Context-Picker: Dynamic context selection using multi-stage reinforcement learning
This addresses a key challenge in long-context QA for factoid questions, offering a dynamic method to improve answer quality by reducing noise, though it is incremental as it builds on existing retrieval and reinforcement learning techniques.
The paper tackled the problem of selecting appropriate context for long-context question answering by proposing Context-Picker, a framework that uses multi-stage reinforcement learning to identify minimal sufficient evidence subsets, resulting in higher answer accuracy on five datasets compared to strong RAG baselines.
In long-context question answering, selecting the appropriate scope of context for a query remains a key and unresolved challenge. Insufficient context can lead to missing essential information, whereas excessive context often introduces noise and degrades answer quality. Conventional methods, such as retrieving a fixed number of passages or applying reranking, struggle to dynamically determine which context to include. This is especially problematic for factoid questions, which typically depend only on a few precise pieces of evidence. To overcome this limitation, we propose Context-Picker, a reasoning-aware framework that reframes context selection as the task of identifying a minimal sufficient evidence subset, moving beyond conventional similarity-based ranking. Context-Picker uses a human-inspired two-stage reinforcement learning schedule: stage 1 focuses on improving the recall rate of critical passages, and stage 2 prioritizes pruning redundancy to distill a compact evidence set. To resolve reward sparsity, we propose an offline evidence distillation pipeline that mines ``minimal sufficient sets" via a Leave-One-Out (LOO) procedure, providing dense and task-aligned supervision. Experiments on five long-context and multi-hop QA datasets demonstrate that our method outperforms strong RAG baselines and achieved higher answer accuracy. Ablation studies also indicate that our coarse-to-fine optimization schedule, the redundancy-aware reward shaping, along with the rationale generated by the policy, all contribute substantially to these gains.