Announcement
AnomalyXL generates synthetic data for long-context time-series RL
Prime Intellect says a 4B TimeRLM trained on fresh AnomalyXL data recursively called sub-agents to locate anomalies.
TLDR
Prime Intellect says its RL resident built AnomalyXL, a synthetic generator for long-context time series with programmatic ground truth. The company says a 4B TimeRLM trained with prime-rl on fresh data at every step, using nano-rlm to recursively call sub-agents and localize anomalies across 131k datapoints. It claims the model transfers zero-shot to ECG, sleep and observability data.
Combined views
—
1 Source, first seen ago
— likes— comments— saves— reposts
