TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

First AI Benchmark Solved Before Release: The Zero Barrier Has Been Crossed

2 点作者 mrconter114 个月前

1 comment

mrconter114 个月前
Author here! While working on h-matched.com (tracking time between benchmark release and AI achieving human-level performance), I just added the first negative datapoint - LongBench v2 was solved 22 days before its public release.<p>This wasn&#x27;t entirely unexpected given the trend, but it raises fascinating questions about what happens next. The trend line approaching y=0 has been discussed before, but now we&#x27;re in uncharted territory.<p>Mathematically, we can make some interesting observations about where this could go: 1. It won&#x27;t flatten at zero (we&#x27;ve already crossed that) 2. It&#x27;s unlikely to accelerate downward indefinitely (that would imply increasingly trivial benchmarks) 3. It cannot cross y=-x (that would mean benchmarks being solved before they&#x27;re even conceived)<p>My hypothesis is that we&#x27;ll see convergence toward y=-x as an asymptote. I&#x27;ll be honest - I&#x27;m not entirely sure what a world operating at that boundary would even look like. Maybe others here have insights into what existence at that mathematical boundary would mean in practical terms?
评论 #42644115 未加载