Contexts are Never Long Enough: Structured Reasoning for Scalable Question Answering over Long Document Sets

• URL: http://arxiv.org/abs/2604.22294v1
• Date: Fri, 24 Apr 2026 07:16:44 GMT
• Title: Contexts are Never Long Enough: Structured Reasoning for Scalable Question Answering over Long Document Sets
• Abstract summary: We present SLIDERS, a framework for question answering over long document collections through structured reasoning.<n> SLIDERS extracts salient information into a relational database, enabling scalable reasoning over persistent structured state.<n>It outperforms all baselines on three existing long-context benchmarks.
• Score: 7.102370558887478
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Real-world document question answering is challenging. Analysts must synthesize evidence across multiple documents and different parts of each document. However, any fixed LLM context window can be exceeded as document collections grow. A common workaround is to decompose documents into chunks and assemble answers from chunk-level outputs, but this introduces an aggregation bottleneck: as the number of chunks grows, systems must still combine and reason over an increasingly large body of extracted evidence. We present SLIDERS, a framework for question answering over long document collections through structured reasoning. SLIDERS extracts salient information into a relational database, enabling scalable reasoning over persistent structured state via SQL rather than concatenated text. To make this locally extracted representation globally coherent, SLIDERS introduces a data reconciliation stage that leverages provenance, extraction rationales, and metadata to detect and repair duplicated, inconsistent, and incomplete records. SLIDERS outperforms all baselines on three existing long-context benchmarks, despite all of them fitting within the context window of strong base LLMs, exceeding GPT-4.1 by 6.6 points on average. It also improves over the next best baseline by ~19 and ~32 points on two new benchmarks at 3.9M and 36M tokens, respectively.

