SearchHorizonBench was introduced as a benchmark for learning and evaluating the value of additional search in AI agents, addressing the gap between answer-success probabilities, which capture question difficulty, and the decision of whether another search is worth its cost. The benchmark is built on execution experience and targets the estimation of search's additional benefit rather than question difficulty alone.
AI agents must decide when another search is worth its cost. Answer-success probabilities capture question difficulty, while this decision requires estimating the additional benefit of search. We introduce SearchHorizonBench, a benchmark for learning and evaluating search value from execution experi
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.