A Missing Piece for Trustworthy AI Reviewers: From Benchmarking Rhetorical Robustness to SciCore Review

arXiv 2026 31.8 method

TLDR

Introduces rhetorical robustness for AI reviewers, a 1,260-version benchmark, and SciCore reviewer to improve stability and discrimination.

Reasoning

The paper provides a strong, rigorous benchmark and identifies an important failure mode in AI reviewers, supported by extensive evaluations. Its weaknesses include unclear generalizability beyond the tested configurations and limited detail on the SciCore construction process in the abstract.

Read-first score

Read-first score 31.8, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 10.

Recency 8%

100

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

Methodology quality 25%

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

Reproducibility 25%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Topical relevance 42%

8.3

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: 30.

Keyword Scores

AI for scientific research

research automation

automated research

AI scientist

automated scientific discovery

autonomous research agent

literature review agent

survey generation

automated experimentation

experiment design agent

paper writing agent

scientific discovery agent

Resources

Paper