{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PTVABYERW6EMVCA7PAZMF5HARO","short_pith_number":"pith:PTVABYER","schema_version":"1.0","canonical_sha256":"7cea00e091b788ca881f7832c2f4e08b91324f3290cbe97fb539a2a49596ad0e","source":{"kind":"arxiv","id":"2406.16694","version":1},"attestation_state":"computed","paper":{"title":"Task Oriented In-Domain Data Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jian Jiao, Qiang Lou, Shao-Lun Huang, Simiao Zuo, Xiao Liang, Xinyu Hu, Yeyun Gong, Yi Liu","submitted_at":"2024-06-24T14:58:11Z","abstract_excerpt":"Large Language Models (LLMs) have shown superior performance in various applications and fields. To achieve better performance on specialized domains such as law and advertisement, LLMs are often continue pre-trained on in-domain data. However, existing approaches suffer from two major issues. First, in-domain data are scarce compared with general domain-agnostic data. Second, data used for continual pre-training are not task-aware, such that they may not be helpful to downstream applications. We propose TRAIT, a task-oriented in-domain data augmentation framework. Our framework is divided int"},"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":"2406.16694","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-24T14:58:11Z","cross_cats_sorted":[],"title_canon_sha256":"b1ae898fbd85cdffe64d7888a4e37b7df9c86f0c0c7de88f3973a9650ac7a299","abstract_canon_sha256":"0c497d0ab55d017382ede6a645f8595106bcf01a6be1f2ac5ba403f88d093c0c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:36:02.324464Z","signature_b64":"yMmZn/UXMWBqYXbGm34QLf3FFIA5L9flh43DqNtU3JyQQh15HaeFNot+cIgd1GnXRPEzGZ5biszBxcYOiDOlCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7cea00e091b788ca881f7832c2f4e08b91324f3290cbe97fb539a2a49596ad0e","last_reissued_at":"2026-07-05T08:36:02.323996Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:36:02.323996Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Task Oriented In-Domain Data Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jian Jiao, Qiang Lou, Shao-Lun Huang, Simiao Zuo, Xiao Liang, Xinyu Hu, Yeyun Gong, Yi Liu","submitted_at":"2024-06-24T14:58:11Z","abstract_excerpt":"Large Language Models (LLMs) have shown superior performance in various applications and fields. To achieve better performance on specialized domains such as law and advertisement, LLMs are often continue pre-trained on in-domain data. However, existing approaches suffer from two major issues. First, in-domain data are scarce compared with general domain-agnostic data. Second, data used for continual pre-training are not task-aware, such that they may not be helpful to downstream applications. We propose TRAIT, a task-oriented in-domain data augmentation framework. Our framework is divided int"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.16694","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/2406.16694/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":"2406.16694","created_at":"2026-07-05T08:36:02.324052+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.16694v1","created_at":"2026-07-05T08:36:02.324052+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.16694","created_at":"2026-07-05T08:36:02.324052+00:00"},{"alias_kind":"pith_short_12","alias_value":"PTVABYERW6EM","created_at":"2026-07-05T08:36:02.324052+00:00"},{"alias_kind":"pith_short_16","alias_value":"PTVABYERW6EMVCA7","created_at":"2026-07-05T08:36:02.324052+00:00"},{"alias_kind":"pith_short_8","alias_value":"PTVABYER","created_at":"2026-07-05T08:36:02.324052+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2409.06624","citing_title":"A Practice of Post-Training on Llama-3 70B with Optimal Selection of Additional Language Mixture Ratio","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PTVABYERW6EMVCA7PAZMF5HARO","json":"https://pith.science/pith/PTVABYERW6EMVCA7PAZMF5HARO.json","graph_json":"https://pith.science/api/pith-number/PTVABYERW6EMVCA7PAZMF5HARO/graph.json","events_json":"https://pith.science/api/pith-number/PTVABYERW6EMVCA7PAZMF5HARO/events.json","paper":"https://pith.science/paper/PTVABYER"},"agent_actions":{"view_html":"https://pith.science/pith/PTVABYERW6EMVCA7PAZMF5HARO","download_json":"https://pith.science/pith/PTVABYERW6EMVCA7PAZMF5HARO.json","view_paper":"https://pith.science/paper/PTVABYER","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.16694&json=true","fetch_graph":"https://pith.science/api/pith-number/PTVABYERW6EMVCA7PAZMF5HARO/graph.json","fetch_events":"https://pith.science/api/pith-number/PTVABYERW6EMVCA7PAZMF5HARO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PTVABYERW6EMVCA7PAZMF5HARO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PTVABYERW6EMVCA7PAZMF5HARO/action/storage_attestation","attest_author":"https://pith.science/pith/PTVABYERW6EMVCA7PAZMF5HARO/action/author_attestation","sign_citation":"https://pith.science/pith/PTVABYERW6EMVCA7PAZMF5HARO/action/citation_signature","submit_replication":"https://pith.science/pith/PTVABYERW6EMVCA7PAZMF5HARO/action/replication_record"}},"created_at":"2026-07-05T08:36:02.324052+00:00","updated_at":"2026-07-05T08:36:02.324052+00:00"}