{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:F5QJGGYGUWBRKQPKQBPQMTPOLN","short_pith_number":"pith:F5QJGGYG","schema_version":"1.0","canonical_sha256":"2f60931b06a5831541ea805f064dee5b629486d67091e946d5e125dc863e6573","source":{"kind":"arxiv","id":"2402.14526","version":2},"attestation_state":"computed","paper":{"title":"Balanced Data Sampling for Language Model Training with Clustering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dahua Lin, Hang Yan, Linyang Li, Xipeng Qiu, Yunfan Shao, Zhaoye Fei","submitted_at":"2024-02-22T13:20:53Z","abstract_excerpt":"Data plays a fundamental role in the training of Large Language Models (LLMs). While attention has been paid to the collection and composition of datasets, determining the data sampling strategy in training remains an open question. Most LLMs are trained with a simple strategy, random sampling. However, this sampling strategy ignores the unbalanced nature of training data distribution, which can be sub-optimal. In this paper, we propose ClusterClip Sampling to balance the text distribution of training data for better model training. Specifically, ClusterClip Sampling utilizes data clustering t"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2402.14526","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-22T13:20:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"adb7af4482754008712af7056960cb75ce6e3e0e6073332c5835a07cc914a344","abstract_canon_sha256":"90d07053d919617958d311363030d81917be557803a3d8a6b479f0aacc9bf3dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:26:20.305428Z","signature_b64":"gPY/tQw/vkkmMIKc0PQ+mxv+h9ecbn2OReGRyZIX6Y8YTRs2GUeSXKq1ToDPAodS1od+ZtUdDEzZQKfJQfK0Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f60931b06a5831541ea805f064dee5b629486d67091e946d5e125dc863e6573","last_reissued_at":"2026-07-05T08:26:20.304937Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:26:20.304937Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Balanced Data Sampling for Language Model Training with Clustering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dahua Lin, Hang Yan, Linyang Li, Xipeng Qiu, Yunfan Shao, Zhaoye Fei","submitted_at":"2024-02-22T13:20:53Z","abstract_excerpt":"Data plays a fundamental role in the training of Large Language Models (LLMs). While attention has been paid to the collection and composition of datasets, determining the data sampling strategy in training remains an open question. Most LLMs are trained with a simple strategy, random sampling. However, this sampling strategy ignores the unbalanced nature of training data distribution, which can be sub-optimal. In this paper, we propose ClusterClip Sampling to balance the text distribution of training data for better model training. Specifically, ClusterClip Sampling utilizes data clustering t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.14526","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/2402.14526/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2402.14526","created_at":"2026-07-05T08:26:20.304992+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.14526v2","created_at":"2026-07-05T08:26:20.304992+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.14526","created_at":"2026-07-05T08:26:20.304992+00:00"},{"alias_kind":"pith_short_12","alias_value":"F5QJGGYGUWBR","created_at":"2026-07-05T08:26:20.304992+00:00"},{"alias_kind":"pith_short_16","alias_value":"F5QJGGYGUWBRKQPK","created_at":"2026-07-05T08:26:20.304992+00:00"},{"alias_kind":"pith_short_8","alias_value":"F5QJGGYG","created_at":"2026-07-05T08:26:20.304992+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F5QJGGYGUWBRKQPKQBPQMTPOLN","json":"https://pith.science/pith/F5QJGGYGUWBRKQPKQBPQMTPOLN.json","graph_json":"https://pith.science/api/pith-number/F5QJGGYGUWBRKQPKQBPQMTPOLN/graph.json","events_json":"https://pith.science/api/pith-number/F5QJGGYGUWBRKQPKQBPQMTPOLN/events.json","paper":"https://pith.science/paper/F5QJGGYG"},"agent_actions":{"view_html":"https://pith.science/pith/F5QJGGYGUWBRKQPKQBPQMTPOLN","download_json":"https://pith.science/pith/F5QJGGYGUWBRKQPKQBPQMTPOLN.json","view_paper":"https://pith.science/paper/F5QJGGYG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.14526&json=true","fetch_graph":"https://pith.science/api/pith-number/F5QJGGYGUWBRKQPKQBPQMTPOLN/graph.json","fetch_events":"https://pith.science/api/pith-number/F5QJGGYGUWBRKQPKQBPQMTPOLN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F5QJGGYGUWBRKQPKQBPQMTPOLN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F5QJGGYGUWBRKQPKQBPQMTPOLN/action/storage_attestation","attest_author":"https://pith.science/pith/F5QJGGYGUWBRKQPKQBPQMTPOLN/action/author_attestation","sign_citation":"https://pith.science/pith/F5QJGGYGUWBRKQPKQBPQMTPOLN/action/citation_signature","submit_replication":"https://pith.science/pith/F5QJGGYGUWBRKQPKQBPQMTPOLN/action/replication_record"}},"created_at":"2026-07-05T08:26:20.304992+00:00","updated_at":"2026-07-05T08:26:20.304992+00:00"}