Reproduce “GIFT-Eval-CTX: A Benchmark For Contextual Forecasting Model Evaluation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Xu Liu, Hanshu Yan, Taha Aksu, Quang Pham, Zhongzheng Qiao, Jun Hao Liew, Silvio Savarese, Doyen Sahoo and 1 more
GIFT-Eval-CTX tests whether forecasting models can turn numerical history and supplied descriptions of future trends and localized changes into accurate forecasts. Its future-derived Macro and Micro effects are paired with difficult held-out GIFT-Eval windows, isolating realization rather than context discovery. GPT-5.6 Sol’s aggregate relative MASE falls from 0.953 to 0.565, a 40.7% reduction, on selected benchmark instances.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.