KGpipe: Generation and Evaluation of Pipelines for Data Integration into Knowledge Graphs

The emergence of Large Language Model (LLM)-driven multi-agent systems has transformed software development, allowing users with minimal technical skills to create applications through natural language inputs. However, this innovation also raises significant security concerns, particularly through scenarios where malicious users exploit benign agents or vice versa. The introduction of the Implicit Malicious Behavior Injection Attack (IMBIA) highlights these vulnerabilities, with alarming success rates in various frameworks.

FHE-Agent has been introduced as an innovative framework designed to automate the configuration of Fully Homomorphic Encryption (FHE) using the CKKS scheme, addressing the complexities that typically hinder its practical deployment in privacy-preserving machine learning as a service (MLaaS). The framework integrates a Large Language Model (LLM) to streamline the configuration process, making it more accessible to practitioners without deep cryptographic expertise.

Small language models (SLMs) have demonstrated potential in relation extraction (RE) for extracting RDF triples guided by SHACL shapes, particularly focusing on common datatype properties. A recent study identifies the challenge of long-tail distribution of rare properties as a key bottleneck in handling both datatype and object properties for comprehensive RDF graph extraction, proposing several strategies to address this issue.

A new evaluation metric called SMILE has been introduced to enhance the assessment of question-answering systems by integrating both lexical exactness and semantic understanding. This metric aims to address the limitations of traditional methods that rely heavily on n-gram similarity, which often overlook deeper semantic meanings. SMILE combines sentence-level and keyword-level evaluations to provide a more comprehensive assessment of responses.