ResearchBenchmarksMix on scientific experiment datasetFollow52.3Success RateCAPER22.1429.9737.845.63Feb 10, 2026Evaluation ResultsMethodMethodLinksSuccess RateCAPER2026.0252.3MP+MLM+RLBackbone=GPT-4oBackbone=GPT-4o2026.0243.6MLM+RLBackbone=GPT-4oBackbone=GPT-4o2026.0243.2MLM+RLBackbone=LLaMA3.2, Rea...Backbone=LLaMA3.2, Reasoning Strategy=w/o CoT2026.0235.5LLM+RLBackbone=GPT-4oBackbone=GPT-4o2026.0223.8LLM+RLBackbone=Llama-3.1-8B,...Backbone=Llama-3.1-8B, Reasoning Strategy=w/o CoT2026.0223.3