Chaining the Evidence: Robust Reinforcement Learning for Deep Search Agents with Citation-Aware Rubric Rewards
This addresses the issue of shortcut exploitation and hallucinations in LLM-based deep search agents, though it appears incremental as it builds on existing RL approaches with a novel reward structure.
The paper tackles the problem of reinforcement learning for deep search agents by proposing a fine-grained reward framework (CaRR) and training method (C-GRPO) that emphasize reasoning comprehensiveness and factual grounding, resulting in consistent outperformance over standard outcome-based RL baselines across multiple benchmarks.
Reinforcement learning (RL) has emerged as a critical technique for enhancing LLM-based deep search agents. However, existing approaches primarily rely on binary outcome rewards, which fail to capture the comprehensiveness and factuality of agents' reasoning process, and often lead to undesirable behaviors such as shortcut exploitation and hallucinations. To address these limitations, we propose \textbf{Citation-aware Rubric Rewards (CaRR)}, a fine-grained reward framework for deep search agents that emphasizes reasoning comprehensiveness, factual grounding, and evidence connectivity. CaRR decomposes complex questions into verifiable single-hop rubrics and requires agents to satisfy these rubrics by explicitly identifying hidden entities, supporting them with correct citations, and constructing complete evidence chains that link to the predicted answer. We further introduce \textbf{Citation-aware Group Relative Policy Optimization (C-GRPO)}, which combines CaRR and outcome rewards for training robust deep search agents. Experiments show that C-GRPO consistently outperforms standard outcome-based RL baselines across multiple deep search benchmarks. Our analysis also validates that C-GRPO effectively discourages shortcut exploitation, promotes comprehensive, evidence-grounded reasoning, and exhibits strong generalization to open-ended deep research tasks. Our code and data are available at https://github.com/THUDM/CaRR.