Evaluation

NAIAD with UCB acquisition on Norman (Tables 2 and 3, acquisition functions with NAIAD)

Published comparison; transcribed, not reproduced.

Evaluation results

1 evaluation · 2 results. Different protocols are not a single leaderboard.

Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 · Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)

Sorted by Count (marginal gain (P_4 - P_0)/4 in the screen's 200 strongest measured gene pairs found, round 0 to round 4) (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.

Tested configuration	Count (marginal gain (P_4 - P_0)/4 in the screen's 200 strongest measured gene pairs found, round 0 to round 4)	True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning)
NAIAD with UCB acquisitionEvaluation	2.25	97

All 2 result rows with coverage, uncertainty and sources

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: NAIAD with UCB acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	2.25 Count (marginal gain (P_4 - P_0)/4 in the screen's 200 strongest measured gene pairs found, round 0 to round 4) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with UCB acquisition on Norman (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 3, row "UCB", column "Norman"
Configuration: NAIAD with UCB acquisition	Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4 Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)	97 True positives (the screen's 200 strongest measured gene pairs found among the 200 strongest measured-plus-predicted values, round 4 of active learning) count · higher Uncertainty: Not reported by the source: The table prints no error and does not state the number of runs. Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NAIAD with UCB acquisition on Norman (Tables 2 and 3, acquisition functions with NAIAD) pheno-20261022-qin2025-protocol-al-top200 Aggregation: Not reported Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Table 2, row "UCB", column "Norman"

Source checking is not independent reproduction. Release 2026-10-23-5347695500ed.

Research readiness

0 of 4 readiness checks met. These checks assess whether the evidence supports a reproducible investigation; a source-checked score alone does not meet them.

Readiness checks, gaps and artifacts

Release 2026-10-23-5347695500ed · Evidence verified: Not verified

Evidence incomplete

Replay metrics

Exact outcomes, predictions, identifiers and evaluator are connected.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
File checksums match the recorded files: not yet verified
Predictions are matched to the right samples: not yet verified
Score meaning and direction are confirmed: not yet verified
Metrics are recomputed from the saved predictions: not yet verified

Verified: Not verified

Evidence incomplete

Investigate discrepancies

Replay evidence includes annotations and an assessment of dependence. Unknown independence permits descriptive analysis only.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
File checksums match the recorded files: not yet verified
Predictions are matched to the right samples: not yet verified
Score meaning and direction are confirmed: not yet verified
Metrics are recomputed from the saved predictions: not yet verified
Sample annotations are recorded: not yet verified
Dependence between samples is assessed: not yet verified

Verified: Not verified

Evidence incomplete

Run locally

A pinned recipe describes the inputs, environment and resource requirements.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
File checksums match the recorded files: not yet verified
Predictions are matched to the right samples: not yet verified
Score meaning and direction are confirmed: not yet verified
A pinned run recipe exists: not yet verified
Compute requirements are estimated: not yet verified

Verified: Not verified

Evidence incomplete

Validate independently

Separate data and exposure records support an independent test.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
File checksums match the recorded files: not yet verified
Predictions are matched to the right samples: not yet verified
Score meaning and direction are confirmed: not yet verified
Independent validation data exist: not yet verified
Overlap with training data is checked: not yet verified

Verified: Not verified

Readiness describes the evidence in this release. Availability on your computer is checked separately when an investigation runs. Existing data exposure can prevent independent validation even when files are available.

Artifacts and reproduction

No verified artifact manifest is connected to this record yet. The gaps above identify what is needed before analysis can begin.

Evaluation procedure

Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4

Configuration: NAIAD with UCB acquisition
Protocol: Retrospective active learning on a fully measured two-gene CRISPR viability screen: top-200 gene pairs found by round 4
Dataset: Norman et al. 2019 combinatorial CRISPRa in K562, gene-level viability (as processed by Qin et al. 2025)

origin: Author-reported evaluation
configuration: Qin et al. 2025, PMLR 267 proceedings PDF
protocol id: pheno-20261022-qin2025-protocol-al-top200
dataset version: Gene-level viability as processed in Qin et al. 2025 Appendix A.1
split: No held-out split; the whole screen is both the candidate pool and the ground truth
population: 6,328 two-gene combinations of 112 genes
inputs: Single-gene viability effects of every gene, plus the measured viability of pairs acquired so far
adaptation: NAIAD ensemble; UCB acquisition
metric implementation: P_i = |X_i,P ∩ Y_i,P| at P = 200 (Appendix D)
aggregation: Not reported
budget: 100 uniformly sampled seed pairs, then 100 acquired pairs in each of 4 rounds (500 measured pairs at round 4)

Metadata review: source checked. Unreported conditions prevent automatic comparisons.

