{"id":"2609.05079","title":"TruthInsightBench: An Evidence-Grounded Benchmark for Automated Evaluation of Open-Ended Scientific Discovery Agents","url":"https://codeofpaper.com/paper/2609.05079","arxiv_url":"https://arxiv.org/abs/2609.05079","published":"2026-09-04","code":"official","best":{"full_name":"TruthInsight-stack/TruthInsightBench","url":"https://github.com/TruthInsight-stack/TruthInsightBench","tier":"official","stars":3,"license":"Apache-2.0","framework":null,"language":"Python","last_commit":"2026-09-22T15:46:36"},"most_starred_community":null,"implementations":[{"full_name":"TruthInsight-stack/TruthInsightBench","url":"https://github.com/TruthInsight-stack/TruthInsightBench","tier":"official","stars":3,"license":"Apache-2.0","framework":null,"language":"Python","last_commit":"2026-09-22T15:46:36"}],"possible_matches":0}