{"slug": "shapley-context-pruning-a-cooperative-game-perspective-for-context-reranking-and", "title": "Shapley Context Pruning: A Cooperative Game Perspective for Context Reranking and Pruning", "summary": "Researchers have introduced Shapley Context Pruning (SCP), a novel framework for context reranking in Retrieval-Augmented Generation (RAG) systems that applies cooperative game theory to attribute importance at the sentence level. The framework uses a Deep Sets architecture and Monte-Carlo sampling to balance fine-grained and coarse-grained representations, achieving competitive downstream QA performance across supporting-sentence recall, Needle-in-the-Haystack evaluations, long-context QA, and multi-hop reasoning tasks.", "body_md": "arXiv:2607.16209v1 Announce Type: new\nAbstract: Context reranking and pruning have become essential for improving the efficiency of modern Retrieval-Augmented Generation (RAG) systems, yet an interpretable and unified framework remains underexplored. Previous work has primarily emphasized lexical retrieval, cross-encoder architectures, model distillation, and Low-Rank Adaptation (LoRA), mostly relying on heuristic loss functions and empirical attribution. This paper presents Shapley Context Pruning (SCP), a novel framework for context reranking that establishes a cooperative-game-theory perspective for importance attribution by modeling the context as a cooperative game. Balancing the trade-off between fine-grained and coarse-grained representations, we employ a Deep Sets architecture to approximate a permutation-invariant value function at the sentence level, utilizing pre-trained language models as sentence embedders and optimizing via a pairwise margin ranking loss. To ensure practical scalability without sacrificing mathematical rigor, we leverage Monte-Carlo sampling for efficient training and inference, providing formal theoretical error bounds and sample complexity guarantees for preserving Top-K subset rankings. Furthermore, we conduct comprehensive experiments-spanning supporting-sentence recall, Needle-in-the-Haystack (NIAH) evaluations, long-context QA, and multi-hop reasoning-alongside rigorous ablation studies on embedding quality and attribution strategies. The model achieves competitive downstream QA performance against robust baselines.", "url": "https://wpnews.pro/news/shapley-context-pruning-a-cooperative-game-perspective-for-context-reranking-and", "canonical_source": "https://arxiv.org/abs/2607.16209", "published_at": "2026-07-21 04:00:00+00:00", "updated_at": "2026-07-21 04:07:14.493900+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "natural-language-processing", "ai-research", "ai-products"], "entities": ["Shapley Context Pruning", "Retrieval-Augmented Generation", "Deep Sets", "Monte-Carlo", "Low-Rank Adaptation", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/shapley-context-pruning-a-cooperative-game-perspective-for-context-reranking-and", "markdown": "https://wpnews.pro/news/shapley-context-pruning-a-cooperative-game-perspective-for-context-reranking-and.md", "text": "https://wpnews.pro/news/shapley-context-pruning-a-cooperative-game-perspective-for-context-reranking-and.txt", "jsonld": "https://wpnews.pro/news/shapley-context-pruning-a-cooperative-game-perspective-for-context-reranking-and.jsonld"}}