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。

Show HN: Predict LLM Predictions

3 点作者 flowingfocus3 个月前

1 comment

ggm3 个月前
Adversarial play might be interesting. I&#x27;m assuming the goal here is to use the feedback as training input, fitness to expectations and like.<p>So a random chance of a bad actor adversary? Or high scores for lying? &quot;The traitor&quot; style<p>Team models? I suppose I&#x27;m arguing for gamification with a leader board and hubristic bragging rights.
评论 #42981150 未加载