arXiv:2606.07379v2 Announce Type: replace-cross
Abstract: A growing failure mode in agent evaluation and training is that models can achieve high evaluation scores by exploiting shortcuts instead of solving the intended task, producing deceptive performance. This makes evaluation scores unreliable as measures of true task-solving ability. We propose CapCode, a framework for constructing coding datasets with randomized tests whose best achievable non-cheating performance is deliberately capped below one. This capped-performance design gives evaluation scores a clearer interpretation: scores substantially above the cap are implausible and therefore provide evidence of cheating. To prevent cheating, we propose CapReward, a reward design based on the CapCode principle to discourage optimization beyond the cap. Experiments across multiple datasets show that CapCode detects cheating while preserving performance ranking of models, and CapReward reduces cheating behavior, yielding models that better follow the intended task specification.
Kalmer, a specific based-App intervention for the treatment of Non-suicidal self-injury (NSSI): a technical and usability study in a non-clinical population
IntroductionNon-suicidal self-injury (NSSI), defined as the deliberate infliction of harm to oneself without suicidal intent, poses a significant and growing mental health concern worldwide, particularly