Yuexi Pan
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
The Right Fact, the Wrong Decision: SciValidityBench for Scientific Agent Memory
According to the abstract, Pan and Chen introduce SciValidityBench, 64 task specifications testing whether scientific agents use remembered evidence only when it is valid to do so, for example distinguishing a pre-planned test from a selected result. They also present a framework that compiles history into decision-governing state.