Tipping the scales: Merging weak agents into a state-of-the-art deep researcher

A new method achieves state-of-the-art results on the DeepResearch Bench II (DRB II) by merging the outputs of multiple weaker AI agents. Instead of building a single superior agent, this approach combines existing reports from lower-ranked agents using an "agglomerative pairwise merging" technique. This process effectively aggregates scattered facts, significantly boosting the "Information Recall" score, which is the most heavily weighted component of the benchmark. The final merged report surpasses the previous best score by leveraging the collective, diverse knowledge of the weaker agents without creating a new agent from scratch.