Chinese AI Model Kimi K3 Tested on U.S. Tax Return Filing
Kimi K3 evaluated on TaxCalcBench for ability to file American tax returns, testing cross-border AI capability in specialized tax domain.
Kimi K3 evaluated on TaxCalcBench for ability to file American tax returns, testing cross-border AI capability in specialized tax domain.
Rillet introduces Aura Memory, a persistent memory system for its AI-native accounting platform that learns and adapts to individual user workflows, enabling personalized automation.
New evaluation framework tests AI systems on multilingual financial question answering across English, Chinese, Japanese, Spanish, and Greek with 256 benchmark items.
Researchers combine gradient boosting, graph features, SHAP explanations, and bounded LLM agents to build auditable fraud detection that scales with transaction volume while remaining reviewable fo...
Researchers propose robust financial statement fraud detection framework using LLMs that addresses overfitting in prior work by testing generalization across companies and time periods, advancing f...
Researchers propose using misinformation detection and explanation techniques to assist financial auditors in identifying material misstatements in balance sheets, income statements, and cash flow ...
Researchers develop LightGBM-based detection system with TreeSHAP attribution to identify money mule accounts at scale, combining 280 transactional and behavioral features for explainable fraud pre...
PA-DSL handles realistic audit scenarios where human labels contain errors and only some items receive expert review, improving automated classifier accuracy in accounting data labeling.
Academic research proposes XAI-enhanced Isolation Forest model for banking transaction anomaly detection, addressing compliance teams' need for interpretable fraud alerts over opaque rule-based sys...
Research paper presents constraint-aware hierarchical search method for fine-grained classification under regulatory rules, addressing customs tariff, export control, and equipment coding tasks whe...
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...
Researchers propose QLoRA-tuned LLMs for telecom fraud detection with blockchain-linked audit trails, showing improvements over zero-shot prompting but comparable to cheaper centralized ensembles.
DocMaster leverages hierarchical document structures and LLMs to analyze complex financial reports and technical documents, preserving formatting context that traditional chunking methods discard.
Researchers propose harness-engineering framework that transforms prototype LLM applications into traceable, auditable agent architectures with deterministic behavior, source boundaries, and reprod...
Accounting firms relying on historical compliance work face disruption from AI like ChatGPT; those adding advisory and forward-looking services will thrive, argues industry opinion piece.
Edgefield Group opens formerly private CPE courses to the public and debuts a new learning platform focused on Microsoft Copilot workflows for accounting teams.
Virginia Society of CPAs teams with Blue J to provide members access to AI tools for tax, compliance, advisory, and training workflows.
Opinion piece argues accounting requires AI systems capable of traceability and defensibility—generic models fail to meet audit and regulatory standards.
Roberto Dias Duarte argues accounting firms are converging with software companies as AI enables service providers to adopt software-like operations and anyone can build software with proper method...
Microsoft deepens Excel Copilot integration into financial workflows, automating formula creation and data analysis for accounting and finance professionals.