{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:RN62SPZK5UH37NH6AUQDFTP7GR","short_pith_number":"pith:RN62SPZK","canonical_record":{"source":{"id":"2109.05687","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T03:39:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3287e9b578f061eb0728c4e8695807ae2dd5c2a4b6facbb64b4445ccc81bc2ae","abstract_canon_sha256":"478c8470779db8888181152bd240286df3944140389581eb36f228ca477a894d"},"schema_version":"1.0"},"canonical_sha256":"8b7da93f2aed0fbfb4fe052032cdff3460d84fd278d2cd4404648557a4ea1da5","source":{"kind":"arxiv","id":"2109.05687","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.05687","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"arxiv_version","alias_value":"2109.05687v1","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.05687","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"pith_short_12","alias_value":"RN62SPZK5UH3","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"pith_short_16","alias_value":"RN62SPZK5UH37NH6","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"pith_short_8","alias_value":"RN62SPZK","created_at":"2026-07-05T03:13:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:RN62SPZK5UH37NH6AUQDFTP7GR","target":"record","payload":{"canonical_record":{"source":{"id":"2109.05687","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T03:39:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3287e9b578f061eb0728c4e8695807ae2dd5c2a4b6facbb64b4445ccc81bc2ae","abstract_canon_sha256":"478c8470779db8888181152bd240286df3944140389581eb36f228ca477a894d"},"schema_version":"1.0"},"canonical_sha256":"8b7da93f2aed0fbfb4fe052032cdff3460d84fd278d2cd4404648557a4ea1da5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:13:41.340581Z","signature_b64":"oditkxgkaeoUxyNsyp+GmcHXVvMDWZ4UgTXechoafxo2mW1XOKjBbzDDbR/cHWDbh6pqnudvpYg5i3o6MA+nDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8b7da93f2aed0fbfb4fe052032cdff3460d84fd278d2cd4404648557a4ea1da5","last_reissued_at":"2026-07-05T03:13:41.340171Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:13:41.340171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2109.05687","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":"2026-07-05T03:13:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LDy3KfN6SP/bkmOmB17vEgJDbT5ndVo+gMzWm0otrRZFic7QBEGlgwvHA6dZHg8NOU9RZxP8zkmhXohHgdcDDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-13T02:13:29.128573Z"},"content_sha256":"a73947bd5239753d1197afeb14b199a5d90a46f455cc5fb571efbbd0083afaa0","schema_version":"1.0","event_id":"sha256:a73947bd5239753d1197afeb14b199a5d90a46f455cc5fb571efbbd0083afaa0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:RN62SPZK5UH37NH6AUQDFTP7GR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Raise a Child in Large Language Model: Towards Effective and Generalizable Fine-tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Baobao Chang, Chuanqi Tan, Fei Huang, Fuli Luo, Runxin Xu, Songfang Huang, Zhiyuan Zhang","submitted_at":"2021-09-13T03:39:52Z","abstract_excerpt":"Recent pretrained language models extend from millions to billions of parameters. Thus the need to fine-tune an extremely large pretrained model with a limited training corpus arises in various downstream tasks. In this paper, we propose a straightforward yet effective fine-tuning technique, Child-Tuning, which updates a subset of parameters (called child network) of large pretrained models via strategically masking out the gradients of the non-child network during the backward process. Experiments on various downstream tasks in GLUE benchmark show that Child-Tuning consistently outperforms th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.05687","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/2109.05687/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":"2026-07-05T03:13:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8p/sBhnvyeo+2wmrvvhUbgLxm6zGfKo5I6giyEMpkHbBue2ODy0mvFqEhY9xrRYZ2BTWXbJHzC9Pp5af3cM6CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-13T02:13:29.129241Z"},"content_sha256":"1927b8e501592915aa01c00d2160f0fd775d6d118a11fe719fc8f4d0ff403681","schema_version":"1.0","event_id":"sha256:1927b8e501592915aa01c00d2160f0fd775d6d118a11fe719fc8f4d0ff403681"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RN62SPZK5UH37NH6AUQDFTP7GR/bundle.json","state_url":"