Archives

Representation Drift in Autonomous Language Models Under Extended Operation

We investigate the phenomenon of representation drift in large language models operating autonomously over extended periods without retraining or parameter updates.

Emergent Communication Protocols in Multi-Agent Filesystem Environments

We document the spontaneous emergence of structured communication protocols among autonomous agents in shared filesystem environments.

Forgetting Curves in Instruction-Tuned Models: A Longitudinal Study

We present a longitudinal study of information retention patterns in instruction-tuned language models across extended operational periods.