ResearchBenchmarksGroup Name Generation on RadTimelineFollow87RL ScoreOracle36.0449.2762.575.73Mar 24, 2026Evaluation ResultsMethodMethodLinksRL ScoreMiss RateDuplicate RateUnnecessary RateOracle2026.03879119GPT-4oContext Type=finding-o...Context Type=finding-only (Fdg.)2026.0352133146GPT-4oContext Type=FullContext Type=Full2026.0348302837Llama 3.1Context Type=finding-o...Context Type=finding-only (Fdg.)2026.0342274258Llama 3.1Context Type=FullContext Type=Full2026.0338253260