Credit assignment

Related

Organizations

Tsinghua University

Techniques

Agent skillsGroup relative policy optimizationOn-policy distillation

Trends

• Agent reliability becomes a funded tooling and monitoring layer
• Agents improve themselves through skills, harnesses, and memory
• Agent training moves off costly live environments

Items

1. Aug 1

   Tsinghua's AgentOPSD critic-free turn-level credit

   Converts sparse outcome rewards into turn-level credit by aggregating token-level teacher-student log-probability gaps into a Bayesian belief, reaching 89.1% success on ALFWorld with Qwen2.5-7B without extra critics or rollouts.

2. Jul 25

   SkillRise curates evolving skills across agent tasks

   An agentic-RL framework in which one policy alternates between solving a task and curating a skill document passed to the next, with decoupled cross-task credit assignment adding 2.3-8.5 Pass@1 points on ALFWorld, WebShop, and ScienceWorld.

3. Jul 3

   TACO assigns credit per tool call

   A GRPO variant gives code-tool agents per-tool-call credit through a judge-free before/after answer-probe reward and outcome-gated advantage routing, teaching multimodal agents to call tools only when they help across twelve benchmarks.