This work proposes MARS, a scalable knowledge graph question answering (KGQA) approach that requires no model fine-tuning, and performs a structured retrieval procedure that links question entities to the KG and iteratively retrieves relevant next-hop information.
Abstract
Large language models (LLMs) have demonstrated strong reasoning performance, but their tendency to hallucinate limits their reliability in knowledge-intensive tasks requiring up-to-date and grounded information. Combining knowledge graphs (KGs) with LLMs facilitates the use of explicit symbolic knowledge that can be continuously updated without costly fine-tuning, while benefiting from rapidly advancing LLM reasoning. We propose MARS, a scalable knowledge graph question answering (KGQA) approach that requires no model fine-tuning. Rather than relying on open-ended agentic exploration, MARS performs a structured retrieval procedure that links question entities to the KG and iteratively retrieves relevant next-hop information. At each step, MARS decides whether to continue graph traversal or to generate the final SPARQL query, allowing the model to adapt the retrieval depth to the question while keeping the overall pipeline more predictable than fully agentic approaches. We evaluate MARS on three established KGQA benchmarks across several LLMs and settings, including multilingual evaluation, and provide insights through ablation studies and error analysis. Our approach achieves competitive performance relative to state-of-the-art methods while remaining efficient and scalable. The evaluation results, code and resources are publicly available: https://github.com/dice-group/mars-kgqa.
Large language models can answer knowledge-intensive questions more reliably when they are grounded with knowledge graphs, but systems such as Think-on-Graph and Reasoning-on-Graph repeatedly query the same graph neighborhoods across different questions. In this work, we study this repeated retrieval in Knowledge Graph Question Answering~(KGQA) workloads and propose KGCache, an in-memory cache for one-hop knowledge graph neighborhoods. KGCache is designed to be compatible with both iterative traversal (ToG) and one shot planning (RoG) KGQA paradigms. KGCache is placed between the KGQA engine and the backend serving the KG, so repeated entity requests can be served from cache instead of issuing new KG queries. We evaluate KGCache on WebQSP and CWQ using LRU, LFU, and a trace-aware Oracle policy. Our analysis shows that both datasets contain substantial entity reuse among starting entities and entities reached during traversal. We also explore semantic caching for similar queries, which shows additional hit-rate gains on WebQSP and needs further accuracy testing on CWQ. Entity caching accelerates KG retrieval by up to $1.91\times$, while semantic-context caching achieves up to $1.06\times$ full-system speedup in the evaluated WebQSP configurations, with each hit being up to $3.73\times$ faster.
Uros Stanic, Chang-He Yuan, Sabuj Laskar et al.· 0 citations
Integrating Knowledge Graphs (KGs) into Retrieval-Augmented Generation (RAG) can substantially improve LLM performance on complex question answering (QA) by reducing hallucinations and supplying structured context. However, building high-quality KGs over large corpora for edge scenarios is challenging: cloud-based processing introduces latency and dependency on remote services, while exhaustive on-device construction with LLMs is often computationally infeasible under limited hardware budgets. We observe that traditional non-LLM methods can efficiently capture explicit knowledge, and that real-world queries typically touch only a small, highly concentrated portion of the graph. As a result, static and exhaustive KG construction is redundant and inefficient. We propose Edge-AdaptiveKG, a resource-aware framework that combines an offline Seed KG (S-KG) with an online Query-driven KG (Q-KG). Lightweight non-LLM methods build the S-KG, while the LLM is invoked on demand during question answering to incrementally expand the Q-KG only when complex relations are needed. Experiments show that Edge-AdaptiveKG reduces computational overhead and inference latency, enabling KG-enhanced RAG on resource-constrained devices while maintaining competitive QA accuracy.
Yuyu Du, Juxin Niu, Chun Jason Xue et al.· IEEE International Conferenc...· 0 citations
Knowledge graph question answering (KGQA) focuses on answering questions with data retrieved from knowledge graphs (KGs). Although Large Language Models (LLMs) have enabled the development of innovative KGQA systems, serious challenges remain. Approaches that utilize supervised finetuning are accurate and responsive, but also require large training datasets that provide good coverage of both the target knowledge graph and expected user questions. Likewise, integrating KG-specific optimizations improves accuracy and responsiveness but limits versatility and is susceptible to updates to the KG's ontology. LLMs have facilitated the development of training-free and KG-agnostic solutions that utilize reasoning and real-time decision making to explore the target KG, but the semi-structured nature of KGs makes this a complex and unpredictable task, leading to performance degradation and longer response times. In this paper we present PYTHIA, a training-free, plug-and-play solution for KGQA over any RDF KG. At its core, PYTHIA consists of an LLM agent guided by a relation-centric conceptual model of the KGQA task which is acted upon through a suite of tools for entity linking, graph exploration and query execution. The agent is responsible for interpreting user intent, orchestrating tool calls, and evaluating tool output, while the tools themselves present simple interfaces that abstract complex processes, reducing cognitive load on the agent. To assess its viability, we perform a comprehensive evaluation of PYTHIA and prior systems across four knowledge graphs and six datasets. The results show PYTHIA achieving a median performance improvement of 15.31% while improving cost and time efficiency at the performance frontier.
Sergios-Anestis Kefalidis, Konstantinos Plas, Manolis Koubarakis· Proceedings of the 32nd ACM...· 0 citations
An LLM reasoning framework with hierarchical relational retrieval for large-scale knowledge updating, named G-HiRel, which achieves superiority in terms of accuracy and interpretability and handles the knowledge inconsistency between the KG and LLM to obtain entity independence.
Yudai Pan, Jiajie Hong, Tianzhe Zhao et al.· Annual Meeting of the Associ...· 0 citations
This paper proposes DocTrace, a hierarchical framework that progressively performs evidence localization, structured document parsing, and evidence graph reasoning to enable explicit evidence provenance, and develops a two-stage training framework.
Lei Xiang, Zhicheng Guan, Hong Chen et al.· 0 citations
As Large Language Models (LLMs) become foundational to next-generation Intelligent Information Systems, the bridge between natural language interfaces and structured database systems remains a critical bottleneck. While Text-to-SQL generation enables cooperative support for complex query formulation, ensuring the reliability of these generated queries at inference time is a central challenge. Conventional methods rely on coarse execution-based signals, which may limit their ability to capture the nuanced semantic alignment required for high-stakes database environments. In this work, we propose the use of Outcome Reward Models (ORMs) as a fine-grained, probabilistic feedback mechanism for test-time verification in Text-to-SQL tasks. We introduce GradeSQL, a framework for training task-specific ORMs that assign scalar utility scores to candidate SQL queries based on their semantic correctness and alignment with database schema. Our approach is evaluated on the BIRD and Spider benchmarks across multiple open-source LLM families. Experimental results demonstrate that ORM-based verification consistently outperforms traditional execution-based heuristics.
M. Tritto, G. Farano, Dario Di Palma et al.· Journal of Intelligence and...· 2 citations