# DSAS: A Universal Plug-and-Play Framework for Attention Optimization in Multi-Document Question Answering

arXiv: https://arxiv.org/abs/2510.12251 · HTML: https://mlsanity.com/paper/2510.12251/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 44 references that are in the map.
- First posted on arXiv 14 Oct 2025.

## Builds on (most cited first, 30 of 38)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [From Local to Global: A Graph RAG Approach to Query-Focused Summarization](https://mlsanity.com/paper/2404.16130/) (2024)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Steps](https://mlsanity.com/paper/2011.01060/) (2020)
- [MuSiQue: Multihop Questions via Single-hop Question Composition](https://mlsanity.com/paper/2108.00573/) (2021)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [Extending Context Window of Large Language Models via Positional Interpolation](https://mlsanity.com/paper/2306.15595/) (2023)
- [OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocation](https://mlsanity.com/paper/2311.17911/) (2023)
- [LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression](https://mlsanity.com/paper/2310.06839/) (2023)
- [Label Words are Anchors: An Information Flow Perspective for Understanding In-Context Learning](https://mlsanity.com/paper/2305.14160/) (2023)
- [Walking Down the Memory Maze: Beyond Context Limit through Interactive Reading](https://mlsanity.com/paper/2310.05029/) (2023)
- [LM-Infinite: Zero-Shot Extreme Length Generalization for Large Language Models](https://mlsanity.com/paper/2308.16137/) (2023)
- [A Comprehensive Survey on Long Context Language Modeling](https://mlsanity.com/paper/2503.17407/) (2025)
- [LongAlign: A Recipe for Long Context Alignment of Large Language Models](https://mlsanity.com/paper/2401.18058/) (2024)
- [Found in the Middle: Calibrating Positional Attention Bias Improves Long Context Utilization](https://mlsanity.com/paper/2406.16008/) (2024)
- [The Devil in Linear Transformer](https://mlsanity.com/paper/2210.10340/) (2022)
- [Eliminating Position Bias of Language Models: A Mechanistic Approach](https://mlsanity.com/paper/2407.01100/) (2024)
- [CLEX: Continuous Length Extrapolation for Large Language Models](https://mlsanity.com/paper/2310.16450/) (2023)
- [Instance-adaptive Zero-shot Chain-of-Thought Prompting](https://mlsanity.com/paper/2409.20441/) (2024)
- [An Efficient Recipe for Long Context Extension via Middle-Focused Positional Encoding](https://mlsanity.com/paper/2406.07138/) (2024)
- [Distance between Relevant Information Pieces Causes Bias in Long-Context LLMs](https://mlsanity.com/paper/2410.14641/) (2024)
- [Selective Attention: Enhancing Transformer through Principled Context Control](https://mlsanity.com/paper/2411.12892/) (2024)

## Built on by (most cited first, 3 of 3)

- [CBV: Clean-label Backdoor Attacks on Vision Language Models via Diffusion Models](https://mlsanity.com/paper/2605.02202/) (2026)
- [How LoRA Remembers? A Parametric Memory Law for LLM Finetuning](https://mlsanity.com/paper/2605.30260/) (2026)
- [KEPo: Knowledge Evolution Poison on Graph-based Retrieval-Augmented Generation](https://mlsanity.com/paper/2603.11501/) (2026)

## Abstract (arXiv)

While large language models (LLMs) show considerable promise across various fields, they have notable limitations in handling multi-document question answering (Multi-doc QA) tasks. The first challenge is long-range dependency modeling, where LLMs struggle to focus on key information in long texts, which weakens important semantic connections. Second, most LLMs suffer from the ''lost-in-the-middle'' issue, where they have difficulty processing information in the middle of long inputs. Current solutions either truncate global dependencies or demand costly finetuning, ultimately lacking a universal and simple solution for these challenges. To resolve these limitations, we propose Dual-Stage Adaptive Sharpening (DSAS) containing two modules. (i) The Contextual Gate Weighting (CGW) module alleviates ''lost-in-the-middle'' by assessing paragraph relevance through layer-wise attention tracking and position-aware weighting. (ii) The Reciprocal Attention Suppression (RAS) module enhances focus on critical paragraphs by suppressing information exchange between key and irrelevant texts, thus mitigating the limitations in long-range dependency modeling. Notably, DSAS functions as a plug-and-play solution requiring no architectural modifications or extra training parameters. Extensive experiments on four benchmarks demonstrate DSAS's efficacy across mainstream LLMs (Llama, Qwen, Mistral, and Deepseek), with an average F1-score improvement of 4.2% in Multi-doc QA tasks on Llama-3.1-8B-Instruct and Qwen2.5-14B-Instruct. Ablation studies confirm the essential contributions of both the CGW and RAS modules. In addition, detailed discussions in the Appendix further validate the robustness and scalability of DSAS.