[Paper] Impermanent: A Live Benchmark for Temporal Generalization in Time Series Forecasting

Summary

Recent time-series forecasting models, especially pre-trained foundation models, claim broad generalization, but current evaluation methods using static train-test splits are flawed. These static splits can lead to data contamination and inflated performance by allowing models to inadvertently train on or select based on test data. To address this, a new live benchmark called Impermanent has been introduced to provide more robust evidence for temporal generalization.

Continue Reading

Explore related coverage about research paper and adjacent AI developments: [Paper] Ruka-v2: Tendon Driven Open-Source Dexterous Hand with Wrist and Abduction for Robot Learning, [Paper] MedObvious: Exposing the Medical Moravec's Paradox in VLMs via Clinical Triage, [Paper] In-Place Test-Time Training, [Paper] HaloProbe: Bayesian Detection and Mitigation of Object Hallucinations in Vision-Language Models.

Related Articles

• [Paper] Ruka-v2: Tendon Driven Open-Source Dexterous Hand with Wrist and Abduction for Robot Learning

  March 30, 2026

• [Paper] MedObvious: Exposing the Medical Moravec's Paradox in VLMs via Clinical Triage

  March 25, 2026

• [Paper] In-Place Test-Time Training

  April 8, 2026

• [Paper] HaloProbe: Bayesian Detection and Mitigation of Object Hallucinations in Vision-Language Models

  April 8, 2026