530638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.00808","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":"2026-07-05T11:47:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Q7onI2RlsOxyJ5ggiRrpKflPJ3vId6imRGAyOEASmOng/sIKFU0nsLCzXEZAqkcgYdfsYPYnyVVXBkatLJRMAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T19:52:03.642604Z"},"content_sha256":"84ce5b62c98c2afaea2d6403d32cea63448650a9094c138351305299716a73ef","schema_version":"1.0","event_id":"sha256:84ce5b62c98c2afaea2d6403d32cea63448650a9094c138351305299716a73ef"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YAJTQIQK24N5AD62KZXRFGOUY7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Predictive Data Selection: The Data That Predicts Is the Data That Teaches","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hongjian Zou, Junxian He, Kashun Shum, Qian Liu, Qi Ding, Xiaoxin Chen, Yixuan Liao, Yuzhen Huang","submitted_at":"2025-03-02T09:21:28Z","abstract_excerpt":"Language model pretraining involves training on extensive corpora, where data quality plays a pivotal role. In this work, we aim to directly estimate the contribution of data during pretraining and select pretraining data in an efficient manner. Specifically, we draw inspiration from recent findings showing that compression efficiency (i.e., the normalized loss) of diverse models on certain text correlates strongly with their downstream performance, when the text domain aligns with the downstream benchmarks(Huang et al., 2024). Building on this observation, we hypothesize that data on which mo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.00808","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/2503.00808/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":"2026-07-05T11:47:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0RH35A/yGTaLzIW3BI0pE/w9kALq/9HoKGVPfaSJF5hVscC2cL9BjFHfd90J71uSLV58ZZSAKF6DdqwNpDCUDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T19:52:03.643437Z"},"content_sha256":"5412b2bd6ce24f9a7751553c4cccdb8d90d68df428a2b93a2c3b2209fb983ddb","schema_version":"1.0","event_id":"sha256:5412b2bd6ce24f9a7751553c4cccdb8d90d68df428a2b93a2c3b2209fb983ddb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YAJTQIQK24N5AD62KZXRFGOUY7/bundle.json","state_url":"https://pith.science/pith/YAJTQIQK24N5AD62KZXRFGOUY7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YAJTQIQK24N5AD62KZXRFGOUY7/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":"2026-08-07T19:52:03Z","links":{"resolver":"https://pith.science/pith/YAJTQIQK24N5AD62KZXRFGOUY7","bundle":"https://pith.science/pith/YAJTQIQK24N5AD62KZXRFGOUY7/bundle.json","state":"https://pith.science/pith/YAJTQIQK24N5AD62KZXRFGOUY7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YAJTQIQK24N5AD62KZXRFGOUY7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YAJTQIQK24N5AD62KZXRFGOUY7","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":"c91a8a0ee602f47417fab02625160df0d98fe081d045d4419d3293e5536eb12a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-02T09:21:28Z","title_canon_sha256":"a33da15dfb1a3e981954d171b034ef0f5fc5c99aa47b9bc0b1e3fc88944c4b11"},"schema_version":"1.0","source":{"id":"2503.00808","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.00808","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"arxiv_version","alias_value":"2503.00808v4","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.00808","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"pith_short_12","alias_value":"YAJTQIQK24N5","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"pith_short_16","alias_value":"YAJTQIQK24N5AD62","created_at":"2026-07-05T11:47:19Z"},{"alias_kind":"pith_short_8","alias_value":"YAJTQIQK","created_at":"2026-07-05T11:47:19Z"}],"graph_snapshots":[{"event_id":"sha256:5412b2bd6ce24f9a7751553c4cccdb8d90d68df428a2b93a2c3b2209fb983ddb","target":"graph","created_at":"2026-07-05T11:47:19Z","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/2503.00808/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language model pretraining involves training on extensive corpora, where data quality plays a pivotal role. In this work, we aim to directly estimate the contribution of data during pretraining and select pretraining data in an efficient manner. Specifically, we draw inspiration from recent findings showing that compression efficiency (i.e., the normalized loss) of diverse models on certain text correlates strongly with their downstream performance, when the text domain aligns with the downstream benchmarks(Huang et al., 2024). Building on this observation, we hypothesize that data on which mo","authors_text":"Hongjian Zou, Junxian He, Kashun Shum, Qian Liu, Qi Ding, Xiaoxin Chen, Yixuan Liao, Yuzhen Huang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-02T09:21:28Z","title":"Predictive Data Selection: The Data That Predicts Is the Data That Teaches"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.00808","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":"sha256:84ce5b62c98c2afaea2d6403d32cea63448650a9094c138351305299716a73ef","target":"record","created_at":"2026-07-05T11:47:19Z","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":"c91a8a0ee602f47417fab02625160df0d98fe081d045d4419d3293e5536eb12a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-02T09:21:28Z","title_canon_sha256":"a33da15dfb1a3e981954d171b034ef0f5fc5c99aa47b9bc0b1e3fc88944c4b11"},"schema_version":"1.0","source":{"id":"2503.00808","kind":"arxiv","version":4}},"canonical_sha256":"c01338220ad71bd00fda566f1299d4c7e02c5106d2e4686b8d32ff2bd15df251","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c01338220ad71bd00fda566f1299d4c7e02c5106d2e4686b8d32ff2bd15df251","first_computed_at":"2026-07-05T11:47:19.530638Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:47:19.530638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lQSyy2BOcJqGGWIoNdUJRJsgJbqGO9KVmDm25mE9cZzGGWsXrKMCqgmU16KU8dPGGEKIAADXgVP9xhkvgvZaCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:47:19.531099Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.00808","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:84ce5b62c98c2afaea2d6403d32cea63448650a9094c138351305299716a73ef","sha256:5412b2bd6ce24f9a7751553c4cccdb8d90d68df428a2b93a2c3b2209fb983ddb"],"state_sha256":"6de507936d4ed129e9b4887cb5b3a8b25511e138e13e709e375338a6387487b2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EcIbspmhs8a4C7qnsl26ZBKxWQXDXAYfXh1791a+CEGI6Yfvx7pWWaBvqAyLyyJRb6vpRv4EnWRGxQx0vBtpAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T19:52:03.648027Z","bundle_sha256":"4cb641a7b59976b16fb2d071e8374be62b14a033913a55da062b6abc77c916c1"}}