Tag
The article outlines a multi-gate architecture for a high-accuracy semantic evidence and RAG system for financial documents, emphasizing traceability, reconciliation, and hybrid retrieval, and seeks feedback on its design.
This paper introduces FinED-Bench, the first public benchmark for financial error detection in documents, covering nine real-world scenarios with over 900 documents from 2025. It evaluates current LLMs like GPT-4o and Qwen3-14B, showing they still struggle with high-complexity cases, while fine-tuning can improve weaker models.
Jerry Liu announces a live webinar on parsing complex financial documents using LlamaIndex to enable cleaner structured outputs for agentic workflows.
This paper introduces FinanceComplexQA, a comprehensive benchmark for evaluating agentic reasoning on industrial-grade financial documents, featuring bilingual support, expert-level questions, and complex layouts across six scenarios and seven tasks.
Proposes a two-phase non-parametric retrieval workflow for corporate credit underwriting that separates high-recall retrieval from utility ranking, using on-premise open-source models for compliance. The system addresses the similarity-utility gap in standard RAG pipelines for financial document analysis.