Termsslug: process-reward

Process Reward

过程奖励

Supervision on intermediate reasoning steps rather than only final answers, separating valid searches from critical errors.

Curated Intelligence

1 items · Page 1/1
量子位 (微信公众号)T2·Agents
74 pts
Alibaba Neos Open-Sources 9B Multimodal Search Agent Matching Closed Models

Alibaba's Accio team, with Fudan University and Shanghai Innovation Institute, introduces Neos, a unified training framework that couples data construction, process rewards and environment co-evolution. Its 9B and 35B models approach frontier closed-source performance on four multimodal search benchmarks by building multi-hop tasks with visual checkpoints and filtering out shortcut solutions.