Parallel Context-of-Experts Decoding for Retrieval Augmented Generation
On the map: Retrieval-augmented generation → RAG robustness and adaptive retrieval
also in 1 more
Abstract (arXiv)
Retrieval Augmented Generation faces a trade-off: concatenating documents in a long prompt enables multi-document reasoning but creates prefill bottlenecks, while encoding document KV caches separately offers speed but breaks cross-document interaction. We propose Parallel Context-of-Experts Decoding (Pced), a training-free framework that shifts evidence aggregation from the attention mechanism to the decoding. Pced treats retrieved documents as isolated "experts", synchronizing their predictions via a novel retrieval-aware contrastive decoding rule that weighs expert logits against the model prior. This approach recovers cross-document reasoning capabilities without constructing a shared attention across documents.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation. Builds on 20 of its 31 references that are in the map. First posted on arXiv 13 Jan 2026.
Builds on
20 of 31 references.
Closest ancestors · 1
Same direction, newest first
Other references · 19
- 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
- 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
- 2022Classifier-Free Diffusion Guidancecited by 5,847
- 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
- 2019The Curious Case of Neural Text Degenerationcited by 2,916
- 2023LongBench: A Bilingual, Multitask Benchmark for Long Context Understandingcited by 1,417
- 2024DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Servingcited by 617
- 2022Contrastive Decoding: Open-ended Text Generation as Optimizationcited by 609
- 2023Trusting Your Evidence: Hallucinate Less with Context-aware Decodingcited by 327
- 2024CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusioncited by 223
- 2025LMCache: An Efficient KV Cache Layer for Enterprise-Scale LLM Inferencecited by 126
- 2024Long-Context Language Modeling with Parallel Context Encodingcited by 84
7 more
- 2022Parallel Context Windows for Large Language Modelscited by 81
- 2025KVLink: Accelerating Large Language Models via Efficient KV Cache Reusecited by 66
- 2024TurboRAG: Accelerating Retrieval-Augmented Generation with Precomputed KV Caches for Chunked Textcited by 63
- 2024Don't Do RAG: When Cache-Augmented Generation is All You Need for Knowledge Taskscited by 38
- 2024AdaCAD: Adaptively Decoding to Balance Conflicts between Contextual and Parametric Knowledgecited by 37
- 2025APE: Faster and Longer Context-Augmented Generation via Adaptive Parallel Encodingcited by 29
- 2024Entropy-Based Decoding for Retrieval-Augmented Large Language Modelscited by 19
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2026Guaranteeing Knowledge Integration with Joint Decoding for Retrieval-Augmented Generation
- 2026Retrieval from Within: An Intrinsic Capability of Attention-Based Models
- 2022Multi-Task Retrieval-Augmented Text Generation with Relevance Sampling
- 2026Dual-Confidence Contrastive Decoding for Retrieval-Augmented Generation
- 2025LFD: Layer Fused Decoding to Exploit External Knowledge in Retrieval-Augmented Generation
- 2025Direct Retrieval-augmented Optimization: Synergizing Knowledge Selection and Language Models
- 2026MoG: Mixture of Experts for Graph-based Retrieval-Augmented Generation
- 2024Adaptive Contrastive Decoding in Retrieval-Augmented Generation for Handling Noisy Contexts