读完 64,000 次赋值，最好的也只剩 13%

Ai2 每一级只多允许一种写法。四家前沿模型前两级满分，第四级最长的输入只剩 13%。下回看到「100 万上下文」，先问它做的是哪一级。1 2

1
StateBench: Testing how language models track state changes over long documents
allenai.org
2
A Fine-Grained Hierarchy and Benchmark for State Tracking in Language Models
arxiv.org

This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.

Comments

1/5

2026-11-05

Ai2 每一级只多允许一种写法。四家前沿模型前两级满分，第四级最长的输入只剩 13%。下回看到「100 万上下文」，先问它做的是哪一级。1 2

1
StateBench: Testing how language models track state changes over long documents
allenai.org
2
A Fine-Grained Hierarchy and Benchmark for State Tracking in Language Models
arxiv.org

This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.