Interactive demonstration
A durable, human-gated auto-research pipeline
Name a target journal, and pose a research question or leave it to the pipeline. It reviews the scholarly literature with live search, proposes testable ideas and pauses for you to choose one; it then analyses a real, public dataset in an isolated sandbox, interprets the results, drafts a manuscript to that journal's conventions, reviews it, and pauses again for your sign-off. Every stage is checkpointed, and a run may wait at a gate for as long as you like.
- Stages
- 6
- Human gates
- 2
- Disciplines
- 9
- Example papers
- 9
How it works
You set two things: the target journal or conference, and — optionally — the question. Leave the question blank and the pipeline proposes one that venue would want, answerable with a dataset it can actually obtain. The six stages below then run on their own, each checkpointed, pausing only at the two gates.
- 01
Literature review
A grounded search of the scholarly record — journals, proceedings and preprints — distilled into key papers, methods in use and open gaps, with every source kept.
- 02
IdeationHuman gate
Ranked candidate studies, each with a falsifiable hypothesis, a named real dataset and a concrete test plan. The run pauses here for you to choose one.
- 03
Experimentation
The chosen dataset is downloaded, checksum-verified and placed in a fresh sandbox VM with a loader. The model writes a Python analysis, which runs under a time limit; a failure triggers a repair round.
- 04
Analysis
An honest results section built only from the numbers the run produced — verdict, key figures, caveats. A refuted hypothesis is reported as refuted.
- 05
Paper writing
A manuscript written to the target venue's structure and reporting conventions, describing the real data it analysed and citing only retrieved sources.
- 06
ReviewHuman gate
A referee report against the venue's standards: 1–5 scores, weaknesses and required changes. The run pauses for your sign-off; sending it back rewrites the paper and re-reviews it.
Example research by discipline
One complete study per discipline, each analysing a real public dataset and written to the conventions of the field's leading journal, with every artifact preserved: sources, ideas, executed code, results, manuscript and referee report.
Social Sciences & Business
Finance
Modelled on Journal of Finance
7af2280f389b8 September 2026CompleteVolatility Timing or Factor Momentum? A Real-Time Horse Race on a Century of U.S. Factor Returns
Do volatility-managed factor portfolios survive real-time implementation and a horse race against time-series factor momentum? A century of US market, size and value factors
Review: Minor revision · Novelty 3/5 · Rigor 4/5 · Clarity 4/5 · Reproducibility 5/5
Economics
Modelled on American Economic Review
e968530de6708 September 2026CompleteMargin Composition and the IV–OLS Gap in the Return to Schooling
Why is the college-proximity IV estimate of the return to schooling nearly twice the OLS estimate? Testing whether the gap is an Angrist-Imbens margin-weighting artefact
Review: Minor revision · Novelty 4/5 · Rigor 4/5 · Clarity 4/5 · Reproducibility 5/5
Psychology
Modelled on Psychological Science
44d2d5b89d848 September 2026CompleteDomain Scores Discard Half the Age Signal in Their Own Items: A Loading-Conditioned Test of the Domain-General Model
Are age differences in personality domain-general or nuance-specific? Testing whether Big Five domain scores capture the age-related information carried by their own items
Review: Minor revision · Novelty 4/5 · Rigor 4/5 · Clarity 4/5 · Reproducibility 5/5
Management Science & Operations
Modelled on Management Science
c2291bbbc8a08 September 2026CompleteContact Fatigue or Dynamic Selection? Decomposing the Declining Conversion Hazard in Telemarketing and the Value of Client-Specific Stopping Rules
Contact fatigue or dynamic selection? Decomposing the declining conversion hazard across repeated telemarketing contacts and designing client-specific stopping rules
Review: Minor revision · Novelty 4/5 · Rigor 4/5 · Clarity 4/5 · Reproducibility 5/5
Health Sciences
Biostatistics
Modelled on Statistics in Medicine
8ef59f2d4a2f8 September 2026CompleteCovariate adjustment does not automatically buy power in a randomised survival trial: a plasmode resampling study of the NCCTG adjuvant colon cancer trial
Does prespecified covariate adjustment in the Cox model really increase power over the unadjusted log-rank test in a randomised survival trial? A real-data resampling study of the NCCTG adjuvant colon cancer trial
Review: Minor revision · Novelty 4/5 · Rigor 4/5 · Clarity 5/5 · Reproducibility 5/5
Epidemiology & Public Health
Modelled on American Journal of Epidemiology
b1eefce326718 September 2026CompleteTesting Indication, Not Age: A Replication Attempt of the Reported SARS-CoV-2 Viral-Load Excess in Children Under 5 Years
Do children under 5 years have higher SARS-CoV-2 upper-respiratory viral loads (lower RT-PCR cycle threshold values) than adults, once testing setting and calendar time are accounted for?
Review: Minor revision · Novelty 3/5 · Rigor 4/5 · Clarity 4/5 · Reproducibility 4/5
Physical Sciences & Engineering
Physics
Modelled on Physical Review Letters
32eaf23e6fe28 September 2026CompleteNo Chronometer in the Sun: the Phase of the Schwabe Cycle Diffuses as a Random Walk
Does the solar activity cycle keep time like a clock, or does its phase perform a random walk? A likelihood test on sunspot-cycle minima epochs in the SILSO v2.0 record with propagated measurement uncertainty
Review: Minor revision · Novelty 4/5 · Rigor 4/5 · Clarity 4/5 · Reproducibility 4/5
Computer Science — Machine Learning
Modelled on NeurIPS
193af5f072588 September 2026CompleteCalibrate or Tune the Cut-off? A Matched-Budget Contest Between Two Recipes for Cost-Sensitive Thresholding
Under asymmetric misclassification costs, does post-hoc probability calibration with the theoretical cost-optimal threshold beat directly tuning the decision threshold on the same held-out validation data?
Review: Minor revision · Novelty 3/5 · Rigor 4/5 · Clarity 4/5 · Reproducibility 5/5
Life Sciences
Ecology & Evolutionary Biology
Modelled on Ecology Letters
ad88c89b80ff8 September 2026CompleteConspecifics, not heterospecifics, limit individual body condition in a desert rodent community
Does intraspecific competition exceed interspecific competition in a desert rodent community, and is interspecific competition stronger between body-size-similar species?
Review: Minor revision · Novelty 4/5 · Rigor 4/5 · Clarity 5/5 · Reproducibility 5/5
Live demonstration
Live runs call the model API on your behalf, so they are behind a PIN. The PIN identifies your account, which carries a run allowance set by the administrator; the archived runs remain open to everyone.