{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:NCTF5MNDTSQX7RVKCBESIC3K4N","short_pith_number":"pith:NCTF5MND","canonical_record":{"source":{"id":"2309.17425","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-09-29T17:37:29Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"15de1a7cd2396d4249aee61e0598d5906cc4bb5893053ff4ceaea65d0f4ca61b","abstract_canon_sha256":"76ccb0d64fa214d8defb0ba38b359e9893c8250db467673b322d2d2983dfd164"},"schema_version":"1.0"},"canonical_sha256":"68a65eb1a39ca17fc6aa1049240b6ae36a2de59417183c31208ca3de85489902","source":{"kind":"arxiv","id":"2309.17425","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.17425","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"arxiv_version","alias_value":"2309.17425v3","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.17425","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"pith_short_12","alias_value":"NCTF5MNDTSQX","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"pith_short_16","alias_value":"NCTF5MNDTSQX7RVK","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"pith_short_8","alias_value":"NCTF5MND","created_at":"2026-07-05T07:09:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:NCTF5MNDTSQX7RVKCBESIC3K4N","target":"record","payload":{"canonical_record":{"source":{"id":"2309.17425","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-09-29T17:37:29Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"15de1a7cd2396d4249aee61e0598d5906cc4bb5893053ff4ceaea65d0f4ca61b","abstract_canon_sha256":"76ccb0d64fa214d8defb0ba38b359e9893c8250db467673b322d2d2983dfd164"},"schema_version":"1.0"},"canonical_sha256":"68a65eb1a39ca17fc6aa1049240b6ae36a2de59417183c31208ca3de85489902","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:09:02.020015Z","signature_b64":"dAm6liGxs5mcauTHITXJNsUSnodCoUGvSsk3kQhAkxRFebO5BjkgqU/MR5doxlMNPlyktKw0KgyYLUxhinffBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"68a65eb1a39ca17fc6aa1049240b6ae36a2de59417183c31208ca3de85489902","last_reissued_at":"2026-07-05T07:09:02.019524Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:09:02.019524Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2309.17425","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:09:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fVevZhipGSnM6D75Ku5J6H1QtjNpYILPyk7IMxDaGSpQyJ0kcCEimlYdHkaWitvds0Ew4t48mYxXYOj81X79DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T23:26:31.824826Z"},"content_sha256":"be6bd21b8b0151d0069b0b661e88c2018370ac4ce0cf9246308e30572eb56c33","schema_version":"1.0","event_id":"sha256:be6bd21b8b0151d0069b0b661e88c2018370ac4ce0cf9246308e30572eb56c33"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:NCTF5MNDTSQX7RVKCBESIC3K4N","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Data Filtering Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Albin Madappally Jose, Alexander Toshev, Alex Fang, Amit Jain, Ludwig Schmidt, Vaishaal Shankar","submitted_at":"2023-09-29T17:37:29Z","abstract_excerpt":"Large training sets have become a cornerstone of machine learning and are the foundation for recent advances in language modeling and multimodal learning. While data curation for pre-training is often still ad-hoc, one common paradigm is to first collect a massive pool of data from the Web and then filter this candidate pool down to an actual training set via various heuristics. In this work, we study the problem of learning a data filtering network (DFN) for this second step of filtering a large uncurated dataset. Our key finding is that the quality of a network for filtering is distinct from"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.17425","kind":"arxiv","version":3},"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/2309.17425/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":"2026-07-05T07:09:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"I7GhYtfGhUzcLXWZkhBB1ZBu4zmV8ruDDlW4nWrgXR7sYqVt2t+5+t6aTQljiiMqy6god+TDfwSPNXFCARX6Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T23:26:31.825203Z"},"content_sha256":"0d9e23464703ef3cb764b060f90810f772806fa42a1602b8c062581b2b400945","schema_version":"1.0","event_id":"sha256:0d9e23464703ef3cb764b060f90810f772806fa42a1602b8c062581b2b400945"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NCTF5MNDTSQX7RVKCBESIC3K4N/bundle.json","state_url":"https://pith