{"record": {"aliases": ["IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO"], "artifacts": [{"id": "arxiv:2606.23032", "kind": "paper", "url": "https://arxiv.org/abs/2606.23032"}, {"id": "gh:benstaf/ipoagent", "kind": "repo", "url": "https://github.com/benstaf/ipoagent"}], "categories": ["Agents", "Finance", "Language & Knowledge"], "description": {"en": "IPO Finance Agent evaluates language-model financial analysts on IPO due-diligence questions using expert-reviewed evaluation rubrics and accuracy reporting."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.23032", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.23032", "title": "IPO Finance Agent"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.23032", "modality": null, "name": "IPO Finance Agent", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-09-25T04:40:22Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_ipo-finance-agent-benchmark-of-llm-financi_c2b9d3e2", "origin_source": "arxiv", "recognition_confidence": "0.7", "record_sha256": "09e0a6cc9b532522b0a3d2481e33a201c8d0c477a9e6986f5cb43cd33c5048f3", "relation": "extends", "review_model": "codex-cli-default", "review_state": "ai-reviewed", "reviewed_at": "2026-09-24T17:17:36.502749Z", "source_url": "https://arxiv.org/abs/2606.23032"}, "publisher": null, "released": "2026-06-22", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.23032", "source_url": "https://arxiv.org/abs/2606.23032"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-23032", "source": "claire_radar", "source_benchmark_id": "2606.23032", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO"], "annotation": "Unknown", "applicationDomains": ["Finance"], "area": "Language & Knowledge", "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The combination of a prominent IPO case, agent evaluation, public questions, and cost-efficiency comparisons could draw near-term practitioner and research interest.", "score": 68}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-24T17:17:36.502749Z", "inputHash": "1af656cbc1e4ea771ba7d58a9a877e196bb2f81a369addcd587b9bba93271170", "model": "codex-cli-default"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO", "canonicalNameSource": "paper_title", "decisionReason": "The source formally names the benchmark, specifies its IPO-diligence task, reports comparable accuracy scores, describes rubric construction, and provides a public question subset plus released code and data.", "evaluationEvidence": {"reuse": "publicly release 70 questions on the SpaceX (SPCX) S-1 filing to support reproducibility", "scoring": "the best-performing evaluated model, Zhipu GLM-5.2, reaches 79.8% accuracy", "task": "a dataset of 1,000 IPO-diligence questions"}, "model": "codex-cli-default", "reviewedAt": "2026-09-24T17:17:36.502749Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "IPO Finance Agent evaluates language-model financial analysts on IPO due-diligence questions using expert-reviewed evaluation rubrics and accuracy reporting.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "In addition, we introduce an evaluator-optimizer pipeline to automatically generate evaluation rubrics for the benchmark: candidate facts are extracted from model answers, consolidated into draft criteria, then automatically audited for omissions, hallucinations, mistiered items, and redundancy, with LLM feedback driving iterative repair, targeted enrichment, and deduplication."}, "familyId": "bmf_25c0ad103da4", "firstSeenAt": "2026-08-19", "heat": null, "id": "bm_ipo-finance-agent-benchmark-of-llm-financi_c2b9d3e2", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Financial Services"], "links": {"code": "https://github.com/benstaf/ipoagent", "data": null, "paper": "https://arxiv.org/abs/2606.23032", "pdf": "https://arxiv.org/pdf/2606.23032", "project": null, "report": "https://arxiv.org/abs/2606.23032"}, "metrics": [], "motivation": "Finance Agent v2 (by Vals AI) has emerged as the reference benchmark for evaluating both Anthropic Claude and OpenAI ChatGPT frontier language models on financial tasks.", "name": "IPO Finance Agent", "oneLine": "IPO Finance Agent evaluates language-model financial analysts on IPO due-diligence questions using expert-reviewed evaluation rubrics and accuracy reporting.", "paperTitle": "IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO", "primaryDomain": "Finance", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.23032", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-19T11:05:13.395391Z"}, "publications": [], "publishers": [{"name": "benstaf", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://github.com/benstaf/ipoagent"}], "readiness": "Runnable", "recognitionConfidence": 0.7, "relation": "extends", "releasedAt": "2026-06-22", "reviewLifecycle": {"state": "pending"}, "source": {"authors": ["Mostapha Benhenda"], "categories": ["cs.AI", "q-fin.GN"], "id": "2606.23032", "title": "IPO Finance Agent: Benchmark of LLM Financial Analysts Beyond Finance Agent v2, with Automated Rubric Generation, on the SpaceX (SPCX) IPO", "type": "arxiv", "url": "https://arxiv.org/abs/2606.23032"}, "sourceUpdatedAt": "2026-06-30T10:07:05Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "It extends financial-agent evaluation from periodic company reporting to the distinct accounting, governance, disclosure, and retrieval demands of IPO filings."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
