{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RLNMVC3E5LQZLLE3R74A2XSZ3Y","short_pith_number":"pith:RLNMVC3E","schema_version":"1.0","canonical_sha256":"8adaca8b64eae195ac9b8ff80d5e59de34572333b30fcaebd02a77f34efba89f","source":{"kind":"arxiv","id":"2505.24768","version":1},"attestation_state":"computed","paper":{"title":"From Macro to Micro: Probing Dataset Diversity in Language Model Fine-Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Haoyu Li, Kun Liu, Xuhong Li, Yiming Dong","submitted_at":"2025-05-30T16:31:05Z","abstract_excerpt":"Dataset diversity plays a pivotal role for the successful training of many machine learning models, particularly in the supervised fine-tuning (SFT) stage of large language model (LLM) development. Despite increasing recognition of its importance, systematic analyses of dataset diversity still remain underexplored. To address this gap, this work presents a systematic taxonomy of existing diversity-control strategies, which primarily focus on the instruction component, operating at either macroscopic (entire instruction semantics) or mesoscopic levels (instruction units), and furthermore introd"},"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":"2505.24768","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T16:31:05Z","cross_cats_sorted":[],"title_canon_sha256":"2983a0f422071d46be3ab68164a85c92d6cc3fc86f06b420c9dea5ee4d37920f","abstract_canon_sha256":"800c8c9bc48deaa7f903bf08849e1a63fafcf1c1f487867d2b0b0cc1eb1b2aab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:56.305452Z","signature_b64":"KrKNUcCFvKg65cI1RmLb/kYNnWSNRQnH/oka/7gtPwWr6g8bN28nmE71ZtIEBHEAdryvLujpVb11FD2EXPenCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8adaca8b64eae195ac9b8ff80d5e59de34572333b30fcaebd02a77f34efba89f","last_reissued_at":"2026-07-05T11:12:56.304844Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:56.304844Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Macro to Micro: Probing Dataset Diversity in Language Model Fine-Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Haoyu Li, Kun Liu, Xuhong Li, Yiming Dong","submitted_at":"2025-05-30T16:31:05Z","abstract_excerpt":"Dataset diversity plays a pivotal role for the successful training of many machine learning models, particularly in the supervised fine-tuning (SFT) stage of large language model (LLM) development. Despite increasing recognition of its importance, systematic analyses of dataset diversity still remain underexplored. To address this gap, this work presents a systematic taxonomy of existing diversity-control strategies, which primarily focus on the instruction component, operating at either macroscopic (entire instruction semantics) or mesoscopic levels (instruction units), and furthermore introd"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24768","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/2505.24768/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":"2505.24768","created_at":"2026-07-05T11:12:56.304923+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24768v1","created_at":"2026-07-05T11:12:56.304923+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24768","created_at":"2026-07-05T11:12:56.304923+00:00"},{"alias_kind":"pith_short_12","alias_value":"RLNMVC3E5LQZ","created_at":"2026-07-05T11:12:56.304923+00:00"},{"alias_kind":"pith_short_16","alias_value":"RLNMVC3E5LQZLLE3","created_at":"2026-07-05T11:12:56.304923+00:00"},{"alias_kind":"pith_short_8","alias_value":"RLNMVC3E","created_at":"2026-07-05T11:12:56.304923+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.27115","citing_title":"Exploring the Limits of Pruning: Task-Specific Neurons, Model Collapse, and Recovery in Task-Specific Large Language Models","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RLNMVC3E5LQZLLE3R74A2XSZ3Y","json":"https://pith.science/pith/RLNMVC3E5LQZLLE3R74A2XSZ3Y.json","graph_json":"https://pith.science/api/pith-number/RLNMVC3E5LQZLLE3R74A2XSZ3Y/graph.json","events_json":"https://pith.science/api/pith-number/RLNMVC3E5LQZLLE3R74A2XSZ3Y/events.json","paper":"https://pith.science/paper/RLNMVC3E"},"agent_actions":{"view_html":"https://pith.science/pith/RLNMVC3E5LQZLLE3R74A2XSZ3Y","download_json":"https://pith.science/pith/RLNMVC3E5LQZLLE3R74A2XSZ3Y.json","view_paper":"https://pith.science/paper/RLNMVC3E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24768&json=true","fetch_graph":"https://pith.science/api/pith-number/RLNMVC3E5LQZLLE3R74A2XSZ3Y/graph.json","fetch_events":"https://pith.science/api/pith-number/RLNMVC3E5LQZLLE3R74A2XSZ3Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RLNMVC3E5LQZLLE3R74A2XSZ3Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RLNMVC3E5LQZLLE3R74A2XSZ3Y/action/storage_attestation","attest_author":"https://pith.science/pith/RLNMVC3E5LQZLLE3R74A2XSZ3Y/action/author_attestation","sign_citation":"https://pith.science/pith/RLNMVC3E5LQZLLE3R74A2XSZ3Y/action/citation_signature","submit_replication":"https://pith.science/pith/RLNMVC3E5LQZLLE3R74A2XSZ3Y/action/replication_record"}},"created_at":"2026-07-05T11:12:56.304923+00:00","updated_at":"2026-07-05T11:12:56.304923+00:00"}