PromptLab: Tested prompts for using large language models in life science research
The DOI matches its DataCite record.
Abstract
PromptLab v1.1.0 completes the cross-model evaluation with three independent human reviewers and reorganizes the repository so main stays focused on the prompts. Archived on Zenodo under concept DOI 10.5281/zenodo.20855199. Highlights Three independent reviewers — Tom Sharpton, Alex Alexiev, and Ed Davis — now set verdicts across the 144-cell panel, up from one reviewer in v1.0.0. This fulfills the v1.1 plan for per-cell inter-rater agreement. 24 of 24 runnable prompts pass at the ≥ 4/6 panel threshold: 17 at 6/6, 6 at 5/6, 1 at 4/6, none below. Panel-wide pass rate 94.4% (136/144). Inter-rater agreement: Fleiss's κ = 0.554 (moderate; 3-way, 118-cell overlap); pairwise Cohen's κ ranges 0.378–0.717. κ is depressed by the high base pass rate (> 93%), so raw agreement (95–98%) is the more intuitive summary. Changed Regenerated the canonical verdicts, audit trail, per-prompt table, agreement report, per-prompt ## Model Notes, and tests/SUMMARY.md from the three-reviewer merge. The merge is abstention-aware: Skip / Cannot-judge do not vote, and Needs-revision wins any genuine Pass-vs-Needs-revision split. Repository restructure: the evaluation evidence (evaluation/) and the raw per-reviewer verdict files (verdicts/) now live on the dedicated resubmission-evidence branch, keeping main — and this archived snapshot — focused on the prompts. The reviewer-contribution pipeline (README, review SPA, REVIEWER.md, PI.md, and the verdict-review Action) was redirected to that branch. Where to find things Prompts with their updated ## Test Input / example outputs and inline ## Model Notes — this release, on main. Cross-model matrix — tests/SUMMARY.md. Full audit trail, per-reviewer verdicts, agreement report — the resubmission-evidence branch. See CHANGELOG.md for the complete history.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Full per-model and per-cell results are archived with the repository”
From A practical risk framework for large language model use in life science research (2026) · cites this paper as [27]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cited by
A practical risk framework for large language model use in life science research
Sharpton, T., Davis II, E. W., & Alexiev, A. (2026) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1014776
Self-citation
Metadata
- Authors
- Sharpton, T.
- Affiliations
- Oregon State University
- Venue
- Zenodo
- Publisher
- Zenodo
- Published
- 2026-07-03
- References
- 0
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: PromptLab: Tested prompts for using large language models in life science research. https://app.cochat.ai/papers/10.5281/zenodo.20855198
- The paper
- Sharpton, T. (2026). PromptLab: Tested prompts for using large language models in life science research. Zenodo. https://doi.org/10.5281/zenodo.20855198
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.