{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:FPGUZMYZKTDSDXFUI5O7XYWHWF","short_pith_number":"pith:FPGUZMYZ","canonical_record":{"source":{"id":"2502.05551","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-08T12:46:33Z","cross_cats_sorted":[],"title_canon_sha256":"72a329493a872cbbfc5f4b5a16644bf5409c0d651a9143caae1cc0d0f1c06f34","abstract_canon_sha256":"a570615191e31b3c92df22bfcfbd4ee55e8af158a84b36f6b3bad5762d279024"},"schema_version":"1.0"},"canonical_sha256":"2bcd4cb31954c721dcb4475dfbe2c7b143ee0d0526255f608559bf969cb2cddf","source":{"kind":"arxiv","id":"2502.05551","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.05551","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"arxiv_version","alias_value":"2502.05551v4","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.05551","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"pith_short_12","alias_value":"FPGUZMYZKTDS","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"pith_short_16","alias_value":"FPGUZMYZKTDSDXFU","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"pith_short_8","alias_value":"FPGUZMYZ","created_at":"2026-07-05T11:13:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:FPGUZMYZKTDSDXFUI5O7XYWHWF","target":"record","payload":{"canonical_record":{"source":{"id":"2502.05551","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-08T12:46:33Z","cross_cats_sorted":[],"title_canon_sha256":"72a329493a872cbbfc5f4b5a16644bf5409c0d651a9143caae1cc0d0f1c06f34","abstract_canon_sha256":"a570615191e31b3c92df22bfcfbd4ee55e8af158a84b36f6b3bad5762d279024"},"schema_version":"1.0"},"canonical_sha256":"2bcd4cb31954c721dcb4475dfbe2c7b143ee0d0526255f608559bf969cb2cddf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:20.065568Z","signature_b64":"ud9xRxLh+1Ag5tAi+fCHg25ULGTGE8bxnGzuZkK9RDU5mzW3ZYuxHTKvfxRQ1J+4q8mydCA+LkmJVQUgadAkCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2bcd4cb31954c721dcb4475dfbe2c7b143ee0d0526255f608559bf969cb2cddf","last_reissued_at":"2026-07-05T11:13:20.065103Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:20.065103Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.05551","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":"2026-07-05T11:13:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5AvbHFbjVh5GAAw6gGI3rPo6S8WlA7PY83H25ZuOOveVik7HDlQayTgCsTj+1LPDk8o0BM2l+jlPzccebwIpCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T02:14:44.078289Z"},"content_sha256":"aee96ee7333b4e29f56655b1fc40d9e36aba18a2ef0682c1dbc399a961aed397","schema_version":"1.0","event_id":"sha256:aee96ee7333b4e29f56655b1fc40d9e36aba18a2ef0682c1dbc399a961aed397"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:FPGUZMYZKTDSDXFUI5O7XYWHWF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FRAME: Boosting LLMs with A Four-Quadrant Multi-Stage Pretraining Strategy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Feiyu Duan, Jingang Wang, Liangyu Xu, Rongxiang Weng, Sirui Wang, Xuemiao Zhang, Xunliang Cai, Yongwei Zhou","submitted_at":"2025-02-08T12:46:33Z","abstract_excerpt":"Large language models (LLMs) have significantly advanced human language understanding and generation, with pretraining data quality and organization being crucial to their performance. Multi-stage pretraining is a promising approach, but existing methods often lack quantitative criteria for data partitioning and instead rely on intuitive heuristics. In this paper, we propose the novel Four-quadRAnt Multi-stage prEtraining strategy (FRAME), guided by the established principle of organizing the pretraining process into four stages to achieve significant loss reductions four times. This principle"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.05551","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/2502.05551/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":"2026-07-05T11:13:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"U8TuqAgWdppq1DQFXCSscDhZvA0CdfOaV+r8ZtimwjPOBhgQksJAvfOo4SvIbRLlNZxnPeb2UNyqs5PgOpO/Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T02:14:44.078686Z"},"content_sha256":"cba3974d87b14bbcf1d6cf80d27f25fdbf56a04f8d4f281055742bf16ce06b32","schema_version":"1.0","event_id":"sha256:cba3974d87b14bbcf1d6cf80d27f25fdbf56a04f8d4f281055742bf16ce06b32"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FPGUZMYZKTDSDXFUI5O7XYWHWF/bundle.json","state_url