Related papers

• EnSI-RAG: Entity-Structure-Indexed Retrieval-Augmented Generation for Long-Document Question Answering [15.557103631480445]
  We propose EnSI-RAG, a framework that constructs a query-independent, entity-centered index.<n>At query time, these records serve as retrieval handles, and an LLM synthesizes the retrieved passages into the final answer.<n>Across Loong and Oolong, EnSI-RAG achieves an average accuracy of 78.24.
  arXiv  Detail & Related papers  (2026-08-21T16:05:00Z)
• Structure then Query: Enabling Precise Analytical Queries over Unstructured Documents [11.689256498133446]
  Unstructured documents constitute the majority of enterprise and web data.<n>Mainstream retrieval methods still relies on fuzzy matching based on vector similarity.<n> AnnoIndex turns scattered unstructured text into a materialized, structured index that enables low-cost filtering and querying.
  arXiv  Detail & Related papers  (2026-08-13T15:46:20Z)
• M3DocDep: Multi-modal, Multi-page, Multi-document Dependency Chunking with Large Vision-Language Models [20.44093242908907]
  We propose M3DocDep, an LVLM-based pipeline that first recovers block-level dependencies and then constructs chunks along the recovered document tree.<n>Under a shared-block evaluation protocol, M3DocDep improves STEDS by +28.5 to +39.6 percent on DHP benchmarks, retrieval nDCG by +1.1 to +15.3 percent, and QA ANLS by +4.5 to +15.3 percent on corpus-level RAG benchmarks.
  arXiv  Detail & Related papers  (2026-04-17T05:33:45Z)
• MoDora: Tree-Based Semi-Structured Document Analysis System [62.01015188258797]
  Semi-structured documents integrate diverse interleaved data elements arranged in various and often irregular layouts.<n>MoDora is an LLM-powered system for semi-structured document analysis.<n> Experiments show MoDora outperforms baselines by 5.97%-61.07% in accuracy.
  arXiv  Detail & Related papers  (2026-02-26T14:48:49Z)
• Panini: Continual Learning in Token Space via Structured Memory [4.979820180013486]
  Language models are increasingly used to reason over content they were not trained on.<n>A common approach is retrieval-augmented generation (RAG), which stores verbatim documents externally (as chunks) and retrieves only a relevant subset at inference time.<n>We propose a human-like non-parametric continual learning framework, where the base model remains fixed, and learning occurs by integrating each new experience into an external semantic memory state.
  arXiv  Detail & Related papers  (2026-02-16T19:58:03Z)
• Aggregation Queries over Unstructured Text: Benchmark and Agentic Method [20.80318496130298]
  We formalize entity-level aggregation over text in a corpus-bounded setting with strict completeness requirement.<n>AGGBench is a benchmark designed to evaluate completeness-oriented aggregation under realistic large-scale corpus.<n>DFA is a modular agentic baseline that decomposes aggregation querying into interpretable stages.
  arXiv  Detail & Related papers  (2026-02-01T17:49:33Z)
• Beyond Isolated Dots: Benchmarking Structured Table Construction as Deep Knowledge Extraction [80.88654868264645]
  Arranged and Organized Extraction Benchmark designed to evaluate ability of large language models to comprehend fragmented documents.<n>AOE includes 11 carefully crafted tasks across three diverse domains, requiring models to generate context-specific schema tailored to varied input queries.<n>Results show that even the most advanced models struggled significantly.
  arXiv  Detail & Related papers  (2025-07-22T06:37:51Z)
• Mixture-of-RAG: Integrating Text and Tables with Large Language Models [5.038576104344948]
  Heterogeneous Document RAG requires joint retrieval and reasoning across textual and hierarchical data.<n>We propose MixRAG, a novel three-stage framework that preserves hierarchical structure and heterogeneous relationships.<n>Experiments show that MixRAG boosts top-1 retrieval by 46% over strong text-only, table-only, and naive-mixture baselines.
  arXiv  Detail & Related papers  (2025-04-13T13:02:33Z)
• Unstructured Evidence Attribution for Long Context Query Focused Summarization [53.08341620504465]
  We propose to extract unstructured (i.e., spans of any length) evidence in order to acquire more relevant and consistent evidence than in the fixed granularity case.<n>We show how existing systems struggle to copy and properly cite unstructured evidence, which also tends to be "lost-in-the-middle"
  arXiv  Detail & Related papers  (2025-02-20T09:57:42Z)
• HERA: Improving Long Document Summarization using Large Language Models with Context Packaging and Reordering [6.876612430571396]
  We propose a novel summary generation framework, called HERA.<n>We first segment a long document by its semantic structure and retrieve text segments about the same event, and finally reorder them to form the input context.<n>The experimental results show that HERA outperforms foundation models in ROUGE, BERTScore and faithfulness metrics.
  arXiv  Detail & Related papers  (2025-02-01T14:55:06Z)
• LLM$\times$MapReduce: Simplified Long-Sequence Processing using Large Language Models [73.13933847198395]
  We propose a training-free framework for processing long texts, utilizing a divide-and-conquer strategy to achieve comprehensive document understanding. The proposed LLM$times$MapReduce framework splits the entire document into several chunks for LLMs to read and then aggregates the intermediate answers to produce the final output.
  arXiv  Detail & Related papers  (2024-10-12T03:13:44Z)
• PDFTriage: Question Answering over Long, Structured Documents [60.96667912964659]
  Representing structured documents as plain text is incongruous with the user's mental model of these documents with rich structure. We propose PDFTriage that enables models to retrieve the context based on either structure or content. Our benchmark dataset consists of 900+ human-generated questions over 80 structured documents.
  arXiv  Detail & Related papers  (2023-09-16T04:29:05Z)
• Text Summarization with Latent Queries [60.468323530248945]
  We introduce LaQSum, the first unified text summarization system that learns Latent Queries from documents for abstractive summarization with any existing query forms. Under a deep generative framework, our system jointly optimize a latent query model and a conditional language model, allowing users to plug-and-play queries of any type at test time. Our system robustly outperforms strong comparison systems across summarization benchmarks with different query types, document settings, and target domains.
  arXiv  Detail & Related papers  (2021-05-31T21:14:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.