{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:BFIHXHMWCBIWSUCZYJNN6J5N5P","short_pith_number":"pith:BFIHXHMW","canonical_record":{"source":{"id":"2406.07515","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-11T17:46:16Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"04824ba928a54a86a5aec9c48a91892f36ed28659cd3bce23ac2031bf8efd45b","abstract_canon_sha256":"4b0a528f618e01e34926123b9ccf638cac260884eee3b37b1b3a23619f0d01c4"},"schema_version":"1.0"},"canonical_sha256":"09507b9d961051695059c25adf27adebe5000830654021c05274e167efa7dda7","source":{"kind":"arxiv","id":"2406.07515","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.07515","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"arxiv_version","alias_value":"2406.07515v2","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07515","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"pith_short_12","alias_value":"BFIHXHMWCBIW","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"pith_short_16","alias_value":"BFIHXHMWCBIWSUCZ","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"pith_short_8","alias_value":"BFIHXHMW","created_at":"2026-07-05T09:25:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:BFIHXHMWCBIWSUCZYJNN6J5N5P","target":"record","payload":{"canonical_record":{"source":{"id":"2406.07515","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-11T17:46:16Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"04824ba928a54a86a5aec9c48a91892f36ed28659cd3bce23ac2031bf8efd45b","abstract_canon_sha256":"4b0a528f618e01e34926123b9ccf638cac260884eee3b37b1b3a23619f0d01c4"},"schema_version":"1.0"},"canonical_sha256":"09507b9d961051695059c25adf27adebe5000830654021c05274e167efa7dda7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:25:44.252065Z","signature_b64":"KX+X8zYB08RxLee0ct13ufbG/6eW3l3MBRNtK7zvPl3I/Ng95YStVl/5GLHoEyfkP4j72ZpEwo49++75KgEVCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"09507b9d961051695059c25adf27adebe5000830654021c05274e167efa7dda7","last_reissued_at":"2026-07-05T09:25:44.251566Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:25:44.251566Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.07515","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:25:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y7LFgFcuwX9mQz6iG/eB7SygRXpqLG2ZJSl43UnydIhMmwKHBpvNOe+r6HpxVZzZynmTTWxhY4GEHWGYClEiCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T19:38:30.992242Z"},"content_sha256":"0a48aae8842f87a6eb002e115f8592ecd73a1070a2cc81b00994536fdd9ba1c1","schema_version":"1.0","event_id":"sha256:0a48aae8842f87a6eb002e115f8592ecd73a1070a2cc81b00994536fdd9ba1c1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:BFIHXHMWCBIWSUCZYJNN6J5N5P","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Beyond Model Collapse: Scaling Up with Synthesized Data Requires Verification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Elvis Dohmatob, Francois Charton, Julia Kempe, Pu Yang, Yunzhen Feng","submitted_at":"2024-06-11T17:46:16Z","abstract_excerpt":"Large Language Models (LLM) are increasingly trained on data generated by other LLM, either because generated text and images become part of the pre-training corpus, or because synthetized data is used as a replacement for expensive human-annotation. This raises concerns about \\emph{model collapse}, a drop in model performance when their training sets include generated data. Considering that it is easier for both humans and machines to tell between good and bad examples than to generate high-quality samples, we investigate the use of verification on synthesized data to prevent model collapse. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07515","kind":"arxiv","version":2},"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/2406.07515/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":"2026-07-05T09:25:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mxOI+GeBWkWiRYVJwx7H8IFS6o3jhbpVlHofh0sU++P/6x7bck8cNQWYRXLUZzCJkLDXOC/JPpBPDMdiVmz7Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T19:38:30.992759Z"},"content_sha256":"bdab568f5a076735dfbe166f186ecabd53a738bdf8baad0c16a7327f46b568c0","schema_version":"1.0","event_id":"sha256:bdab568f5a076735dfbe166f186ecabd53a738bdf8baad0c16a7327f46b568c0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BFIHXHMWCBIWSUCZYJNN6J5N5P/bundle.json","state_url":"https://