{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6OSYVNJRJKR5D245QFS73D7X3V","short_pith_number":"pith:6OSYVNJR","schema_version":"1.0","canonical_sha256":"f3a58ab5314aa3d1eb9d8165fd8ff7dd7fdf8cc01ce18aeef8dfcf5b03e75cb1","source":{"kind":"arxiv","id":"2502.15854","version":1},"attestation_state":"computed","paper":{"title":"Enhancing Domain-Specific Retrieval-Augmented Generation: Synthetic Data Generation and Evaluation using Reasoning Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Aryan Jadon, Avinash Patil, Shashank Kumar","submitted_at":"2025-02-21T06:38:57Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) systems face significant performance gaps when applied to technical domains requiring precise information extraction from complex documents. Current evaluation methodologies relying on document-level metrics inadequately capture token-resolution retrieval accuracy that is critical for domain-related documents. We propose a framework combining granular evaluation metrics with synthetic data generation to optimize domain-specific RAG performance. First, we introduce token-aware metrics Precision $\\Omega$ and Intersection-over-Union (IoU) that quantify context"},"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":"2502.15854","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-21T06:38:57Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"80946faec86b2344b8b4dcf9e680ef2d4993fa52c0a5e5c9c2118bad953ed52b","abstract_canon_sha256":"5090a0c86c8297c58e995212c1005805570d336419df6979add02f652789b5c9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:18:52.145359Z","signature_b64":"Ro5uNdrwHJqoYKfGbpj+Mqd1+SzO8KbL4swWapJ3QmWwAWW9ot4PlhYxLEtoeK2dWZmtv8PXIoIMhXt23Z/LCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f3a58ab5314aa3d1eb9d8165fd8ff7dd7fdf8cc01ce18aeef8dfcf5b03e75cb1","last_reissued_at":"2026-07-05T10:18:52.144865Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:18:52.144865Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Domain-Specific Retrieval-Augmented Generation: Synthetic Data Generation and Evaluation using Reasoning Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Aryan Jadon, Avinash Patil, Shashank Kumar","submitted_at":"2025-02-21T06:38:57Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) systems face significant performance gaps when applied to technical domains requiring precise information extraction from complex documents. Current evaluation methodologies relying on document-level metrics inadequately capture token-resolution retrieval accuracy that is critical for domain-related documents. We propose a framework combining granular evaluation metrics with synthetic data generation to optimize domain-specific RAG performance. First, we introduce token-aware metrics Precision $\\Omega$ and Intersection-over-Union (IoU) that quantify context"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.15854","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/2502.15854/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":"2502.15854","created_at":"2026-07-05T10:18:52.144916+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.15854v1","created_at":"2026-07-05T10:18:52.144916+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.15854","created_at":"2026-07-05T10:18:52.144916+00:00"},{"alias_kind":"pith_short_12","alias_value":"6OSYVNJRJKR5","created_at":"2026-07-05T10:18:52.144916+00:00"},{"alias_kind":"pith_short_16","alias_value":"6OSYVNJRJKR5D245","created_at":"2026-07-05T10:18:52.144916+00:00"},{"alias_kind":"pith_short_8","alias_value":"6OSYVNJR","created_at":"2026-07-05T10:18:52.144916+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.15647","citing_title":"Neuro-Symbolic AI for LEED compliance: Document-Centric Benchmarking, Deterministic Numeric Checking, and When Multimodal Hurts","ref_index":2025,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6OSYVNJRJKR5D245QFS73D7X3V","json":"https://pith.science/pith/6OSYVNJRJKR5D245QFS73D7X3V.json","graph_json":"https://pith.science/api/pith-number/6OSYVNJRJKR5D245QFS73D7X3V/graph.json","events_json":"https://pith.science/api/pith-number/6OSYVNJRJKR5D245QFS73D7X3V/events.json","paper":"https://pith.science/paper/6OSYVNJR"},"agent_actions":{"view_html":"https://pith.science/pith/6OSYVNJRJKR5D245QFS73D7X3V","download_json":"https://pith.science/pith/6OSYVNJRJKR5D245QFS73D7X3V.json","view_paper":"https://pith.science/paper/6OSYVNJR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.15854&json=true","fetch_graph":"https://pith.science/api/pith-number/6OSYVNJRJKR5D245QFS73D7X3V/graph.json","fetch_events":"https://pith.science/api/pith-number/6OSYVNJRJKR5D245QFS73D7X3V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6OSYVNJRJKR5D245QFS73D7X3V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6OSYVNJRJKR5D245QFS73D7X3V/action/storage_attestation","attest_author":"https://pith.science/pith/6OSYVNJRJKR5D245QFS73D7X3V/action/author_attestation","sign_citation":"https://pith.science/pith/6OSYVNJRJKR5D245QFS73D7X3V/action/citation_signature","submit_replication":"https://pith.science/pith/6OSYVNJRJKR5D245QFS73D7X3V/action/replication_record"}},"created_at":"2026-07-05T10:18:52.144916+00:00","updated_at":"2026-07-05T10:18:52.144916+00:00"}