Think Locally, Explain Globally: Graph-Guided LLM Investigations via Local Reasoning and Belief Propagation

TL;DR

This paper introduces EoG, a graph-guided framework that enhances large language model investigations by enabling local reasoning and belief propagation, significantly improving accuracy and consistency in evidence-based diagnostics.

Contribution

The paper proposes a novel abductive reasoning framework over dependency graphs, separating local evidence mining from global belief management, to address limitations of existing ReAct-style agents.

Findings

EoG outperforms ReAct baselines in accuracy and consistency.

Achieves a 7x average gain in entity F1 on diagnostics tasks.

Improves reliability of evidence-based reasoning in open-ended investigations.

Abstract

LLM agents excel when environments are mostly static and the needed information fits in a model's context window, but they often fail in open-ended investigations where explanations must be constructed by iteratively mining evidence from massive, heterogeneous operational data. These investigations exhibit hidden dependency structure: entities interact, signals co-vary, and the importance of a fact may only become clear after other evidence is discovered. Because the context window is bounded, agents must summarize intermediate findings before their significance is known, increasing the risk of discarding key evidence. ReAct-style agents are especially brittle in this regime. Their retrieve-summarize-reason loop makes conclusions sensitive to exploration order and introduces run-to-run non-determinism, producing a reliability gap where Pass-at-k may be high but Majority-at-k remains low. Simply sampling more rollouts or generating longer reasoning traces does not reliably stabilize results, since hypotheses cannot be autonomously checked as new evidence arrives and there is no explicit mechanism for belief bookkeeping and revision. In addition, ReAct entangles semantic reasoning with controller duties such as tool orchestration and state tracking, so execution errors and plan drift degrade reasoning while consuming scarce context. We address these issues by formulating investigation as abductive reasoning over a dependency graph and proposing EoG (Explanations over Graphs), a disaggregated framework in which an LLM performs bounded local evidence mining and labeling (cause vs symptom) while a deterministic controller manages traversal, state, and belief propagation to compute a minimal explanatory frontier. On a representative ITBench diagnostics task, EoG improves both accuracy and run-to-run consistency over ReAct baselines, including a 7x average gain in Majority-at-k entity F1.