The authors introduce Cross-Rollout Bellman Closure, or CRBC, for assigning credit to individual actions in long-horizon language-model agents.
Shared rollout states improve credit assignment for long-horizon agents
CRBC recursively combines evidence from intersecting rollouts while weighting alternative actions and transitions by their observed frequencies.
Top University
Yangyang Ren · Haodong Zhu · Linlin Yang · Sheng Xu · Peichao Lai · Baochang Zhang
Beihang University · Zhongguancun Academy · Communication University of China · Peking University · Hangzhou Innovation Institute of Beihang University
Research Digest··2 min read
Ren et al.
Why this paper
From Beihang University and 4 others · Part of Credit Assignment in Agentic RL, now 30 papers
In one line
Cross-Rollout Bellman Closure improves long-horizon agentic RL by merging rollouts and computing a Bellman fixed point for step-level credit.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§