{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GF2WXYTCQF7EDREO2ENNR6D3EU","short_pith_number":"pith:GF2WXYTC","canonical_record":{"source":{"id":"2402.10430","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T03:39:37Z","cross_cats_sorted":[],"title_canon_sha256":"5b778c6a27738bc0ef18bd25f8fe2084d2a393f893ffe4bb1c949b7f90955fda","abstract_canon_sha256":"524bceac18820ba9c8ebef90d55f53639bb32e08166b7a51c40bf6c6ece81cc1"},"schema_version":"1.0"},"canonical_sha256":"31756be262817e41c48ed11ad8f87b2524e9afa5eb76aed715376c75a4337723","source":{"kind":"arxiv","id":"2402.10430","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.10430","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"arxiv_version","alias_value":"2402.10430v1","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.10430","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"pith_short_12","alias_value":"GF2WXYTCQF7E","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"pith_short_16","alias_value":"GF2WXYTCQF7EDREO","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"pith_short_8","alias_value":"GF2WXYTC","created_at":"2026-07-05T07:45:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GF2WXYTCQF7EDREO2ENNR6D3EU","target":"record","payload":{"canonical_record":{"source":{"id":"2402.10430","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T03:39:37Z","cross_cats_sorted":[],"title_canon_sha256":"5b778c6a27738bc0ef18bd25f8fe2084d2a393f893ffe4bb1c949b7f90955fda","abstract_canon_sha256":"524bceac18820ba9c8ebef90d55f53639bb32e08166b7a51c40bf6c6ece81cc1"},"schema_version":"1.0"},"canonical_sha256":"31756be262817e41c48ed11ad8f87b2524e9afa5eb76aed715376c75a4337723","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:45:56.299443Z","signature_b64":"Q2YQ2kdJxEp6c38uxxxE80ZUL4kKZNTh4I9X2M350JpXQnSy05bkBpqPJ8wb2/8DwwUjnZIcOTaUXQ5YMAB8BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31756be262817e41c48ed11ad8f87b2524e9afa5eb76aed715376c75a4337723","last_reissued_at":"2026-07-05T07:45:56.298894Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:45:56.298894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.10430","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":"2026-07-05T07:45:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+n1SAI9QVpE0EvQveGtmU3rHFMSrLspDKDE2m+/OwVDScemi9NtoK65w7e8Zv3thLX/62Cg+yj7IgkzJzpxiBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T05:33:31.980515Z"},"content_sha256":"b6f48775e318994654a1e198b09298d2d7019abb05964da7373005e415118a64","schema_version":"1.0","event_id":"sha256:b6f48775e318994654a1e198b09298d2d7019abb05964da7373005e415118a64"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GF2WXYTCQF7EDREO2ENNR6D3EU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Smaller Language Models are capable of selecting Instruction-Tuning Training Data for Larger Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alex Nguyen, Dheeraj Mekala, Jingbo Shang","submitted_at":"2024-02-16T03:39:37Z","abstract_excerpt":"Instruction-tuning language models has become a crucial step in aligning them for general use. Typically, this process involves extensive training on large datasets, incurring high training costs. In this paper, we introduce a novel training data selection based on the learning percentage of the samples. We assert that current language models possess the capability to autonomously select high-quality training data, leading to comparable or improved performance compared to training on the entire dataset. Our experiments span different-sized models, revealing that this characteristic holds for m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.10430","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/2402.10430/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":"2026-07-05T07:45:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1buO+YGM0j3B+i8nNFxFyIKAP26CcdqexNdMeuc/LoK6nWZNt7n7TTJnUiZzJQvoxHK/FVFmu04XkxTCZwsNDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T05:33:31.981137Z"},"content_sha256":"84c50e2f400e42292a4b23719e8a95d6a9ce39f1a2fb1d463bfff006d838be6d","schema_version":"1.0","event_id":"sha256:84c50e2f400e42292a4b23719e8a95d6a9ce39f1a2fb1d463bfff006d838be6d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GF2WXYTCQF7EDREO2ENNR6D3EU/bundle.json","state_url":"https://