Taxonomy-Aligned Risk Extraction from 10-K Filings with Autonomous Improvement Using LLMs
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908780042649600 |
|---|---|
| author | Dolphin, Rian Dursun, Joe Blankenship, Jarrett Adams, Katie Pike, Quinton |
| author_facet | Dolphin, Rian Dursun, Joe Blankenship, Jarrett Adams, Katie Pike, Quinton |
| contents | We present a methodology for extracting structured risk factors from corporate 10-K filings while maintaining adherence to a predefined hierarchical taxonomy. Our three-stage pipeline combines LLM extraction with supporting quotes, embedding-based semantic mapping to taxonomy categories, and LLM-as-a-judge validation that filters spurious assignments. To evaluate our approach, we extract 10,688 risk factors from S&P 500 companies and examine risk profile similarity across industry clusters. Beyond extraction, we introduce autonomous taxonomy maintenance where an AI agent analyzes evaluation feedback to identify problematic categories, diagnose failure patterns, and propose refinements, achieving 104.7% improvement in embedding separation in a case study. External validation confirms the taxonomy captures economically meaningful structure: same-industry companies exhibit 63% higher risk profile similarity than cross-industry pairs (Cohen's d=1.06, AUC 0.82, p<0.001). The methodology generalizes to any domain requiring taxonomy-aligned extraction from unstructured text, with autonomous improvement enabling continuous quality maintenance and enhancement as systems process more documents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_15247 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Taxonomy-Aligned Risk Extraction from 10-K Filings with Autonomous Improvement Using LLMs Dolphin, Rian Dursun, Joe Blankenship, Jarrett Adams, Katie Pike, Quinton Computation and Language We present a methodology for extracting structured risk factors from corporate 10-K filings while maintaining adherence to a predefined hierarchical taxonomy. Our three-stage pipeline combines LLM extraction with supporting quotes, embedding-based semantic mapping to taxonomy categories, and LLM-as-a-judge validation that filters spurious assignments. To evaluate our approach, we extract 10,688 risk factors from S&P 500 companies and examine risk profile similarity across industry clusters. Beyond extraction, we introduce autonomous taxonomy maintenance where an AI agent analyzes evaluation feedback to identify problematic categories, diagnose failure patterns, and propose refinements, achieving 104.7% improvement in embedding separation in a case study. External validation confirms the taxonomy captures economically meaningful structure: same-industry companies exhibit 63% higher risk profile similarity than cross-industry pairs (Cohen's d=1.06, AUC 0.82, p<0.001). The methodology generalizes to any domain requiring taxonomy-aligned extraction from unstructured text, with autonomous improvement enabling continuous quality maintenance and enhancement as systems process more documents. |
| title | Taxonomy-Aligned Risk Extraction from 10-K Filings with Autonomous Improvement Using LLMs |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.15247 |