Summary Generation on InsightBench 2025
58.7RecallDataSTORM
Evaluation results
| Method | Links | |
|---|---|---|
| DataSTORMBackbone LLM=gpt-5, LLM Judge=Qwen3-30B judge2026.04 | 58.7 | |
| DataSTORMBackbone LLM=gpt-5, LLM Judge=GPT-4o judge2026.04 | 52.5 | |
| AgentPoirotBackbone LLM=gpt-5, LLM Judge=Qwen3-30B judge2026.04 | 51.5 | |
| AgentPoirotBackbone LLM=gpt-5, LLM Judge=GPT-4o judge2026.04 | 46.6 |