Loading the edition…
· Past edition · 5 stories
Retrospective edition reconstructed from later captures. Paper versions may include later updates; coverage labels describe the evidence used.
Research & academia
A paper distinguishes an undetected benchmark leak from a low-power audit.
Why it matters: Contamination audits need adequate statistical power; small-sample evidence may be less decisive than it appears.
Research & academia
WorldTrace assigns compressed memory slots virtual positions and evaluates temporal consistency and scene recall using LoopBench.
Open each story in the reader. Summaries are shown when available; some sources provide an abstract only. With an account you can keep your own notes.
Why it matters: Long-horizon video models need ways to retrieve earlier visual context.