pith.science/pith/GF2WXYTCQF7EDREO2ENNR6D3EU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GF2WXYTCQF7EDREO2ENNR6D3EU/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":"2026-07-09T05:33:31Z","links":{"resolver":"https://pith.science/pith/GF2WXYTCQF7EDREO2ENNR6D3EU","bundle":"https://pith.science/pith/GF2WXYTCQF7EDREO2ENNR6D3EU/bundle.json","state":"https://pith.science/pith/GF2WXYTCQF7EDREO2ENNR6D3EU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GF2WXYTCQF7EDREO2ENNR6D3EU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GF2WXYTCQF7EDREO2ENNR6D3EU","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":"524bceac18820ba9c8ebef90d55f53639bb32e08166b7a51c40bf6c6ece81cc1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T03:39:37Z","title_canon_sha256":"5b778c6a27738bc0ef18bd25f8fe2084d2a393f893ffe4bb1c949b7f90955fda"},"schema_version":"1.0","source":{"id":"2402.10430","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.10430","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"arxiv_version","alias_value":"2402.10430v1","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.10430","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"pith_short_12","alias_value":"GF2WXYTCQF7E","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"pith_short_16","alias_value":"GF2WXYTCQF7EDREO","created_at":"2026-07-05T07:45:56Z"},{"alias_kind":"pith_short_8","alias_value":"GF2WXYTC","created_at":"2026-07-05T07:45:56Z"}],"graph_snapshots":[{"event_id":"sha256:84c50e2f400e42292a4b23719e8a95d6a9ce39f1a2fb1d463bfff006d838be6d","target":"graph","created_at":"2026-07-05T07:45:56Z","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/2402.10430/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Instruction-tuning language models has become a crucial step in aligning them for general use. Typically, this process involves extensive training on large datasets, incurring high training costs. In this paper, we introduce a novel training data selection based on the learning percentage of the samples. We assert that current language models possess the capability to autonomously select high-quality training data, leading to comparable or improved performance compared to training on the entire dataset. Our experiments span different-sized models, revealing that this characteristic holds for m","authors_text":"Alex Nguyen, Dheeraj Mekala, Jingbo Shang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T03:39:37Z","title":"Smaller Language Models are capable of selecting Instruction-Tuning Training Data for Larger Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.10430","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:b6f48775e318994654a1e198b09298d2d7019abb05964da7373005e415118a64","target":"record","created_at":"2026-07-05T07:45:56Z","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":"524bceac18820ba9c8ebef90d55f53639bb32e08166b7a51c40bf6c6ece81cc1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T03:39:37Z","title_canon_sha256":"5b778c6a27738bc0ef18bd25f8fe2084d2a393f893ffe4bb1c949b7f90955fda"},"schema_version":"1.0","source":{"id":"2402.10430","kind":"arxiv","version":1}},"canonical_sha256":"31756be262817e41c48ed11ad8f87b2524e9afa5eb76aed715376c75a4337723","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"31756be262817e41c48ed11ad8f87b2524e9afa5eb76aed715376c75a4337723","first_computed_at":"2026-07-05T07:45:56.298894Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:45:56.298894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Q2YQ2kdJxEp6c38uxxxE80ZUL4kKZNTh4I9X2M350JpXQnSy05bkBpqPJ8wb2/8DwwUjnZIcOTaUXQ5YMAB8BA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:45:56.299443Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.10430","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b6f48775e318994654a1e198b09298d2d7019abb05964da7373005e415118a64","sha256:84c50e2f400e42292a4b23719e8a95d6a9ce39f1a2fb1d463bfff006d838be6d"],"state_sha256":"752ef5490686717b1f631251e2c5fab0e26984953265a2611b2aaea6a98f0267"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TdFFDKpAFyX5MqiPOfm7K8QFgQc2rA6bw3rd2/7Kx2817xvgGS5aX71hjfWuHipX4ch9cgYcUUB6/KSp3RT7CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-09T05:33:31.985346Z","bundle_sha256":"9971eb4308002b5a509ae7667bf1e6f2b525c6f52f95d9011a45ce60c72e732b"}}