RISED: Rubrics for Agentic Multi-Environment Selection and Self-Distillation

Key points

• Training a single LLM agent jointly across diverse interactive environments has attracted increasing attention as a route to generalist agents.1
• We introduce a rubric-based reward framework that generates query-specific rubrics grounded in retrieved evidence and decomposed into multiple quality dimensions, providing fine-grained supervision during...1
• On-policy distillation post-trains large language model agents by supplying dense, step-level guidance from a teacher policy when the reinforcement-learning reward is sparse and arrives only once per trajectory.2
• To this end, we introduce GraphOPD, the first method to bring graph-based structural augmentation into on-policy distillation for agent capabilities.2

Sources (2)

• [1]RISED: Rubrics for Agentic Multi-Environment Selection and Self-Distillation
  Apple Machine Learning Research · Oct 6, 12:00 AM

  RISED: Rubrics for Agentic Multi-Environment Selection and Self-Distillation

  Training a single LLM agent jointly across diverse interactive environments has attracted increasing attention as a route to generalist agents.

• [2]GraphOPD: Graph-Augmented On-Policy Distillation for LLM Agents
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 06:25 PM

  On-policy distillation post-trains large language model agents by supplying dense, step-level guidance from a teacher policy when the reinforcement-learning reward is sparse and arrives only once per trajectory.

  To this end, we introduce GraphOPD, the first method to bring graph-based structural augmentation into on-policy distillation for agent capabilities.

Extractive summary: sentences quoted from the sources.