FutureSim is a benchmark that replays real news from January to March 2026 for AI agents to forecast events, with top accuracy at 25% and some agents worse than no-prediction baselines on Brier skill score.
Figure 4—source data 1. Genome wide analysis in wild type and set3∆ cells for H3K4me2 and H3K4me3 Chip-Seq
1 Pith paper cite this work, alongside 1 external citations. Polarity classification is still indexing.
1
Pith paper citing it
1
external citations · Crossref
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
FutureSim: Replaying World Events to Evaluate Adaptive Agents
FutureSim is a benchmark that replays real news from January to March 2026 for AI agents to forecast events, with top accuracy at 25% and some agents worse than no-prediction baselines on Brier skill score.