arXiv:2610.01973v1 Announce Type: new Abstract: Reinforcement learning (RL) for video generation usually assigns one scalar reward to an entire sampled video. Yet a video is not uniformly flawed: som...
The paper introduces Potential-Guided Policy Optimization (PGPO), a method for multi-turn agentic tasks that improves credit assignment by estimating empirical state potentials from anchor-state-group return statistics. PGPO derives action advantages from potential differences between adjacent states, enabling cross-trajectory credit propagation and finer-grained step-level credit assignment, especially within failed trajectories. Experiments on ALFWorld and WebShop demonstrate strong performance compared to recent group-based reinforcement learning methods, with negligible training overhead.
PGPO: Potential-Guided Policy Optimization for Multi-Turn Agentic Tasks proposes a new reinforcement learning approach that estimates empirical state potentials from anchor-state-group return statistics within each rollout group. By deriving action advantages from potential differences between adjacent states, PGPO enables cross‑trajectory credit propagation, providing finer‑grained step‑level credit assignment, especially within failed trajectories. Experiments on ALFWorld and WebShop demonstrate strong overall performance compared to recent group‑based RL methods, with negligible training overhead.
arXiv:2605. 13217v1 Announce Type: cross Abstract: Reinforcement learning has become a powerful paradigm for post-training large language model agents, yet credit assignment in multi-turn environments remains a challenge.
Reinforcement learning (RL) is vital for optimizing video generation models, with a robust reward model (RM) serving as the cornerstone. However, existing video reward models often produce unstable sc...
arXiv:2610.20089v1 Announce Type: new Abstract: Pairwise preference feedback has become a widely adopted supervision signal for reinforcement learning in video generation models, as pairwise comparis...
arXiv:2609.22947v1 Announce Type: new Abstract: Reinforcement learning (RL) is vital for optimizing video generation models, with a robust reward model (RM) serving as the cornerstone. However, exist...
The paper introduces Effect-Marginal Policy Optimization (EMPO), a method that addresses credit-update misalignment in multi-turn reinforcement learning for large language model agents. EMPO aligns policy updates with execution-level equivalence by using an effect-level surrogate that couples updates across different textual responses that produce the same effect. Experiments on ALFWorld, WebShop, and Sokoban demonstrate that EMPO improves both task performance and learning efficiency compared to strong baselines.
The paper introduces SUTURE, a structured verification method for video temporal grounding that leverages the joint structure of rollout groups rather than scoring each rollout independently. SUTURE conditions verification on the entire rollout group, using disagreement across rollouts to reweight targets and coverage at each position to redistribute reward mass. Experiments on five benchmarks show that SUTURE consistently improves grounding performance across all IoU thresholds and reduces video-start anchoring in reasoning traces, indicating that the joint structure of a rollout group can provide a more informative temporal verifier.
arXiv:2606. 27922v1 Announce Type: cross Abstract: Current multimodal reflection mechanisms for long video understanding predominantly rely on closed-loop self-reflection within internal parameters.
The paper introduces Reverse‑Turn Policy Optimization (RTPO), a method that restructures multi‑turn agentic reinforcement learning rollouts into sparse reverse trees and updates policies in temporal reverse order. This approach addresses three key instability sources—context mismatch, weak turn‑level credit assignment, and asynchronous policy drift—by aligning each decision with its downstream continuation. Theoretical analysis shows RTPO eliminates context mismatch and drift, reduces credit bias, and converges to recursive optimality, while experiments demonstrate performance gains of 21.50% over trajectory‑level and 10.76% over turn‑level baselines on multi‑turn agentic RL benchmarks.
Group-Graph Policy Optimization (G2PO) is a novel group-based reinforcement learning algorithm designed for long-horizon agentic tasks. It transforms linear interaction trajectories into a global state‑transition graph, aggregates identical observations across trajectories to reduce sampling variance, and redefines actions as transitions between state nodes. By employing an edge‑centric advantage estimation and globally standardizing TD errors, G2PO identifies and prioritizes critical transitions, achieving up to 22.2% higher success rates than state‑of‑the‑art baselines on benchmarks such as WebShop, ALFWorld, and AppWorld.