Financial agents need to do more than answer questions about filings: they must locate evidence, parse documents, and use tools along the way. This dataset focuses on that interaction pattern, pairing finance questions with multi-turn tool-use trajectories rather than supplying only question-and-answer text. Its narrow scope makes it useful for studying agent behavior on filing-based tasks, but not a substitute for broad financial knowledge data.
AI Dataset2026
Nemotron-SFT-Agentic-Finance-v1
Provides multi-turn tool-calling examples grounded in SEC filings for fine-tuning financial reasoning agents, covering tasks such as numerical reasoning, financial modeling, and market analysis.
Introduction
Information
- Websitehuggingface.co
- OrganizationsNVIDIA Corporation
- Published date2026/07/15
Categories
More Items
AI Agent Papers2026
Evaluates whether agents can search, browse, and combine evidence across the live Web to produce unique factual answers. Its 200 tasks expose persistent weaknesses in long-horizon search and tool coordination.
AI Dataset2026
Provides 1 million synthetic UK adult records, each with multiple lifestyle and professional persona descriptions grounded in official demographic distributions for model training and AI agent evaluation.
AI Dataset2026
Evaluates language models that generate 3D objects from text by comparing their outputs in human preference tests. Scores cover creativity, coherence, prompt alignment, and visual preference.