Large Language Models are Easily Confused: A Quantitative Metric, Security Implications and Typological Analysis

Language Confusion is a phenomenon where Large Language Models (LLMs) generate text that is neither in the desired language, nor in a contextually appropriate language. This phenomenon presents a critical challenge in text generation by LLMs, often appearing as erratic and unpredictable behavior. We hypothesize that there are linguistic regularities to this inherent vulnerability in LLMs and shed light on patterns of language confusion across LLMs. We introduce a novel metric, Language Confusion Entropy, designed to directly measure and quantify this confusion, based on language distributions informed by linguistic typology and lexical variation. Comprehensive comparisons with the Language Confusion Benchmark (Marchisio et al., 2024) confirm the effectiveness of our metric, revealing patterns of language confusion across LLMs. We further link language confusion to LLM security, and find patterns in the case of multilingual embedding inversion attacks. Our analysis demonstrates that linguistic typology offers theoretically grounded interpretation, and valuable insights into leveraging language similarities as a prior for LLM alignment and security.

Against All Odds: Overcoming Typology, Script, and Language Confusion in Multilingual Embedding Inversion Attacks

Large Language Models (LLMs) are susceptible to malicious influence by cyber attackers through intrusions such as adversarial, backdoor, and embedding inversion attacks. In response, the burgeoning field of LLM Security aims to study and defend against such threats. Thus far, the majority of works in this area have focused on monolingual English models, however, emerging research suggests that multilingual LLMs may be more vulnerable to various attacks than their monolingual counterparts. While previous work has investigated embedding inversion over a small subset of European languages, it is challenging to extrapolate these findings to languages from different linguistic families and with differing scripts. To this end, we explore the security of multilingual LLMs in the context of embedding inversion attacks and investigate cross-lingual and cross-script inversion across 20 languages, spanning over 8 language families and 12 scripts. Our findings indicate that languages written in Arabic script and Cyrillic script are particularly vulnerable to embedding inversion, as are languages within the Indo-Aryan language family. We further observe that inversion models tend to suffer from language confusion, sometimes greatly reducing the efficacy of an attack. Accordingly, we systematically explore this bottleneck for inversion models, uncovering predictable patterns which could be leveraged by attackers. Ultimately, this study aims to further the field's understanding of the outstanding security vulnerabilities facing multilingual LLMs and raise awareness for the languages most at risk of negative impact from these attacks.

Large Language Models and Knowledge Graphs: Opportunities and Challenges

Large Language Models (LLMs) have taken Knowledge Representation -- and the world -- by storm. This inflection point marks a shift from explicit knowledge representation to a renewed focus on the hybrid representation of both explicit knowledge and parametric knowledge. In this position paper, we will discuss some of the common debate points within the community on LLMs (parametric knowledge) and Knowledge Graphs (explicit knowledge) and speculate on opportunities and visions that the renewed focus brings, as well as related research topics and challenges.

Entity Type Prediction Leveraging Graph Walks and Entity Descriptions

The entity type information in Knowledge Graphs (KGs) such as DBpedia, Freebase, etc. is often incomplete due to automated generation or human curation. Entity typing is the task of assigning or inferring the semantic type of an entity in a KG. This paper presents \textit{GRAND}, a novel approach for entity typing leveraging different graph walk strategies in RDF2vec together with textual entity descriptions. RDF2vec first generates graph walks and then uses a language model to obtain embeddings for each node in the graph. This study shows that the walk generation strategy and the embedding model have a significant effect on the performance of the entity typing task. The proposed approach outperforms the baseline approaches on the benchmark datasets DBpedia and FIGER for entity typing in KGs for both fine-grained and coarse-grained classes. The results show that the combination of order-aware RDF2vec variants together with the contextual embeddings of the textual entity descriptions achieve the best results.

Knowledge Graphs Evolution and Preservation -- A Technical Report from ISWS 2019

One of the grand challenges discussed during the Dagstuhl Seminar "Knowledge Graphs: New Directions for Knowledge Representation on the Semantic Web" and described in its report is that of a: "Public FAIR Knowledge Graph of Everything: We increasingly see the creation of knowledge graphs that capture information about the entirety of a class of entities. [...] This grand challenge extends this further by asking if we can create a knowledge graph of "everything" ranging from common sense concepts to location based entities. This knowledge graph should be "open to the public" in a FAIR manner democratizing this mass amount of knowledge." Although linked open data (LOD) is one knowledge graph, it is the closest realisation (and probably the only one) to a public FAIR Knowledge Graph (KG) of everything. Surely, LOD provides a unique testbed for experimenting and evaluating research hypotheses on open and FAIR KG. One of the most neglected FAIR issues about KGs is their ongoing evolution and long term preservation. We want to investigate this problem, that is to understand what preserving and supporting the evolution of KGs means and how these problems can be addressed. Clearly, the problem can be approached from different perspectives and may require the development of different approaches, including new theories, ontologies, metrics, strategies, procedures, etc. This document reports a collaborative effort performed by 9 teams of students, each guided by a senior researcher as their mentor, attending the International Semantic Web Research School (ISWS 2019). Each team provides a different perspective to the problem of knowledge graph evolution substantiated by a set of research questions as the main subject of their investigation. In addition, they provide their working definition for KG preservation and evolution.

Entity Type Prediction in Knowledge Graphs using Embeddings

Open Knowledge Graphs (such as DBpedia, Wikidata, YAGO) have been recognized as the backbone of diverse applications in the field of data mining and information retrieval. Hence, the completeness and correctness of the Knowledge Graphs (KGs) are vital. Most of these KGs are mostly created either via an automated information extraction from Wikipedia snapshots or information accumulation provided by the users or using heuristics. However, it has been observed that the type information of these KGs is often noisy, incomplete, and incorrect. To deal with this problem a multi-label classification approach is proposed in this work for entity typing using KG embeddings. We compare our approach with the current state-of-the-art type prediction method and report on experiments with the KGs.

Is Aligning Embedding Spaces a Challenging Task? An Analysis of the Existing Methods

Representation Learning of words and Knowledge Graphs (KG) into low dimensional vector spaces along with its applications to many real-world scenarios have recently gained momentum. In order to make use of multiple KG embeddings for knowledge-driven applications such as question answering, named entity disambiguation, knowledge graph completion, etc., alignment of different KG embedding spaces is necessary. In addition to multilinguality and domain-specific information, different KGs pose the problem of structural differences making the alignment of the KG embeddings more challenging. This paper provides a theoretical analysis and comparison of the state-of-the-art alignment methods between two embedding spaces representing entity-entity and entity-word. This paper also aims at assessing the capability and short-comings of the existing alignment methods on the pretext of different applications.

A Survey on Knowledge Graph Embeddings with Literals: Which model links better Literal-ly?

Knowledge Graphs (KGs) are composed of structured information about a particular domain in the form of entities and relations. In addition to the structured information KGs help in facilitating interconnectivity and interoperability between different resources represented in the Linked Data Cloud. KGs have been used in a variety of applications such as entity linking, question answering, recommender systems, etc. However, KG applications suffer from high computational and storage costs. Hence, there arises the necessity for a representation able to map the high dimensional KGs into low dimensional spaces, i.e., embedding space, preserving structural as well as relational information. This paper conducts a survey of KG embedding models which not only consider the structured information contained in the form of entities and relations in a KG but also the unstructured information represented as literals such as text, numerical values, images, etc. Along with a theoretical analysis and comparison of the methods proposed so far for generating KG embeddings with literals, an empirical evaluation of the different methods under identical settings has been performed for the general task of link prediction.