amortized-search

Tag

Cards List
#amortized-search

From Pareto to Preference: Personalized Test-Time Scaling via Amortized Agentic Policy Discovery

Hugging Face Daily Papers ↗ · 2026-10-07 Cached

提出了 PersonTTS,一个通过需求匹配的控制器初始化和源蒸馏程序化指导,实现个性化测试时缩放(多维用户需求联合满足)的摊销式智能体策略发现框架,在 AIME 和 HMMT 上显著优于现有 TTS 基线。

0 favorites 0 likes
← Back to home

Submit Feedback