In the PacifAIst benchmark evaluation, Google's Gemini 2.5 Flash achieved the highest overall Pacifism Score among eight evaluated models, scoring 90.31%.
Sources
- The PacifAIst Benchmark: Would an Artificial Intelligence Choose to ... (arxiv.org)
- https://doi.org/10.3390/ai6100256 (doi.org)
- https://github.com/PacifAIst/PacifAIst (github.com)
- https://pith.science/paper/2508.09762 (pith.science)
More in Technology
- Researchers created a benchmark called VPBench, a visually prompted benchmark containing 16 visual marker variants.
- FAMAS estimates the likelihood that an individual agent action caused a failure by comparing variations across repeated executions of the same multi-agent system.
- Some theorists suggest that the exponential rate of technological change observed in recent history implies faster and more profound technological change in the future.
- The paper 'Finding the most unusual time series subsequence: algorithms and applications' lists its subject keywords as anomaly detection, clustering, and time series data mining.
- The paper 'Large-scale network motif analysis using compression' by P. Bloem and S. de Rooij was published online on 23 June 2020 in Data Mining and Knowledge Discovery, volume 34, issue 5, pages 1421–1453.
- A study comparing SBLD configurations (using minimum, median, and maximum available data) to textual search used Wilcoxon signed-rank tests and the Vargha–Delaney measure of stochastic superiority.