REVIEW 8 cited by
KG-Agent: An Efficient Autonomous Agent Framework for Complex Reasoning over Knowledge Graph
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
In this paper, we aim to improve the reasoning ability of large language models (LLMs) over knowledge graphs (KGs) to answer complex questions. Inspired by existing methods that design the interaction strategy between LLMs and KG, we propose an autonomous LLM-based agent framework, called KG-Agent, which enables a small LLM to actively make decisions until finishing the reasoning process over KGs. In KG-Agent, we integrate the LLM, multifunctional toolbox, KG-based executor, and knowledge memory, and develop an iteration mechanism that autonomously selects the tool then updates the memory for reasoning over KG. To guarantee the effectiveness, we leverage program language to formulate the multi-hop reasoning process over the KG, and synthesize a code-based instruction dataset to fine-tune the base LLM. Extensive experiments demonstrate that only using 10K samples for tuning LLaMA-7B can outperform state-of-the-art methods using larger LLMs or more data, on both in-domain and out-domain datasets. Our code and data will be publicly released.
Forward citations
Cited by 8 Pith papers
-
Leveraging Knowledge Graphs and LLM Reasoning to Identify Operational Bottlenecks for Warehouse Planning Assistance
An LLM agent querying a knowledge graph built from discrete-event simulation output identified warehouse bottlenecks and answered operational questions more reliably than single-pass query baselines.
-
BYOKG-RAG: Multi-Strategy Graph Retrieval for Knowledge Graph Question Answering
BYOKG-RAG combines LLM-generated entities, paths, queries, and candidate answers with multiple graph retrieval tools to answer questions over custom knowledge graphs without training data.
-
Learning Efficient and Generalizable Graph Retriever for Knowledge-Graph Question Answering
RAPL combines LLM-rationalized path labels, line graph transformation, and path-based decoding to improve graph retrieval for KGQA, reporting state-of-the-art results on WebQSP and CWQ.
-
Search-on-Graph: Iterative Informed Navigation for Large Language Model Reasoning on Knowledge Graphs
An LLM that iteratively inspects 1-hop neighbors of a knowledge-graph entity and chooses the next relation achieves state-of-the-art KGQA scores on six Freebase/Wikidata benchmarks without fine-tuning.
-
Text-to-SPARQL Goes Beyond English: Multilingual Question Answering Over Knowledge Graphs through Human-Inspired Reasoning
mKGQAgent, a modular LLM agent with planning, entity linking, feedback, and an experience pool, reports the best multilingual Text-to-SPARQL results on QALD-9-plus.
-
KnowMap: Efficient Knowledge-Driven Task Adaptation for LLMs
Fine-tuning a 0.56B-parameter embedding model to retrieve environment and experience knowledge improves gpt-4-turbo's ScienceWorld task score from 64.78 to 76.25.
-
Graphs Meet AI Agents: Taxonomy, Progress, and Future Opportunities
A survey that groups graph-empowered AI agent research into planning, execution, memory, and multi-agent coordination, plus agents-for-graphs and applications.
-
Autonomous Computer Vision Development with Agentic AI
An LLM-based agent autonomously generated a SimpleMind configuration, trained, and ran inference for chest X-ray lungs, heart, and ribs segmentation, achieving mean dice scores of 0.96, 0.82, and 0.83.
Discussion (0). Continue with ORCID to comment.