ResearchBenchmarksStructured video reasoning on Sliding Puzzle Med.Follow89Success RateTBS-2043.2455.126778.88Jun 11, 2026Evaluation ResultsMethodMethodLinksSuccess RateTBS-20Budget=20Budget=202026.0689TBS-10Budget=10Budget=102026.0686BoN-20Budget=20Budget=202026.0686BoN-10Budget=10Budget=102026.0680.5BoN-5Budget=5Budget=52026.0675.5BoN-1Budget=1Budget=12026.0645