Reproduction

Split: No held-out split; the whole screen is both the candidate pool and the ground truth
Adaptation: NAIAD ensemble; UCB acquisition
Scoring implementation: P_i = |X_i,P ∩ Y_i,P| at P = 200 (Appendix D)

No execution recipe has been verified for this exact configuration and evaluation. A benchmark's general instructions may use different inputs, splits or model settings.

Reproducing this published result requires matching its model configuration, data, split and scorer. Source checking or a successful smoke test does not establish score reproduction.

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

21 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-23-5347695500ed
Property and statement	Original source and location	Review and provenance
Comparison: adaptation NAIAD ensemble; UCB acquisition Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.adaptation` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact
Comparison: aggregation Not reported Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	missing or unspecified No individual claim review recorded author reported Audit details Field: `attributes.comparison.aggregation` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact
Comparison: budget 100 uniformly sampled seed pairs, then 100 acquired pairs in each of 4 rounds (500 measured pairs at round 4) Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.budget` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact
Comparison: dataset version Gene-level viability as processed in Qin et al. 2025 Appendix A.1 Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.dataset_version` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact
Comparison: inputs Single-gene viability effects of every gene, plus the measured viability of pairs acquired so far Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.inputs` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact
Comparison: metric implementation P_i = \|X_i,P ∩ Y_i,P\| at P = 200 (Appendix D) Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.metric_implementation` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact
Comparison: population 6,328 two-gene combinations of 112 genes Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.population` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact
Comparison: protocol id pheno-20261022-qin2025-protocol-al-top200 Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.protocol_id` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact
Comparison: split No held-out split; the whole screen is both the candidate pool and the ground truth Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.split` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact
Evaluation group id qin2025-table2-3 Context-only references	Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) Original source ↗ Tables 2 and 3, row 'UCB', column 'Norman' Version: Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read. Retrieved: 2026-10-22T13:57:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.evaluation_group_id` Source artifact SHA-256: `983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb` Hash scope: pdftotext -layout text layer, parsed by extract/qin2025naiad.py Inspected artifact

How to read the evidence

Sources and history

Release 2026-10-23-5347695500ed · Record review: source checked

1 source record and release history

Qin et al. 2025, Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations (ICML 2025, PMLR 267) · Original source · Proceedings of the 42nd ICML, PMLR 267:50363-50383, 2025; proceedings PDF qin25g.pdf. The arXiv preprint 2411.12010 was not read.

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: pheno-20261022-qin2025-eval-table2-3-naiad-ucb-norman2019

areas: cells-tissues
contexts: research
origin: author_reported
protocol: pheno-20261022-qin2025-protocol-al-top200
version: Qin et al. 2025, PMLR 267 proceedings PDF
comparison: protocol id: pheno-20261022-qin2025-protocol-al-top200; dataset version: Gene-level viability as processed in Qin et al. 2025 Appendix A.1; split: No held-out split; the whole screen is both the candidate pool and the ground truth; population: 6,328 two-gene combinations of 112 genes; inputs: Single-gene viability effects of every gene, plus the measured viability of pairs acquired so far; adaptation: NAIAD ensemble; UCB acquisition; metric implementation: P_i = |X_i,P ∩ Y_i,P| at P = 200 (Appendix D); aggregation: Not reported; budget: 100 uniformly sampled seed pairs, then 100 acquired pairs in each of 4 rounds (500 measured pairs at round 4)
evaluation group id: qin2025-table2-3
evaluation group name: Tables 2 and 3, acquisition functions with NAIAD
retrospective: true
source locator: Tables 2 and 3, row 'UCB', column 'Norman'
review: method: source-hash-verification; scope-review; independent-cell-check; reviewer: claude; reviewer note: Claude (Opus 5.5) review agent, independent of the extraction agent; no human review claimed; date: 2026-10-22; artifact sha256: 983b2da8aed78f0aced27053301c15f0f67ebd0fac706435bb6800e309dd64eb; retrieval url: https://raw.githubusercontent.com/mlresearch/v267/main/assets/qin25g/qin25g.pdf; method note: Re-hashed the PDF in the scratchpad (SHA-256 matches the source record). Read the captions and methods text in the pdftotext layer and on the rendered pages.; note: Origin, comparison fields, group, locator and links checked against the table caption, Sections 4.2, 4.3 and 5.3 and Appendices B to D.
missing metadata: aggregation: reason: unreported; note: The table prints no error and does not state how many runs or replicates it summarises.

Related records

Suggest a correction