paper-with-me

홈 › Papers

Bounded Path Context: A Controlled Study of Visible Path History in LLM-Based Knowledge Graph Question Answering

2026-05-26 · Xihang Shan, Ye Luo arxiv

LLM-based knowledge-graph question answering (KGQA) delegates graph traversal to language models, turning each question into a sequence of local relation-selection decisions repeated across beams and hops. A common but untested default is to serialize the complete partial path into every routing prompt, even though the controller already maintains this path as exact symbolic state. Bounded Path Context (BPC) decouples these two roles: the controller retains full paths in symbolic memory for answer extraction and audit, while the relation-selection prompt exposes only the question, the current entity, outgoing relation candidates, and at most the last K hops. A controlled sweep over K -- fixing graph neighborhoods, beam budget, depth, decoding, and answer-extraction format -- shows that bounded histories match or exceed full-history prompting on complete WebQSP and CWQ test sets with Qwen3.5-9B-AWQ: K=1 achieves 0.487 answer-set F1 on WebQSP versus 0.472 for full history, and K=0 reaches 0.287 on CWQ versus 0.274, with 9.7% and 12.1% fewer input tokens respectively. At the 4B scale, K=1 remains the strongest setting on both benchmarks. Per-example analysis reveals that 71-84% of examples are unaffected by history length, while the affected cases expose when prior hops disambiguate versus distract. These results suggest that path serialization length is better treated as a tunable interface variable than as a default assumption in LLM-based graph controllers.

📄 PDF Abstract BibTeX arXiv:2605.26645

Code (0)

등록된 구현이 없습니다.

Tasks

Graph Question Answering

Similar Papers 제목 키워드 기반

Attention is Just Another Name for Coupling? A Fast-Slow ODE Perspective on Hierarchical Pretraining

2026-06-15 · Zhengyuan Gao arxiv

We re-interpret Transformer pretraining as a fast-slow, singularly perturbed flow along depth, with untied weights as its non-autonomous feature. The linearised dynamics is a depth-ordered product of layer maps. Along a …

When Does Gene Regulatory Network Inference Break? A Controlled Diagnostic Study of Causal and Correlational Methods on Single-Cell Data

2026-05-06 · Miguel Fernandez-de-Retana, Ruben Sanchez-Corcuera, Unai Zulaika, Aritz Bilbao-Jayo 외 arxiv

Despite theoretical advantages, causal methods for Gene Regulatory Network (GRN) inference from single-cell RNA-seq data consistently fail to match or outperform correlation-based baselines in many realistic benchmarks, …

ToolMenuBench: Benchmarking Tool-Menu Filtering Strategies for Reliable and Efficient LLM Agents

2026-06-13 · Rahul Suresh Babu, Laxmipriya Ganesh Iyer arxiv

Tool-augmented large language model agents increasingly operate over large tool libraries, but existing evaluations often focus on whether a model can call a tool correctly rather than how the visible tool menu shapes re…

TRACE: Trajectory-Routed Causal Memory for Delayed-Evidence Visuomotor Imitation

2026-06-12 · Zihao Li, Ranpeng Qiu, Yincong Chen, Guoqiang Ren 외 arxiv

Robots under autonomous operation may require decisions based on evidence that is no longer visible. We study delayed-evidence tasks, where an early cue disappears before a later decision point, so visually similar obser…

EchoPath: Execution-Level Replayable Memory for GUI Agents

2026-09-15 · Yao Zhao, Aditya Shanmugham, Swastik Roy, Yanxun Xu arxiv

Computer-use agents increasingly operate browsers, software, and desktop applications via CLI or API portals, but graphical user interface (GUI) still plays an important role in common industrial production scenarios. GU…