Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.14002 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913126030508032 |
|---|---|
| author | Zhu, Yifei |
| author_facet | Zhu, Yifei |
| contents | Large Reasoning Models (LRMs) embedded in agentic frameworks have transformed information retrieval from static, long context question answering into open-ended exploration. Yet real world use requires models to discover and synthesize "long-tail" facts from dispersed sources, a capability that remains under-evaluated. We introduce PolitNuggets, a multilingual benchmark for agentic information synthesis via constructing political biographies for 400 global elites, covering over 10000 political facts. We standardize evaluation with an optimized multi agent system and propose FactNet, an evidence conditional protocol that scores discovery, fine-grained accuracy, and efficiency. Across models and settings, we find that current systems often struggle with fine-grained details, and vary substantially in efficiency. Finally, using benchmark diagnostics, we relate agent performance to underlying model capabilities, highlighting the importance of short-context extraction, multilingual robustness, and reliable tool use. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_14002 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PolitNuggets: Benchmarking Agentic Discovery of Long-Tail Political Facts Zhu, Yifei Artificial Intelligence Large Reasoning Models (LRMs) embedded in agentic frameworks have transformed information retrieval from static, long context question answering into open-ended exploration. Yet real world use requires models to discover and synthesize "long-tail" facts from dispersed sources, a capability that remains under-evaluated. We introduce PolitNuggets, a multilingual benchmark for agentic information synthesis via constructing political biographies for 400 global elites, covering over 10000 political facts. We standardize evaluation with an optimized multi agent system and propose FactNet, an evidence conditional protocol that scores discovery, fine-grained accuracy, and efficiency. Across models and settings, we find that current systems often struggle with fine-grained details, and vary substantially in efficiency. Finally, using benchmark diagnostics, we relate agent performance to underlying model capabilities, highlighting the importance of short-context extraction, multilingual robustness, and reliable tool use. |
| title | PolitNuggets: Benchmarking Agentic Discovery of Long-Tail Political Facts |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2605.14002 |