BIRD: GPT-4 hits only 54.89% EX on real databases
On BIRD, GPT-4 reaches 54.89% execution accuracy with domain hints and 34.88% without—a 20-point gap any Beancount NL→BQL interface must close.
#ai
AI
Artificial intelligence research and applications in finance and accounting
Verifiably Safe Tool Use for LLM Agents: STPA Meets MCP
CMU and NC State researchers propose using System-Theoretic Process Analysis (STPA) and a capability-enhanced Model Context Protocol to derive formal safety specifications for LLM agent tool use, with Alloy-based verification demonstrating absence of unsafe flows in a calendar scheduling case study.
GraphRAG: 72–83% wins vanish after LLM-judge bias fix
Microsoft's GraphRAG posts 72–83% sensemaking wins over vector RAG; a 2025 audit collapses those after correcting judge bias—caution for multi-doc ledger QA.
FinAuditing: LLMs Score Under 14% on Real SEC XBRL Auditing Tasks
FinAuditing tests 13 LLMs zero-shot on 1,102 real SEC XBRL filing instances; top scores are 13.86% on financial math verification and 12.42% on concept retrieval—results that directly bound what AI accounting tools can be trusted to automate without external tooling.
InvestorBench: Qwen2.5-72B tops stocks at 46.15% CR
InvestorBench: Qwen2.5-72B leads stock trading at 46.15% CR; finance-tuned Palmyra-Fin backfires on equities—size beats domain fine-tuning.
StructRAG (ICLR 2025): Picking the Right Document Structure Beats GraphRAG by 28 Points
StructRAG (ICLR 2025) routes each query to a task-appropriate structure type — table, graph, catalogue, algorithm, or chunk — before reasoning, scoring 28 points higher than GraphRAG on the Loong benchmark while running 22× faster, with the DPO-trained router alone accounting for a 15-point accuracy gain.
Single-Agent: equals MAS on multi-hop at equal tokens
Under equal thinking-token budgets, single-agent LLMs match or beat multi-agent systems on multi-hop reasoning—favor simpler finance agent designs.
M3MAD-Bench: Are Multi-Agent Debates Really Effective Across Domains and Modalities?
M3MAD-Bench stress-tests Multi-Agent Debate across 9 models, 5 domains, and vision-language settings, finding that Collective Delusion causes 65% of failures, adversarial debate cuts accuracy by up to 12.8%, and Self-Consistency typically matches debate accuracy at lower token cost.
AGrail: Adaptive Safety Guardrails for LLM Agents That Learn Across Tasks
AGrail (ACL 2025) introduces a two-LLM cooperative guardrail that adapts safety checks at inference time via test-time adaptation, achieving 0% prompt injection attack success and 95.6% benign action preservation on Safe-OS — compared to GuardAgent and LLaMA-Guard blocking up to 49.2% of legitimate actions.
ShieldAgent: Verifiable Safety Policy Reasoning for LLM Agents
ShieldAgent (ICML 2025) replaces LLM-based guardrails with probabilistic rule circuits built on Markov Logic Networks, achieving 90.4% accuracy on agent attacks with 64.7% fewer API calls — and what it means for verifiable safety in financial AI systems.
Atlas: Joint Retriever-Reader Pre-Training Beats 540B-Parameter LLMs with 11B Parameters
Atlas (JMLR 2023) achieves 42.4% accuracy on Natural Questions with only 64 training examples—beating PaLM 540B by 3 points using 11B parameters—by jointly pre-training a Contriever-based dense retriever with a T5 Fusion-in-Decoder reader. Analysis covers retrieval accuracy limits, 587GB index infrastructure costs, and implications for Beancount ledger QA systems.
Fusion-in-Decoder: How Multi-Passage Retrieval Improves Generative QA
Izacard and Grave's FiD architecture independently encodes retrieved passages then fuses them in the decoder, outperforming RAG-Sequence by 4–11 points on NQ and TriviaQA. This post examines the design and its implications for Beancount ledger QA, where multi-entry synthesis across transactions is the norm.