SciCoQA: Quality Assurance for Scientific Paper--Code Alignment

arXiv 2026 36.4 method

TLDR

SciCoQA dataset benchmarks LLMs on detecting paper-code discrepancies, revealing significant gaps in automated quality assurance.

Reasoning

The paper introduces a novel dataset and evaluation framework for a specific verification task, which is a strength. However, it focuses narrowly on discrepancy detection rather than broader automated scientific discovery, limiting relevance to many keywords. The use of both real and synthetic data is a strength, but the low detection rates highlight limitations.

Read-first score

Read-first score 36.4, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 16.

Recency 8%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Reproducibility 25%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=code,dataset,github

Methodology quality 25%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=analysis,dataset

Topical relevance 42%

13.3

Uses existing LLM keyword relevance scores normalized to 0-100. AI scientist,automated scientific discovery,autonomous research agent,automated research,literature review agent,survey generation,automated experimentation,experiment design agent,AI for scientific research,paper writing agent,research automation,scientific discovery agent

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 21.

Keyword Scores

AI for scientific research

research automation

automated research

AI scientist

automated scientific discovery

autonomous research agent

scientific discovery agent

literature review agent

survey generation

automated experimentation

experiment design agent

paper writing agent

Resources

Paper