Knowledge-Intensive QA

QA: Question Answering

Knowledge-intensive question answering studies how language models can find and use information beyond what they reliably recall from their parameters, including across documents, tables, knowledge graphs, and changing sources. Common approaches include sparse, dense, and hybrid retrieval-augmented generation; hierarchical or graph-based knowledge organization; and agents that search, inspect evidence, and use tools over multiple steps. Training methods also aim to improve parametric knowledge access or teach models when to verify, revise, or abstain, with evaluation spanning open-domain benchmarks such as TriviaQA and HotpotQA as well as long-document and enterprise tasks. A central challenge is turning relevant retrieved material into complete, well-supported answers at manageable cost, while reliably detecting missing or conflicting evidence.

Latest papers

74 papers4 papers in the last four weeks, down 64% on the four weeks before. 0.0% of all new papers.

DocSage: An Information Structuring Agent for Multi-Doc Multi-Entity Question Answering
Mar 12, 2026Teng Lin, Yizhang Zhu, Zhengxuan Zhang +2Multi-Document QA Multi-Hop QA
The Hong Kong University of Science and Technology (Guangzhou) · The Hong Kong University of Science and Technology
Citation-Enforced RAG for Fiscal Document Intelligence: Cited, Explainable Knowledge Retrieval in Tax Compliance
Mar 11, 2026Akhil Chandra ShanivendraDocument QA Knowledge-Intensive QA
Independent Researcher
MITRA: An AI Assistant for Knowledge Retrieval in Physics Collaborations
Mar 10, 2026Abhishikth Mallampalli, Sridhara DasuKnowledge-Intensive QA AI for Science
University of Wisconsin-Madison
EnComp: Lightweight Encoder-Only Context Compression for Retrieval-Augmented Question Answering
Mar 10, 2026Thao Do, Dinh Phu Tran, An Vo +2Efficient Language Model Inference Knowledge-Intensive QA
Korea Advanced Institute of Science and Technology (KAIST), Daejeon 34141, South Korea
KohakuRAG: A simple RAG framework with hierarchical document indexing
Mar 8, 2026Shih-Ying Yeh, Yueh-Feng Ku, Ko-Wei Huang +1Retrieval-Augmented Generation Structured RAG
National Tsing Hua University · Comfy Org Research · Kohaku-Lab
iAgentBench: Benchmarking Sensemaking Capabilities of Information-Seeking Agents on High-Traffic Topics
Mar 4, 2026Preetam Prabhu Srikar Dammu, Arnav Palkhiwala, Tanya Roosta +1AI Agent Benchmarks LLM Agent Evaluation
University of Washington Seattle, Washington, USA · UC Berkeley Berkeley, California, USA
$τ$ -Knowledge: Evaluating Conversational Agents over Unstructured Knowledge
Mar 4, 2026Quan Shi, Alexandra Zytek, Pedram Razavi +2LLM Agent Evaluation Agent Evaluation
Sierra · Princeton University
GSI Agent: Domain Knowledge Enhancement for Large Language Models in Green Stormwater Infrastructure
Mar 3, 2026Shaohuang WangKnowledge Augmentation for Language Models Knowledge-Intensive QA
Super Research: Answering Highly Complex Questions with Large Language Models through Super Deep and Super Wide Research
Feb 28, 2026Yubo Dong, Nianhao You, Yuxuan Hou +5Autonomous Research Agents LLM Evaluation
College of Computer Science and Technology · Ant Group, Hangzhou, China
Improving Parametric Knowledge Access in Reasoning Language Models
Feb 25, 2026Melody Ma, John HewittFactual Knowledge in Language Models Knowledge-Intensive QA
Columbia University
Controllable Evidence Selection in Retrieval-Augmented Question Answering via Deterministic Utility Gating
Feb 23, 2026Victor P. UndaEvidence Selection Retrieval-Augmented Generation
Independent Researcher, Applied AI Systems
MAPLE: Metadata Conditioned LLM Pretraining for Locale-Aware Question Answering
Jan 21, 2026Anjishnu Mukherjee, Ziwei Zhu, Antonios AnastasopoulosLanguage Model Pretraining LLM Evaluation
George Mason University
Instruction Retrieval at Inference Time for Small Language Models
Oct 15, 2025Kenan Alkiek, David Jurgens, Vinod VydiswaranKnowledge Augmentation for Language Models Small Language Models
School of Information University of Michigan Ann Arbor, MI 48109, USA
KnowMT-Bench: Benchmarking Knowledge-Intensive Long-Form Question Answering in Multi-Turn Dialogues
Sep 26, 2025Junhao Chen, Yu Huang, Siyuan Li +7Knowledge-Intensive QA Factual Knowledge in Language Models
Hong Kong University of Science and Technology (Guangzhou) · The University of Osaka · Hong Kong University of Science and Technology

Mar 12, 2026J/K move · Enter open · S save

DocSage: An Information Structuring Agent for Multi-Doc Multi-Entity Question Answering

Teng Lin, Yizhang Zhu, Zhengxuan Zhang, Yuyu Luo+1

The Hong Kong University of Science and Technology (Guangzhou) · The Hong Kong University of Science and Technology

Multi-Document QA Multi-Hop QA Knowledge-Intensive QA Agentic RAG Relational Reasoning LLM Information Extraction

Open paper AskPDF arXiv

Multi-document Multi-entity Question Answering inherently demands models to track implicit logic between multiple entities across scattered documents. However, existing Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) frameworks suffer from critical limitations: standard RAG's vector similarity-based coarse-grained retrieval often omits critical facts, graph-based RAG fails to efficiently integrate fragmented complex relationship networks, and both lack schema awareness, leading to inadequate cross-document evidence chain construction and inaccurate entity relationship deduction. To address these challenges, we propose DocSage, an end-to-end agentic framework that integrates dynamic schema discovery, structured information extraction, and schema-aware relational reasoning with error guarantees. DocSage operates through three core modules: (1) A schema discovery module dynamically infers query-specific minimal joinable schemas to capture essential entities and relationships; (2) An extraction module transforms unstructured text into semantically coherent relational tables, enhanced by error-aware correction mechanisms to reduce extraction errors; (3) A reasoning module performs multi-hop relational reasoning over structured tables, leveraging schema awareness to efficiently align cross-document entities and aggregate evidence. This agentic design offers three key advantages: precise fact localization via SQL-powered indexing, natural support for cross-document entity joins through relational tables, and mitigated LLM attention diffusion via structured representation. Evaluations on two MDMEQA benchmarks demonstrate that DocSage significantly outperforms state-of-the-art long-context LLMs and RAG systems, achieving more than 27% accuracy improvements respectively.