{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YGCL732NU7R25VCGBPDCKIDXBP","short_pith_number":"pith:YGCL732N","schema_version":"1.0","canonical_sha256":"c184bfef4da7e3aed4460bc62520770beeb6d643f8c938f92d0067a680a8271d","source":{"kind":"arxiv","id":"2508.10409","version":3},"attestation_state":"computed","paper":{"title":"Dataset Construction for Training LLM to Learn Analog Circuit Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.AR","authors_text":"Fan Yang, Jiangli Huang, Jinyi Shen, Ji Zhuang, Li Shang, Mingjie Zhou, Xiaoyue Ke, Xinyi Yang, Xuan Zeng, Xu Yan, Zhenyu Xu, Zhouyang Wu, Zhuoyao Du, Zihao Chen","submitted_at":"2025-08-14T07:32:07Z","abstract_excerpt":"This paper constructs a textual dataset for training large language models (LLMs) to learn analog circuit knowledge and customizes LLM training techniques. For dataset construction, high-quality textbooks are collected and decomposed into fine-grained learning nodes, which are then used to construct structured question-thinking-solution-answer (QTSA) quadruples using a multi-agent framework to capture both final answers and thought processes. The resulting dataset consists of 7.26M tokens of unlabeled data for continual pre-training (CPT) and 112.65M tokens of labeled data for supervised fine-"},"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":"2508.10409","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-08-14T07:32:07Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fccef85cbe36ee0f345ac74eb7cc6280fefc70e22083e41fa179dfc9fa702430","abstract_canon_sha256":"ee306df1af4620f729cd1518b333e2228424dedce6fe83a5e18fbcbe07e1a618"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T01:17:41.001334Z","signature_b64":"eEZL3PuHT7959BKsfEWRMFIPY4Fysq52+J/ErajNgIyKiNiYZNzd85n9SYNxslR881LSmdx1TSz5ckMHuU1DBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c184bfef4da7e3aed4460bc62520770beeb6d643f8c938f92d0067a680a8271d","last_reissued_at":"2026-07-01T01:17:41.000809Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T01:17:41.000809Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dataset Construction for Training LLM to Learn Analog Circuit Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.AR","authors_text":"Fan Yang, Jiangli Huang, Jinyi Shen, Ji Zhuang, Li Shang, Mingjie Zhou, Xiaoyue Ke, Xinyi Yang, Xuan Zeng, Xu Yan, Zhenyu Xu, Zhouyang Wu, Zhuoyao Du, Zihao Chen","submitted_at":"2025-08-14T07:32:07Z","abstract_excerpt":"This paper constructs a textual dataset for training large language models (LLMs) to learn analog circuit knowledge and customizes LLM training techniques. For dataset construction, high-quality textbooks are collected and decomposed into fine-grained learning nodes, which are then used to construct structured question-thinking-solution-answer (QTSA) quadruples using a multi-agent framework to capture both final answers and thought processes. The resulting dataset consists of 7.26M tokens of unlabeled data for continual pre-training (CPT) and 112.65M tokens of labeled data for supervised fine-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.10409","kind":"arxiv","version":3},"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/2508.10409/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":"2508.10409","created_at":"2026-07-01T01:17:41.000869+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.10409v3","created_at":"2026-07-01T01:17:41.000869+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.10409","created_at":"2026-07-01T01:17:41.000869+00:00"},{"alias_kind":"pith_short_12","alias_value":"YGCL732NU7R2","created_at":"2026-07-01T01:17:41.000869+00:00"},{"alias_kind":"pith_short_16","alias_value":"YGCL732NU7R25VCG","created_at":"2026-07-01T01:17:41.000869+00:00"},{"alias_kind":"pith_short_8","alias_value":"YGCL732N","created_at":"2026-07-01T01:17:41.000869+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.23195","citing_title":"AnalogRetriever: Learning Cross-Modal Representations for Analog Circuit Retrieval","ref_index":21,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YGCL732NU7R25VCGBPDCKIDXBP","json":"https://pith.science/pith/YGCL732NU7R25VCGBPDCKIDXBP.json","graph_json":"https://pith.science/api/pith-number/YGCL732NU7R25VCGBPDCKIDXBP/graph.json","events_json":"https://pith.science/api/pith-number/YGCL732NU7R25VCGBPDCKIDXBP/events.json","paper":"https://pith.science/paper/YGCL732N"},"agent_actions":{"view_html":"https://pith.science/pith/YGCL732NU7R25VCGBPDCKIDXBP","download_json":"https://pith.science/pith/YGCL732NU7R25VCGBPDCKIDXBP.json","view_paper":"https://pith.science/paper/YGCL732N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.10409&json=true","fetch_graph":"https://pith.science/api/pith-number/YGCL732NU7R25VCGBPDCKIDXBP/graph.json","fetch_events":"https://pith.science/api/pith-number/YGCL732NU7R25VCGBPDCKIDXBP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YGCL732NU7R25VCGBPDCKIDXBP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YGCL732NU7R25VCGBPDCKIDXBP/action/storage_attestation","attest_author":"https://pith.science/pith/YGCL732NU7R25VCGBPDCKIDXBP/action/author_attestation","sign_citation":"https://pith.science/pith/YGCL732NU7R25VCGBPDCKIDXBP/action/citation_signature","submit_replication":"https://pith.science/pith/YGCL732NU7R25VCGBPDCKIDXBP/action/replication_record"}},"created_at":"2026-07-01T01:17:41.000869+00:00","updated_at":"2026-07-01T01:17:41.000869+00:00"}