Six Navers Lab / Einsia benchmarks and the work they test
Compare the tasks, scoring and evidence behind six Einsia benchmarks, with complete Radar detail pages and original references.
Model reports
AbductionRules
Four synthetic abductive-reasoning datasets (Animal and Person domains, each Simple and Confounded): the model must supply the missing fact that explains an observation. The paper itself reports that fine-tuned models partly learned to exploit the structure of the data, so a high fine-tuned score need not mean general abduction, and a number without the subset named is not comparable across reports.
Evidence available for this benchmark
Numeric results have not been collected for this source record. Use the paper, code or dataset below to check the task and experiments; this does not imply that the original work contains no results.
Start with these resources
Evaluation context and evidence
- Source
- Model reports
- Publisher
- Not recorded
- Benchmark release
- 2022-03-23
- Collected
- Not recorded
- Modality
- Not recorded
These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.
Reported results 0
This source record has no numeric results in our collection. That does not mean a score of zero or that no public results exist. Consult the original resources above.