CAPF: Guiding Search-Agent Rollouts with Credit-Attenuated Privileged Feedback
Recent LLM search agents use reinforcement learning with verifiable rewards (RLVR) to learn search-augmented reasoning from outcome rewards. On hard problems, these agents rarely sample end-to-end successful rollouts, leaving outcome-only RLVR with few positive-reward trajectories. We argue that improving learning on such problems requires additional guidance during training, and RLVR already contains verifier-side information that can provide it. This information can identify errors or omissions in the agent's submitted answer and guide revision within the rollout. We propose a training-time mechanism called \textbf{Credit-Attenuated Privileged Feedback} (CAPF), which makes this verifier-side information available through a Privileged Feedback call during training. CAPF lets the policy revise zero-reward attempts into positive-reward repair trajectories and attenuates credit for the feedback call and earlier actions to accommodate deployment without this call. Empirical research demonstrates that CAPF improves Qwen3-4B's average exact-match score from 44.7% under outcome-only RLVR to 48.5% on seven open-domain QA benchmarks.
Code (0)
등록된 구현이 없습니다.
Tasks
Reinforcement LearningSimilar Papers 제목 키워드 기반
TRIAGE: Role-Typed Credit Assignment for Agentic Reinforcement Learning
Agentic reinforcement learning requires assigning credit to environment-facing actions such as searches, clicks, edits, navigation commands, and object interactions. Standard GRPO uses the final verifier outcome as a uni…
Reinforcement LearningMetaCaptioner: Towards Generalist Visual Captioning with Open-source Suites
Generalist visual captioning goes beyond a simple appearance description task, but requires integrating a series of visual cues into a caption and handling various visual domains. In this task, current open-source models…
When Denser Credit Is Not Enough: Evidence-Calibrated Policy Optimization for Long-Horizon LLM Agent Training
Long-horizon LLM agents require reinforcement learning methods that can assign credit to intermediate decisions under sparse and delayed rewards. Recent group-based methods such as GiGPO improve over GRPO by constructing…
Reinforcement LearningAgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning
Reinforcement learning (RL) with verifiable rewards constructs trajectory-level advantage estimates, yet it often fails to credit the few pivotal decisions that determine outcomes in long-horizon, multi-turn agentic task…
Reinforcement LearningCapFrame: Text-Instructed Viewpoint Grounding in 3D Gaussian Scenes via Geometric Pseudo Labels
3D Gaussian Splatting (3DGS) enables photorealistic real-time novel view synthesis, yet placing a virtual camera to capture a desired frame remains largely manual. Existing language-guided approaches in 3D scenes mainly …
Novel View Synthesis