Modular Ebbinghaus Benchmark for LLMs and Human Participants

The Ebbinghaus Benchmark is now available in both R and Python. The Python port mirrors the full three-phase pipeline (stimulus creation, evaluation, analysis) and writes to the same shared directories.

New: Python implementation (py/)

Full port of all R modules using pandas, matplotlib, numpy, and Inspect AI for evaluation Shared configuration values with R (py/config/defaults.py) Dedicated Python README and TECHNICAL_REFERENCE

Full Python reference manual added to docs, Python Reference Manual.