The Between-Round Decision in Autonomous Research: An LLM Controller That Reads the Whole Trajectory A new LLM-based controller called BRED (Between-Round Evidence-based Decision) improved on task baselines across three research-agent benchmarks by reading the accumulated trajectory of past rounds, according to an arXiv paper (2611.22661v1). BRED cut MAE on ETTh1 by 4.80% to 0.4172 and raised accuracy by 3.44 points on SST-2 to 95.07% and 3.43 points on CIFAR-100 to 84.63%, ahead of published results; without history the same loop reached only 0.4299 on ETTh1 and 83.71% on CIFAR-100. On a new speech-based depression detection task under a leakage-free protocol, BRED reached a macro-F1 of 0.6164 on held-out seeds, 5.5 points above its baseline and 4.0 above the strongest published system, with code released at github.com/learninginvision/BRED. arXiv:2611.22661v1 Announce Type: cross Abstract: Executing an idea often changes how promising it looks, so the next step in autonomous research should rest on results, not on proposals. After each round of experiments, the system must decide whether to refine the best solution or start a new direction, which solution to build from, and what goal to set. We call this the between-round decision. Existing systems wrap it in machinery such as rules, scores, search trees, and schedules; where an LLM makes it, what the LLM needs to read is left open. We ask whether an LLM can make this decision from accumulated evidence, without a search structure around it, and how much of that evidence it needs to read. We present BRED Between-Round Evidence-based Decision , an LLM-based controller that maintains the best verified solution so far the Champion and a log of past rounds, and makes one decision per round. On three widely used research-agent benchmark tasks, BRED improves on the task baselines by 4.80% in MAE on ETTh1 to 0.4172 and by 3.44 and 3.43 points of accuracy on SST-2 to 95.07% and CIFAR-100 to 84.63% , ahead of the published results. When the controller does not read the accumulated evidence, the same loop gains less: with no history it reaches 0.4299 on ETTh1 and 83.71% on CIFAR-100, whereas the full controller's MAE is 2.95% lower and its accuracy 0.92 points higher. On a new task, speech-based depression detection under a leakage-free protocol, it reaches a macro-F1 of 0.6164 on seeds held out from selection, 5.5 points above its baseline and 4.0 above the strongest published system. Code is available at https://github.com/learninginvision/BRED.