.science/pith/NCTF5MNDTSQX7RVKCBESIC3K4N/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NCTF5MNDTSQX7RVKCBESIC3K4N/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":"2026-07-06T23:26:31Z","links":{"resolver":"https://pith.science/pith/NCTF5MNDTSQX7RVKCBESIC3K4N","bundle":"https://pith.science/pith/NCTF5MNDTSQX7RVKCBESIC3K4N/bundle.json","state":"https://pith.science/pith/NCTF5MNDTSQX7RVKCBESIC3K4N/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NCTF5MNDTSQX7RVKCBESIC3K4N/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:NCTF5MNDTSQX7RVKCBESIC3K4N","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":"76ccb0d64fa214d8defb0ba38b359e9893c8250db467673b322d2d2983dfd164","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-09-29T17:37:29Z","title_canon_sha256":"15de1a7cd2396d4249aee61e0598d5906cc4bb5893053ff4ceaea65d0f4ca61b"},"schema_version":"1.0","source":{"id":"2309.17425","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.17425","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"arxiv_version","alias_value":"2309.17425v3","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.17425","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"pith_short_12","alias_value":"NCTF5MNDTSQX","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"pith_short_16","alias_value":"NCTF5MNDTSQX7RVK","created_at":"2026-07-05T07:09:02Z"},{"alias_kind":"pith_short_8","alias_value":"NCTF5MND","created_at":"2026-07-05T07:09:02Z"}],"graph_snapshots":[{"event_id":"sha256:0d9e23464703ef3cb764b060f90810f772806fa42a1602b8c062581b2b400945","target":"graph","created_at":"2026-07-05T07:09:02Z","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/2309.17425/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large training sets have become a cornerstone of machine learning and are the foundation for recent advances in language modeling and multimodal learning. While data curation for pre-training is often still ad-hoc, one common paradigm is to first collect a massive pool of data from the Web and then filter this candidate pool down to an actual training set via various heuristics. In this work, we study the problem of learning a data filtering network (DFN) for this second step of filtering a large uncurated dataset. Our key finding is that the quality of a network for filtering is distinct from","authors_text":"Albin Madappally Jose, Alexander Toshev, Alex Fang, Amit Jain, Ludwig Schmidt, Vaishaal Shankar","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-09-29T17:37:29Z","title":"Data Filtering Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.17425","kind":"arxiv","version":3},"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:be6bd21b8b0151d0069b0b661e88c2018370ac4ce0cf9246308e30572eb56c33","target":"record","created_at":"2026-07-05T07:09:02Z","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":"76ccb0d64fa214d8defb0ba38b359e9893c8250db467673b322d2d2983dfd164","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-09-29T17:37:29Z","title_canon_sha256":"15de1a7cd2396d4249aee61e0598d5906cc4bb5893053ff4ceaea65d0f4ca61b"},"schema_version":"1.0","source":{"id":"2309.17425","kind":"arxiv","version":3}},"canonical_sha256":"68a65eb1a39ca17fc6aa1049240b6ae36a2de59417183c31208ca3de85489902","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"68a65eb1a39ca17fc6aa1049240b6ae36a2de59417183c31208ca3de85489902","first_computed_at":"2026-07-05T07:09:02.019524Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:09:02.019524Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dAm6liGxs5mcauTHITXJNsUSnodCoUGvSsk3kQhAkxRFebO5BjkgqU/MR5doxlMNPlyktKw0KgyYLUxhinffBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:09:02.020015Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.17425","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:be6bd21b8b0151d0069b0b661e88c2018370ac4ce0cf9246308e30572eb56c33","sha256:0d9e23464703ef3cb764b060f90810f772806fa42a1602b8c062581b2b400945"],"state_sha256":"25c9b971a1676297922689946e5fe7b0d47996d1fecc358619b833d2aade13dd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DrQyWE/VwFZSsFSo1OIjKaYbYPmZlpagckt6WdbK6xcHY0btmdtbYKDXexggptaLljDXrIm1a2Qu28OPX9WFCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T23:26:31.827112Z","bundle_sha256":"2f47731ca5d8cffb41aaf04b9e1698255c4aac87c9b3de4fe106453f6c364722"}}