":"https://pith.science/pith/FPGUZMYZKTDSDXFUI5O7XYWHWF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FPGUZMYZKTDSDXFUI5O7XYWHWF/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":"2026-07-07T02:14:44Z","links":{"resolver":"https://pith.science/pith/FPGUZMYZKTDSDXFUI5O7XYWHWF","bundle":"https://pith.science/pith/FPGUZMYZKTDSDXFUI5O7XYWHWF/bundle.json","state":"https://pith.science/pith/FPGUZMYZKTDSDXFUI5O7XYWHWF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FPGUZMYZKTDSDXFUI5O7XYWHWF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FPGUZMYZKTDSDXFUI5O7XYWHWF","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":"a570615191e31b3c92df22bfcfbd4ee55e8af158a84b36f6b3bad5762d279024","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-08T12:46:33Z","title_canon_sha256":"72a329493a872cbbfc5f4b5a16644bf5409c0d651a9143caae1cc0d0f1c06f34"},"schema_version":"1.0","source":{"id":"2502.05551","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.05551","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"arxiv_version","alias_value":"2502.05551v4","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.05551","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"pith_short_12","alias_value":"FPGUZMYZKTDS","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"pith_short_16","alias_value":"FPGUZMYZKTDSDXFU","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"pith_short_8","alias_value":"FPGUZMYZ","created_at":"2026-07-05T11:13:20Z"}],"graph_snapshots":[{"event_id":"sha256:cba3974d87b14bbcf1d6cf80d27f25fdbf56a04f8d4f281055742bf16ce06b32","target":"graph","created_at":"2026-07-05T11:13:20Z","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/2502.05551/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have significantly advanced human language understanding and generation, with pretraining data quality and organization being crucial to their performance. Multi-stage pretraining is a promising approach, but existing methods often lack quantitative criteria for data partitioning and instead rely on intuitive heuristics. In this paper, we propose the novel Four-quadRAnt Multi-stage prEtraining strategy (FRAME), guided by the established principle of organizing the pretraining process into four stages to achieve significant loss reductions four times. This principle","authors_text":"Feiyu Duan, Jingang Wang, Liangyu Xu, Rongxiang Weng, Sirui Wang, Xuemiao Zhang, Xunliang Cai, Yongwei Zhou","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-08T12:46:33Z","title":"FRAME: Boosting LLMs with A Four-Quadrant Multi-Stage Pretraining Strategy"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.05551","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:aee96ee7333b4e29f56655b1fc40d9e36aba18a2ef0682c1dbc399a961aed397","target":"record","created_at":"2026-07-05T11:13:20Z","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":"a570615191e31b3c92df22bfcfbd4ee55e8af158a84b36f6b3bad5762d279024","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-08T12:46:33Z","title_canon_sha256":"72a329493a872cbbfc5f4b5a16644bf5409c0d651a9143caae1cc0d0f1c06f34"},"schema_version":"1.0","source":{"id":"2502.05551","kind":"arxiv","version":4}},"canonical_sha256":"2bcd4cb31954c721dcb4475dfbe2c7b143ee0d0526255f608559bf969cb2cddf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2bcd4cb31954c721dcb4475dfbe2c7b143ee0d0526255f608559bf969cb2cddf","first_computed_at":"2026-07-05T11:13:20.065103Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:20.065103Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ud9xRxLh+1Ag5tAi+fCHg25ULGTGE8bxnGzuZkK9RDU5mzW3ZYuxHTKvfxRQ1J+4q8mydCA+LkmJVQUgadAkCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:20.065568Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.05551","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aee96ee7333b4e29f56655b1fc40d9e36aba18a2ef0682c1dbc399a961aed397","sha256:cba3974d87b14bbcf1d6cf80d27f25fdbf56a04f8d4f281055742bf16ce06b32"],"state_sha256":"28615d556cc2ad8ec95a97cc1a38395d319f40e7ddd9169fec387e3b548393bd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w1gkjUJpD2bjlvDD6cXmti3tJtZDXcauDGETdTJ38FClWZ65exWGhzkhcicUB/6fz9mXDsIn9vt0fh4eAOFhCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T02:14:44.080637Z","bundle_sha256":"d76fa2be510397b19a1860e7e975f0a82353bc413cb737f6a2e690a95d7d8cb2"}}