3 papers
cs.CL2026
Temporal Leakage in Search-Engine Date-Filtered Web Retrieval: A Retrospective Forecasting Case Study
Ali El Lahib, Ying-Jieh Xia, Zehan Li +2
Search-engine date filters are widely used to enforce pre-cutoff retrieval in retrospective evaluations of search-augmented forecasters. We show this approach is unreliable across…
cs.LG2026
HISA: Efficient Hierarchical Indexing for Fine-Grained Sparse Attention
Yufei Xu, Fanxu Meng, Fan Jiang +11
Token-level sparse attention mechanisms, exemplified by DeepSeek Sparse Attention (DSA), achieve fine-grained key selection by scoring every historical key for each query through a…
cs.CL2026
Simulated Ignorance Fails: A Systematic Study of LLM Behaviors on Forecasting Problems Before Model Knowledge Cutoff
Zehan Li, Yuxuan Wang, Ali El Lahib +2
Evaluating LLM forecasting capabilities is constrained by a fundamental tension: prospective evaluation offers methodological rigor but prohibitive latency, while retrospective for…