Financial LLM Agents Need Better Memory to Handle Market Complexity

Grammar-Railed Decoding enables enterprise SQL generation with compliance gua...

HQ arXiv: AI + Accounting arXiv · 5d ago · 1 min read

GRID, a new grammar-constrained decoding engine, ensures LLM-generated SQL is syntactically valid, role-aware, and audit-compliant—addressing enterprise security and governance requirements for dat...

LLM system uses document hierarchy to improve financial report analysis

HQ arXiv: AI + Accounting arXiv · 7d ago · 1 min read

DocMaster leverages hierarchical document structures and LLMs to analyze complex financial reports and technical documents, preserving formatting context that traditional chunking methods discard.

Harness Engineering Enables Auditable Enterprise LLM Agents

HQ arXiv: AI + Accounting arXiv · 8d ago · 1 min read

Researchers propose harness-engineering framework that transforms prototype LLM applications into traceable, auditable agent architectures with deterministic behavior, source boundaries, and reprod...

New benchmark tests AI agents on realistic spreadsheet workflows for finance

HQ arXiv: AI + Accounting arXiv · 18d ago · 1 min read

SpreadsheetBench 2 evaluates AI agents on end-to-end spreadsheet tasks (generation, debugging, visualization) used in business financial modeling and reporting—moving beyond isolated cell operation...

LLM Summaries Risk Distorting Financial Decisions, Study Finds

HQ arXiv: AI + Accounting arXiv · 19d ago · 1 min read

Academic research reveals that LLM-compressed financial data loses decision-critical fidelity, with errors potentially amplifying across agentic systems—a risk for automated investment analysis.