{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CKHH36OKI46AYUUXNWDYVT2QAL","short_pith_number":"pith:CKHH36OK","schema_version":"1.0","canonical_sha256":"128e7df9ca473c0c52976d878acf5002fbad1e3e4577f067f6365989710abee1","source":{"kind":"arxiv","id":"2605.18642","version":1},"attestation_state":"computed","paper":{"title":"A Benchmark on LLM-Based Power Flow Computation: Do More Structured Prompts Help?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SY"],"primary_cat":"eess.SY","authors_text":"Kai Sun, Kaiyang Huang, Tingwei Chen","submitted_at":"2026-05-18T16:49:17Z","abstract_excerpt":"We present a controlled benchmark evaluating three LLMs -- Claude Sonnet 4.5, Gemini 2.5 Pro, and GPT-3.5 Turbo -- across four prompt formats (from concise narrative to structured JSON with explicit iteration trace) on Gauss--Seidel AC power flow computation for a three-bus system. Against 50 test cases with reference solutions computed numerically, Gemini 2.5 Pro with the simplest narrative prompt achieves the lowest mean absolute error (MAE = 0.257 MW/MVar, 54\\% of cases within 5\\% relative error), while the same model with a JSON-structured prompt raises MAE to 0.789 -- a 3.1$\\times$ increa"},"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":"2605.18642","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.SY","submitted_at":"2026-05-18T16:49:17Z","cross_cats_sorted":["cs.SY"],"title_canon_sha256":"63f326227e2f403efa737600beb9ebfb372cd43b0d822682f80425cd88814b4c","abstract_canon_sha256":"6fe22c97775963d95b4fd5e609dc4b4fa5bcc4e2858bc7f8b7072f0c17b64c13"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:06:12.360134Z","signature_b64":"BXET9q2Z5+irLdBf1Gt/XiGAwSUmM9vd5LQorVNfEsgiW+yaggN9T1X3N+hqjnZw3BBvX3sNa33x6D48GgB/Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"128e7df9ca473c0c52976d878acf5002fbad1e3e4577f067f6365989710abee1","last_reissued_at":"2026-05-20T00:06:12.359306Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:06:12.359306Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Benchmark on LLM-Based Power Flow Computation: Do More Structured Prompts Help?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SY"],"primary_cat":"eess.SY","authors_text":"Kai Sun, Kaiyang Huang, Tingwei Chen","submitted_at":"2026-05-18T16:49:17Z","abstract_excerpt":"We present a controlled benchmark evaluating three LLMs -- Claude Sonnet 4.5, Gemini 2.5 Pro, and GPT-3.5 Turbo -- across four prompt formats (from concise narrative to structured JSON with explicit iteration trace) on Gauss--Seidel AC power flow computation for a three-bus system. Against 50 test cases with reference solutions computed numerically, Gemini 2.5 Pro with the simplest narrative prompt achieves the lowest mean absolute error (MAE = 0.257 MW/MVar, 54\\% of cases within 5\\% relative error), while the same model with a JSON-structured prompt raises MAE to 0.789 -- a 3.1$\\times$ increa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18642","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/2605.18642/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T00:01:59.184386Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"a5d322870600ad1ec0de31bda9413ab6e64a8a66462102ff0da63f5bdf4bbfdf"},"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":"2605.18642","created_at":"2026-05-20T00:06:12.359452+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18642v1","created_at":"2026-05-20T00:06:12.359452+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18642","created_at":"2026-05-20T00:06:12.359452+00:00"},{"alias_kind":"pith_short_12","alias_value":"CKHH36OKI46A","created_at":"2026-05-20T00:06:12.359452+00:00"},{"alias_kind":"pith_short_16","alias_value":"CKHH36OKI46AYUUX","created_at":"2026-05-20T00:06:12.359452+00:00"},{"alias_kind":"pith_short_8","alias_value":"CKHH36OK","created_at":"2026-05-20T00:06:12.359452+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/CKHH36OKI46AYUUXNWDYVT2QAL","json":"https://pith.science/pith/CKHH36OKI46AYUUXNWDYVT2QAL.json","graph_json":"https://pith.science/api/pith-number/CKHH36OKI46AYUUXNWDYVT2QAL/graph.json","events_json":"https://pith.science/api/pith-number/CKHH36OKI46AYUUXNWDYVT2QAL/events.json","paper":"https://pith.science/paper/CKHH36OK"},"agent_actions":{"view_html":"https://pith.science/pith/CKHH36OKI46AYUUXNWDYVT2QAL","download_json":"https://pith.science/pith/CKHH36OKI46AYUUXNWDYVT2QAL.json","view_paper":"https://pith.science/paper/CKHH36OK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18642&json=true","fetch_graph":"https://pith.science/api/pith-number/CKHH36OKI46AYUUXNWDYVT2QAL/graph.json","fetch_events":"https://pith.science/api/pith-number/CKHH36OKI46AYUUXNWDYVT2QAL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CKHH36OKI46AYUUXNWDYVT2QAL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CKHH36OKI46AYUUXNWDYVT2QAL/action/storage_attestation","attest_author":"https://pith.science/pith/CKHH36OKI46AYUUXNWDYVT2QAL/action/author_attestation","sign_citation":"https://pith.science/pith/CKHH36OKI46AYUUXNWDYVT2QAL/action/citation_signature","submit_replication":"https://pith.science/pith/CKHH36OKI46AYUUXNWDYVT2QAL/action/replication_record"}},"created_at":"2026-05-20T00:06:12.359452+00:00","updated_at":"2026-05-20T00:06:12.359452+00:00"}