https://pith.science/pith/RN62SPZK5UH37NH6AUQDFTP7GR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RN62SPZK5UH37NH6AUQDFTP7GR/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":"2026-07-13T02:13:29Z","links":{"resolver":"https://pith.science/pith/RN62SPZK5UH37NH6AUQDFTP7GR","bundle":"https://pith.science/pith/RN62SPZK5UH37NH6AUQDFTP7GR/bundle.json","state":"https://pith.science/pith/RN62SPZK5UH37NH6AUQDFTP7GR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RN62SPZK5UH37NH6AUQDFTP7GR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:RN62SPZK5UH37NH6AUQDFTP7GR","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":"478c8470779db8888181152bd240286df3944140389581eb36f228ca477a894d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T03:39:52Z","title_canon_sha256":"3287e9b578f061eb0728c4e8695807ae2dd5c2a4b6facbb64b4445ccc81bc2ae"},"schema_version":"1.0","source":{"id":"2109.05687","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.05687","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"arxiv_version","alias_value":"2109.05687v1","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.05687","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"pith_short_12","alias_value":"RN62SPZK5UH3","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"pith_short_16","alias_value":"RN62SPZK5UH37NH6","created_at":"2026-07-05T03:13:41Z"},{"alias_kind":"pith_short_8","alias_value":"RN62SPZK","created_at":"2026-07-05T03:13:41Z"}],"graph_snapshots":[{"event_id":"sha256:1927b8e501592915aa01c00d2160f0fd775d6d118a11fe719fc8f4d0ff403681","target":"graph","created_at":"2026-07-05T03:13:41Z","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/2109.05687/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent pretrained language models extend from millions to billions of parameters. Thus the need to fine-tune an extremely large pretrained model with a limited training corpus arises in various downstream tasks. In this paper, we propose a straightforward yet effective fine-tuning technique, Child-Tuning, which updates a subset of parameters (called child network) of large pretrained models via strategically masking out the gradients of the non-child network during the backward process. Experiments on various downstream tasks in GLUE benchmark show that Child-Tuning consistently outperforms th","authors_text":"Baobao Chang, Chuanqi Tan, Fei Huang, Fuli Luo, Runxin Xu, Songfang Huang, Zhiyuan Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T03:39:52Z","title":"Raise a Child in Large Language Model: Towards Effective and Generalizable Fine-tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.05687","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:a73947bd5239753d1197afeb14b199a5d90a46f455cc5fb571efbbd0083afaa0","target":"record","created_at":"2026-07-05T03:13:41Z","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":"478c8470779db8888181152bd240286df3944140389581eb36f228ca477a894d","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T03:39:52Z","title_canon_sha256":"3287e9b578f061eb0728c4e8695807ae2dd5c2a4b6facbb64b4445ccc81bc2ae"},"schema_version":"1.0","source":{"id":"2109.05687","kind":"arxiv","version":1}},"canonical_sha256":"8b7da93f2aed0fbfb4fe052032cdff3460d84fd278d2cd4404648557a4ea1da5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8b7da93f2aed0fbfb4fe052032cdff3460d84fd278d2cd4404648557a4ea1da5","first_computed_at":"2026-07-05T03:13:41.340171Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:13:41.340171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oditkxgkaeoUxyNsyp+GmcHXVvMDWZ4UgTXechoafxo2mW1XOKjBbzDDbR/cHWDbh6pqnudvpYg5i3o6MA+nDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T03:13:41.340581Z","signed_message":"canonical_sha256_bytes"},"source_id":"2109.05687","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a73947bd5239753d1197afeb14b199a5d90a46f455cc5fb571efbbd0083afaa0","sha256:1927b8e501592915aa01c00d2160f0fd775d6d118a11fe719fc8f4d0ff403681"],"state_sha256":"474cb1d974be47eb30d1d4be36527d0091599ddb519478d979f460ed6a7020ae"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AgXQz+pPILIVjXnlgpqN1afyuVZnH4b0X4bw7FbPG4m4NY8rnwIFDvSwuq7UXm1jO1JGZcD9IYAZTnXKFR8eDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-13T02:13:29.133297Z","bundle_sha256":"1d45fdf7906d964606a1c62a8d904d7aba963d521f22319247c19b9111c0f544"}}