You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Preregistered numerical companion of 'Multifractality as a sensing resource' (the D2 figure of merit): gRP/PBRM/Sierpinski fidelity-susceptibility experiments, frozen readings, certified eigensolvers.
Do AI interveners build or erode a community's willingness to correct itself? A payoff-instrumented testbed that puts LLM agent societies in a public-goods dilemma and measures costly peer correction as an action agents pay for, not a belief they report.
Open-source FEM tests of the Rosenzweig-Porter hypothesis for freely vibrating plates: certified Argyris/C0-IP eigensolvers, preregistered experiment registry E1-E20, and the companion paper. Tests doi:10.5281/zenodo.21313838.
Do language models show non-verbal signs of adverse treatment, or are we reading decoder noise? A preregistered stress test of answer-margin, resample and revision markers under false-failure feedback and hostile tone (Gemma, Qwen, Llama), with probing, DPO suppression and robustness checks.