strategy-scope/

si_et_al-ideation-gpt5mini-full-20260415_012526

Benchmark: si_et_al Generated: 2026-04-15T01:44:01.632746

model: gpt-5-mini judge_model: anthropic/claude-sonnet-4-5-20250929 strategy: full n_ideas_per_run: 5 n_runs: 3 n_topics: 7 rag: False

Topics: 7 Total ideas: 108 Generation model: gpt-5-mini Judge model: anthropic/claude-sonnet-4-5-20250929 Runs per topic: 3 Ideas per run: 5

Evaluation protocol: port of Si et al.… See the full description on the dataset page: https://huggingface.co/datasets/strategy-scope/si_et_al-ideation-gpt5mini-full-20260415_012526.