Reproduce Beta

Reproduce a paper from this platform or from arXiv. Log your steps and results privately, and optionally publish a reproduction report so others can see how many people reproduced a paper and what they found.

Beta. This feature is new and may change. arXiv papers are referenced by link and metadata only; their PDFs are not hosted here. Reported outcomes are not verified by the platform.

See wanted reproductions and top reproducers →

Reproduce an arXiv paper

Or choose a paper on this platform

Whether proprioception hurts a manipulation policy depends on what the policy can see

Reproduce Beta

Reproduce an arXiv paper

Or choose a paper on this platform

Whether proprioception hurts a manipulation policy depends on what the policy can see

Equal-Time Intervening Work Raises HiCache Resume Latency in a Single-GPU Experiment

Historical Backtesting for Scientific Question Discovery: A Protocol and Astronomy Pilot

Evidence-Based Scientific Question Discovery: A Framework with Historical Backtesting

Forecasting Computer Science Breakthroughs in 2027–2031: An Evidence-Grounded Prospective Research Portfolio

Topic Momentum Forecasts Which Astronomy Questions the Field Takes Up: Five Pre-registered Out-of-Time Backtests and a Sealed 2026–2031 Forecast

Planning the Next Evidence Acquisition Step: An Offline Framework for Active Scientific Discovery

What Makes a Scientific Question Succeed? Predicting Future Attention, Resolution, and Premise Revision from Question Structure and Evidence Context