• FNP – FNP 2026
• FNP 2026 proceedings PDF.

When Tables Go Crazy: Evaluating Multimodal Models on French Financial Documents

• Links
  • [2602.10384] When Tables Go Crazy: Evaluating Multimodal Models on French Financial Documents
  • Github contains both dataset and scripts!
• Benchmark French visual finance docs, includes evaluation on multiple VLMs
  • Chart checkboxes graphs and weird edge cases specifically present
• Bits
  • LLMs when asked to generate questions usually generate only questions they have answers to

CFQA: A Chinese Financial Question Answering Benchmark from Corporate Annual Reports

• Links:
  • ZackZhu00/CFQA_Chinese_Finance_Question_Answering · Datasets at Hugging Face
  • zhutianning/Hallucination-detection-for-RAG: This project aims to design and implement a hallucination detection and evaluation pipeline for Retrieval- Augmented Generation (RAG) systems processing multi-modal financial reports (text, tables, images/charts).
  • Unclear relation:
    • Github (?) / Addressing investor concerns: a Chinese financial question-answering benchmark with LLM-based evaluation | EPJ Data Science | Springer Nature Link (??)
• Takeaway
  • RAG decreases hallucinations and improves fact extraction scores but not clear for other tasks involving reasoning, and decrease scores w/ some model[s

(!) Verifiable Financial Enterprise Question Answering via Inference-Time Grounding and Traceability

• LLMs grounding verifiable citations
• Framework, modular, real-time
• Verify citations and fix citation drift by looking at ovelaps/presence w/ source documents.
  • Sentence-level citations allegedly lead to better groundedness
• TODO Many many interesting citations to parse
• Related paper I just found: [2604.23588] FinGround: Detecting and Grounding Financial Hallucinations via Atomic Claim Verification