Extending a Parliamentary Corpus with MPs' Tweets: Automatic Annotation and Evaluation Using MultiParTweet

Social media serves as a critical medium in modern politics because it both reflects politicians' ideologies and facilitates communication with younger generations. We present MultiParTweet, a multilingual tweet corpus from X that connects politicians' social media discourse with German political corpus GerParCor, thereby enabling comparative analyses between online communication and parliamentary debates. MultiParTweet contains 39 546 tweets, including 19 056 media items. Furthermore, we enriched the annotation with nine text-based models and one vision-language model (VLM) to annotate MultiParTweet with emotion, sentiment, and topic annotations. Moreover, the automated annotations are evaluated against a manually annotated subset. MultiParTweet can be reconstructed using our tool, TTLABTweetCrawler, which provides a framework for collecting data from X. To demonstrate a methodological demonstration, we examine whether the models can predict each other using the outputs of the remaining models. In summary, we provide MultiParTweet, a resource integrating automatic text and media-based annotations validated with human annotations, and TTLABTweetCrawler, a general-purpose X data collection tool. Our analysis shows that the models are mutually predictable. In addition, VLM-based annotation were preferred by human annotators, suggesting that multimodal representations align more with human interpretation.

FAME: Fictional Actors for Multilingual Erasure

LLMs trained on web-scale data raise concerns about privacy and the right to be forgotten. To address these issues, Machine Unlearning provides techniques to remove specific information from trained models without retraining from scratch. However, existing benchmarks for evaluating unlearning in LLMs face two major limitations: they focus only on English and support only entity-level forgetting (removing all information about a person). We introduce FAME (Fictional Actors for Multilingual Erasure), a synthetic benchmark for evaluating Machine Unlearning across five languages: English, French, German, Italian, and Spanish. FAME contains 1,000 fictional actor biographies and 20,000 question-answer pairs. Each biography includes information on 20 topics organized into structured categories (biography, career, achievements, personal information). This design enables both entity-level unlearning (i.e., forgetting entire identities) and instance-level unlearning (i.e., forgetting specific facts while retaining others). We provide two dataset splits to support these two different unlearning scenarios and enable systematic comparison of unlearning techniques across languages. Since FAME uses entirely fictional data, it ensures that the information was never encountered during model pretraining, allowing for a controlled evaluation of unlearning methods.

FIBER: A Multilingual Evaluation Resource for Factual Inference Bias

Large language models are widely used across domains, yet there are concerns about their factual reliability and biases. Factual knowledge probing offers a systematic means to evaluate these aspects. Most existing benchmarks focus on single-entity facts and monolingual data. We therefore present FIBER, a multilingual benchmark for evaluating factual knowledge in single- and multi-entity settings. The dataset includes sentence completion, question-answering, and object-count prediction tasks in English, Italian, and Turkish. Using FIBER, we examine whether the prompt language induces inference bias in entity selection and how large language models perform on multi-entity versus single-entity questions. The results indicate that the language of the prompt can influence the model's generated output, particularly for entities associated with the country corresponding to that language. However, this effect varies across different topics such that 31% of the topics exhibit factual inference bias score greater than 0.5. Moreover, the level of bias differs across languages such that Turkish prompts show higher bias compared to Italian in 83% of the topics, suggesting a language-dependent pattern. Our findings also show that models face greater difficulty when handling multi-entity questions than the single-entity questions. Model performance differs across both languages and model sizes. The highest mean average precision is achieved in English, while Turkish and Italian lead to noticeably lower scores. Larger models, including Llama-3.1-8B and Qwen-2.5-7B, show consistently better performance than smaller 3B-4B models.

