Result

0.172 top-1-accuracy

lin2025-qwen-refined-civic top-1-accuracy (CIViC evidence level; mean over 10 iterations)

Tested configuration: Qwen 2.5 72B, refined prompt, temperature 0.8 (default) (Lin et al. 2025)
Protocol: Lin et al. 2025 CIViC evidence-level assignment
Dataset: CIViC clinical evidence summary, 4,426 variant-disease associations (accessed 2024-11-20)
Procedure: egfrnsclc-20261009-protocol-lin2025-civic-level-assignment
Evaluation: qwen-refined on civic (Lin et al. 2025)
Coverage: Not reported scored / Not reported eligible
Uncertainty: Not reported by the source
Evidence: Independent external evaluation · source checkedBenchmarking large language models GPT-4o, llama 3.1, and qwen 2.5 for cancer genetic variant classification · Table 2 (Tab2), row 8 ('Qwen2.5', 'Refined prompt + Default temperature (0.8)', 'CIViC'), column 'Accuracy'

A source-checked result verifies the numerical transcription, not every model or protocol detail. Evaluation metadata: source checked. Source checked does not mean independently reproduced.

Reproduction

Split: Whole table; no training split (prompted models)
Adaptation: Prompting only
Scoring implementation: Top-1 answer compared with the reference level

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.

1 evidence row matching the loaded filters

Claims, original sources and review scope · Release 2026-10-09-ba02f2f4a36e
Property and statement	Original source and location	Review and provenance
Reported result 0.1722 Individual claims	Benchmarking large language models GPT-4o, llama 3.1, and qwen 2.5 for cancer genetic variant classification Original source ↗ Table 2 (Tab2), row 8 ('Qwen2.5', 'Refined prompt + Default temperature (0.8)', 'CIViC'), column 'Accuracy' Version: npj Precision Oncology 9:141, published 2025-05-15; PMC12078457 full-text XML Retrieved: 2026-10-09T20:43:36Z	source checked ["source-hash-verification","deterministic-table-parse","independent-cell-check"] · 2026-10-09 independent paper Audit details Extracted by deterministic parse of the article JATS XML tables Tab1 and Tab2 (extract/extract_egfr_nsclc.py) with header and row labels asserted. Independent review 2026-10-09: value and identity match the source. Field: `attributes.printed_value` Source artifact SHA-256: `09c67fcaf7b367d74500db5fd015968389371dcb9ee5a37ccc83241aa63c0f80` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact Extraction artifact SHA-256: `09c67fcaf7b367d74500db5fd015968389371dcb9ee5a37ccc83241aa63c0f80` Extraction artifact

All evidence (JSONL) (gzip)How to read the evidence

Sources and history

Release 2026-10-09-ba02f2f4a36e · Record review: source checked

1 source records and release history

Benchmarking large language models GPT-4o, llama 3.1, and qwen 2.5 for cancer genetic variant classification · Original source · npj Precision Oncology 9:141, published 2025-05-15; PMC12078457 full-text XML

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: egfrnsclc-20261009-result-lin2025-qwen-refined-civic-top1

metric: top-1-accuracy
metric qualifier: CIViC evidence level; mean over 10 iterations
metric direction: higher
unit: fraction
printed value: 0.1722
numeric value: 0.1722
source locator: Table 2 (Tab2), row 8 ('Qwen2.5', 'Refined prompt + Default temperature (0.8)', 'CIViC'), column 'Accuracy'
review: method: source-hash-verification; deterministic-table-parse; independent-cell-check; method note: Read Table 1 and Table 2 from the article XML with a separate parser; checked value, 95% CI, p value, metric, qualifier, unit, direction and the linked configuration and protocol against the row and column headers.; reviewer: claude; reviewer note: Separate Claude review agent, independent of the extractor; no human review claimed; date: 2026-10-09; artifact sha256: 09c67fcaf7b367d74500db5fd015968389371dcb9ee5a37ccc83241aa63c0f80; retrieval url: https://www.ebi.ac.uk/europepmc/webservices/rest/PMC12078457/fullTextXML; note: Extracted by deterministic parse of the article JATS XML tables Tab1 and Tab2 (extract/extract_egfr_nsclc.py) with header and row labels asserted. Independent review 2026-10-09: value and identity match the source.
missing metadata: uncertainty: reason: unreported
scope note: Not marked best in its model and dataset group in Table 2.

Related records

evaluation: qwen-refined on civic (Lin et al. 2025)

Suggest a correction