Fetching from the wire…
01
02
03
04
05
06
07
08
Research2026-08-20 · source-backed
40+ experts and 31,000+ human hours built 60 project-level research tasks across 11 domains, designed so guidance is progressively withdrawn within the same project. Across 18 agent-model configurations: 50.91 with full methodological guidance, 29.10 when only the method is specified, 26.62 when the agent picks the method itself. (arXiv 2608.17271) Half the score was the human's methodology. Same conclusion as the Princeton study from a completely different direction.
Each link below shares sources, entities, or timing with this story.
Shared entities / Same source domain / Shared topic / Earlier coverage
Both cover Bench, Same; reported by the same outlet (arxiv.org); overlapping topics (different, same).
Shared entities / Shared topic / Earlier coverage / Tension
Both cover Bench, Same; overlapping topics (agent, same); earlier Bench coverage from 2026-05-14.
Shared entity: Bench / Same source domain / Shared topic / Earlier coverage / Tension
Both cover Bench; reported by the same outlet (arxiv.org); overlapping topics (agent, expert).
Shared entities / Same source domain / Earlier coverage
Both cover Bench, Same; reported by the same outlet (arxiv.org); earlier Bench coverage from 2026-07-25.
Shared entities / Shared topic / Earlier coverage
Both cover Bench, Same; overlapping topics (agent, same); earlier Bench coverage from 2026-06-09.
Shared entity: Bench / Shared topic / Earlier coverage / Tension
Both cover Bench; overlapping topics (agent, configuration, different); earlier Bench coverage from 2026-06-19.
Shared entity: Same / Shared topic / Earlier coverage / Tension
Both cover Same; overlapping topics (agent, different, same); earlier Same coverage from 2026-04-01.
Shared entity: Same / Same source domain / Shared topic / Earlier coverage
Both cover Same; reported by the same outlet (arxiv.org); overlapping topics (agent, same).