Diagnosing Fine-Grained Inconsistency Classification in Financial Disclosure Text
Quick summary
arXiv:2607.26368v2 Announce Type: replace-cross Abstract: Financial disclosures may contain numerical, temporal, referential, factual, and policy inconsistencies that require different evidence and reasoning to diagnose. We study \emph{fine-grained inconsistency classification}: given a passage known to contain a conflict, the goal is to identify its type among 11 categories. Using a fixed snapshot of the synthetic SBID-FD benchmark, we compare frozen and fine-tuned encoders, evidence-augmented classifiers, prompted large language models, and LoRA-adapted generative models under a shared evalu
Key takeaways
- arXiv:2607.26368v2 Announce Type: replace-cross Abstract: Financial disclosures may contain numerical, temporal, referential, factual, and policy inconsistencies that require different evidence and reasoning to diagnose.
- We study \emph{fine-grained inconsistency classification}: given a passage known to contain a conflict, the goal is to identify its type among 11 categories.
- Using a fixed snapshot of the synthetic SBID-FD benchmark, we compare frozen and fine-tuned encoders, evidence-augmented classifiers, prompted large language models, and LoRA-adapted generative models under a shared evalu
Why it matters
The significance is not only the legal text but how it changes product design. Decisions around “Diagnosing Fine-Grained Inconsistency Classification in Financial Disclosure Text” may reshape data collection, model training, output accountability and market access.
Member comments