Saved in:
Bibliographic Details
Main Authors: Tang, Yihong, Williams, Andrew Robert, Ashok, Arjun, Zheng, Vincent Zhihao, Sun, Lijun, Drouin, Alexandre, Laradji, Issam H., Marcotte, Étienne, Zantedeschi, Valentina
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.27904
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Time series forecasting in real-world settings often depends not only on historical observations, but also on external context that must be actively discovered from noisy, heterogeneous information sources. Yet existing context-aided forecasting benchmarks typically assume that the supporting context is already provided, leaving open whether agents can identify it on their own. Therefore, we introduce Dr-CiK, a benchmark for evaluating whether agents can retrieve forecasting-relevant supporting context from a document corpus, filter out distractors, distill the retrieved context into forecast-useful evidence, and generate forecasts supported by that evidence. Through context ablations and evaluations of state-of-the-art deep research and forecasting methods paired together, we show that high-quality context substantially improves forecasting performance in Dr-CiK. However, most existing DR agents recover only a small fraction of the ground-truth supporting evidence (usually <5%), are frequently misled by distractors (>80% distractor citations), and can cause forecasters to perform worse with retrieved context than without context. Our results motivate research on foresight-driven agents that search for the right context to predict the future.