GRA is presented, a Graph Reasoning Agent that explores hybrid knowledge graphs, whose nodes are either textual concepts or relational tables, with seven generic tools, discovering everything domain-specific at run time.
Abstract
Tool-calling LLM agents navigate unfamiliar codebases with a handful of generic primitives for listing, reading and searching files (ls, cat, grep). A knowledge graph admits the same interface: listing neighbours, reading node content and searching descriptions are the same operations on a different substrate. Building on this correspondence, we present GRA, a Graph Reasoning Agent that explores hybrid knowledge graphs, whose nodes are either textual concepts or relational tables, with seven generic tools, discovering everything domain-specific at run time. On UFK-M (Unified Factory Knowledge Model), an industrial benchmark of 258 analytical questions whose gold answers are produced by executing validated SQL programs, GRA beats a full-context agent by 5.1 pp (88.4% vs. 83.3%), while reading under a third of its input tokens. A graph-free control shows the gain comes chiefly from selective agentic access rather than graph topology, and that the effect depends on a model able to drive tools reliably. Seeing less, the agent answers better: selective navigation over a structured substrate beats exhaustive context.
Translating a natural-language question into a SPARQL query that can be executed against a large knowledge graph requires resolving lexical ambiguity, grounding surface terms in the target ontology, and producing graph patterns that are both syntactically valid and semantically faithful. We present an agentic text-to-SPARQL system that goes one step beyond static tool-using agents: a researcher agent that, after each round of inference on a validation set, proposes and tests changes to its own prompts, rules, and tool-orchestration code. We instantiate the loop on DBpedia, evolve nine successive versions of the agent driven by a low-cost reasoning model, and deploy the best-performing configuration with two stronger backbone models. The study yields three observations: (i) self-improvement converges quickly and then achieves 0.22 overall accuracy on the 2025 DBpedia validation set; (ii) the bottleneck is consistently in basic-graph-pattern predicate selection, not in SPARQL syntax or modifiers; and (iii) several benchmark items appear to penalise correct queries due to property ambiguity in DBpedia, suggesting that future Text-to-SPARQL benchmarks should be scored using a combination of machine translation and information retrieval metrics.
Enterprise IT support knowledge graphs capture rich relationships among cases, users, devices, symptoms, taxonomic categories, root causes, and historical resolutions. Yet querying them in Gremlin requires knowledge of graph schemas, traversal semantics, edge directionality, and property-graph-specific constraints, making them difficult for non-expert operators to use. We introduce SEGRA, an experience-guided agent for enterprise text-to-Gremlin question answering. SEGRA integrates intent routing, schema- and taxonomy-grounded query generation, multi-shot decomposition, execution-aware verification, and a curriculum-bootstrapped skill library that reuses verified query patterns. On an enterprise IT support benchmark, SEGRA achieves a $7.0\times$ higher mean judge score than backbone-only chain-of-thought prompting. Its skill library further reduces LLM calls by $20\%$ and dollar cost by $18\%$ relative to SEGRA without skills, while preserving answer quality. These results show that schema-grounded agent design and reusable execution experience improve both accuracy and efficiency for enterprise graph QA.
Saiyue Lyu, Mariam Dundua, Vishaal Kapoor et al.· 0 citations
This work introduces agentic graph token reasoning, which recasts graph tokenization as part of the reasoning process itself and pushes LLM-based graph analysis from static graph-token encoders towards a graph-native agent paradigm.
This work presents an evidence-carrying validation interface: every selected node-shape check returns either a satisfaction trace or failure witness, and shows how programs combine passing and failing evidence to diagnose missing information and guide repair.
Organisations hold significant knowledge capital that their AI systems cannot access as knowledge. The knowledge bases most organisations maintain — manuals, onboarding documents, process documentation, and accumulated expertise — are structurally incompatible with AI reasoning: there are no dependency structures to traverse, no role classifications to honour, no relational context to surface. This paper introduces meta-books as a knowledge management design response to this structural problem. A meta-book transforms a conventional organisational document into a typed, dependency-linked knowledge graph. Source content is decomposed into typed knowledge objects — concepts, procedures, examples, controversies, and reflective insights — connected by semantically meaningful edges such as prerequisite-of, illustrates, contradicts, and applies-to. Every node carries pedagogical metadata: depth layer, role classification, engagement time, and prerequisite links. Unlike retrieval-augmented generation, meta-books restructure source content itself, making dependency, context, and role visible to AI at the point of reasoning, and support cross-source integration allowing knowledge nodes to migrate across individual, team, and organisational graphs without losing relational context. Meta-books reposition organisational knowledge as a compound, AI-traversable asset with implications across the full employee lifecycle — from recruitment and adaptive onboarding to knowledge retention beyond individual tenure. The paper positions meta-books against learning object metadata, adaptive hypermedia, semantic web standards, and enterprise knowledge graphs, demonstrating that no existing standard simultaneously occupies the intersection of typed knowledge structure, pedagogical metadata, and AI queryability. Five design principles are derived using design science research methodology and evaluated through illustrative decomposition of a representative organisational document. The paper’s most distinctive argument is that meta-books are the first organisational KM framework explicitly aligned with the cognitive architecture of human learning — integrating schema theory, cognitive load theory, dual coding, retrieval practice, the spacing effect, meaningful learning, and connectionist neuroscience — with implications for universities, professional certification, and all structured knowledge delivery systems.
Sima Fatemipour· European Conference on Knowl...· 0 citations