pith.science/pith/BFIHXHMWCBIWSUCZYJNN6J5N5P/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BFIHXHMWCBIWSUCZYJNN6J5N5P/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":"2026-08-19T19:38:30Z","links":{"resolver":"https://pith.science/pith/BFIHXHMWCBIWSUCZYJNN6J5N5P","bundle":"https://pith.science/pith/BFIHXHMWCBIWSUCZYJNN6J5N5P/bundle.json","state":"https://pith.science/pith/BFIHXHMWCBIWSUCZYJNN6J5N5P/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BFIHXHMWCBIWSUCZYJNN6J5N5P/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:BFIHXHMWCBIWSUCZYJNN6J5N5P","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":"4b0a528f618e01e34926123b9ccf638cac260884eee3b37b1b3a23619f0d01c4","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-11T17:46:16Z","title_canon_sha256":"04824ba928a54a86a5aec9c48a91892f36ed28659cd3bce23ac2031bf8efd45b"},"schema_version":"1.0","source":{"id":"2406.07515","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.07515","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"arxiv_version","alias_value":"2406.07515v2","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07515","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"pith_short_12","alias_value":"BFIHXHMWCBIW","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"pith_short_16","alias_value":"BFIHXHMWCBIWSUCZ","created_at":"2026-07-05T09:25:44Z"},{"alias_kind":"pith_short_8","alias_value":"BFIHXHMW","created_at":"2026-07-05T09:25:44Z"}],"graph_snapshots":[{"event_id":"sha256:bdab568f5a076735dfbe166f186ecabd53a738bdf8baad0c16a7327f46b568c0","target":"graph","created_at":"2026-07-05T09:25:44Z","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/2406.07515/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLM) are increasingly trained on data generated by other LLM, either because generated text and images become part of the pre-training corpus, or because synthetized data is used as a replacement for expensive human-annotation. This raises concerns about \\emph{model collapse}, a drop in model performance when their training sets include generated data. Considering that it is easier for both humans and machines to tell between good and bad examples than to generate high-quality samples, we investigate the use of verification on synthesized data to prevent model collapse. ","authors_text":"Elvis Dohmatob, Francois Charton, Julia Kempe, Pu Yang, Yunzhen Feng","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-11T17:46:16Z","title":"Beyond Model Collapse: Scaling Up with Synthesized Data Requires Verification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07515","kind":"arxiv","version":2},"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":"sha256:0a48aae8842f87a6eb002e115f8592ecd73a1070a2cc81b00994536fdd9ba1c1","target":"record","created_at":"2026-07-05T09:25:44Z","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":"4b0a528f618e01e34926123b9ccf638cac260884eee3b37b1b3a23619f0d01c4","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-11T17:46:16Z","title_canon_sha256":"04824ba928a54a86a5aec9c48a91892f36ed28659cd3bce23ac2031bf8efd45b"},"schema_version":"1.0","source":{"id":"2406.07515","kind":"arxiv","version":2}},"canonical_sha256":"09507b9d961051695059c25adf27adebe5000830654021c05274e167efa7dda7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"09507b9d961051695059c25adf27adebe5000830654021c05274e167efa7dda7","first_computed_at":"2026-07-05T09:25:44.251566Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:25:44.251566Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KX+X8zYB08RxLee0ct13ufbG/6eW3l3MBRNtK7zvPl3I/Ng95YStVl/5GLHoEyfkP4j72ZpEwo49++75KgEVCg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:25:44.252065Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.07515","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0a48aae8842f87a6eb002e115f8592ecd73a1070a2cc81b00994536fdd9ba1c1","sha256:bdab568f5a076735dfbe166f186ecabd53a738bdf8baad0c16a7327f46b568c0"],"state_sha256":"22bbf2d553c3743c700d7dde1c193993f8d91f67da533aff102c17cd800e4d59"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HVQ5tLZ++Ajx9uwf8FwNU64Swb+bvDeaDg9xxqSDf2Y9x053l7N1o7LCZFBNgtFCIiDySudKl44S9CjC0IO+Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T19:38:30.998089Z","bundle_sha256":"394d465889dcbb04d24b7f133a83a2d4d3c3c43dd92762e81229caeedb26793d"}}