ResearchBenchmarksPick place on scientific experiment datasetFollow80.3Success RateCAPER37.97248.96159.9570.939Feb 10, 2026Evaluation ResultsMethodMethodLinksSuccess RateCAPER2026.0280.3MP+MLM+RLBackbone=GPT-4oBackbone=GPT-4o2026.0279.3MLM+RLBackbone=GPT-4oBackbone=GPT-4o2026.0278.9LLM+RLBackbone=GPT-4oBackbone=GPT-4o2026.0270.4MLM+RLBackbone=LLaMA3.2, Rea...Backbone=LLaMA3.2, Reasoning Strategy=w/o CoT2026.0245.8LLM+RLBackbone=Llama-3.1-8B,...Backbone=Llama-3.1-8B, Reasoning Strategy=w/o CoT2026.0239.6