Grad: Guided Relation Diffusion Generation for Graph Augmentation in Graph Fraud Detection

Nowadays, Graph Fraud Detection (GFD) in financial scenarios has become an urgent research topic to protect online payment security. However, as organized crime groups are becoming more professional in real-world scenarios, fraudsters are employing more sophisticated camouflage strategies. Specifically, fraudsters disguise themselves by mimicking the behavioral data collected by platforms, ensuring that their key characteristics are consistent with those of benign users to a high degree, which we call Adaptive Camouflage. Consequently, this narrows the differences in behavioral traits between them and benign users within the platform's database, thereby making current GFD models lose efficiency. To address this problem, we propose a relation diffusion-based graph augmentation model Grad. In detail, Grad leverages a supervised graph contrastive learning module to enhance the fraud-benign difference and employs a guided relation diffusion generator to generate auxiliary homophilic relations from scratch. Based on these, weak fraudulent signals would be enhanced during the aggregation process, thus being obvious enough to be captured. Extensive experiments have been conducted on two real-world datasets provided by WeChat Pay, one of the largest online payment platforms with billions of users, and three public datasets. The results show that our proposed model Grad outperforms SOTA methods in both various scenarios, achieving at most 11.10% and 43.95% increases in AUC and AP, respectively. Our code is released at https://github.com/AI4Risk/antifraud and https://github.com/Muyiiiii/WWW25-Grad.

Exploring LLMs for Scientific Information Extraction Using The SciEx Framework

Large language models (LLMs) are increasingly touted as powerful tools for automating scientific information extraction. However, existing methods and tools often struggle with the realities of scientific literature: long-context documents, multi-modal content, and reconciling varied and inconsistent fine-grained information across multiple publications into standardized formats. These challenges are further compounded when the desired data schema or extraction ontology changes rapidly, making it difficult to re-architect or fine-tune existing systems. We present SciEx, a modular and composable framework that decouples key components including PDF parsing, multi-modal retrieval, extraction, and aggregation. This design streamlines on-demand data extraction while enabling extensibility and flexible integration of new models, prompting strategies, and reasoning mechanisms. We evaluate SciEx on datasets spanning three scientific topics for its ability to extract fine-grained information accurately and consistently. Our findings provide practical insights into both the strengths and limitations of current LLM-based pipelines.

Quantifying consistency and accuracy of Latent Dirichlet Allocation

Topic modelling in Natural Language Processing uncovers hidden topics in large, unlabelled text datasets. It is widely applied in fields such as information retrieval, content summarisation, and trend analysis across various disciplines. However, probabilistic topic models can produce different results when rerun due to their stochastic nature, leading to inconsistencies in latent topics. Factors like corpus shuffling, rare text removal, and document elimination contribute to these variations. This instability affects replicability, reliability, and interpretation, raising concerns about whether topic models capture meaningful topics or just noise. To address these problems, we defined a new stability measure that incorporates accuracy and consistency and uses the generative properties of LDA to generate a new corpus with ground truth. These generated corpora are run through LDA 50 times to determine the variability in the output. We show that LDA can correctly determine the underlying number of topics in the documents. We also find that LDA is more internally consistent, as the multiple reruns return similar topics; however, these topics are not the true topics.

Large Causal Models from Large Language Models

We introduce a new paradigm for building large causal models (LCMs) that exploits the enormous potential latent in today's large language models (LLMs). We describe our ongoing experiments with an implemented system called DEMOCRITUS (Decentralized Extraction of Manifold Ontologies of Causal Relations Integrating Topos Universal Slices) aimed at building, organizing, and visualizing LCMs that span disparate domains extracted from carefully targeted textual queries to LLMs. DEMOCRITUS is methodologically distinct from traditional narrow domain and hypothesis centered causal inference that builds causal models from experiments that produce numerical data. A high-quality LLM is used to propose topics, generate causal questions, and extract plausible causal statements from a diverse range of domains. The technical challenge is then to take these isolated, fragmented, potentially ambiguous and possibly conflicting causal claims, and weave them into a coherent whole, converting them into relational causal triples and embedding them into a LCM. Addressing this technical challenge required inventing new categorical machine learning methods, which we can only briefly summarize in this paper, as it is focused more on the systems side of building DEMOCRITUS. We describe the implementation pipeline for DEMOCRITUS comprising of six modules, examine its computational cost profile to determine where the current bottlenecks in scaling the system to larger models. We describe the results of using DEMOCRITUS over a wide range of domains, spanning archaeology, biology, climate change, economics, medicine and technology. We discuss the limitations of the current DEMOCRITUS system, and outline directions for extending its capabilities.

CLINIC: Evaluating Multilingual Trustworthiness in Language Models for Healthcare

Integrating language models (LMs) in healthcare systems holds great promise for improving medical workflows and decision-making. However, a critical barrier to their real-world adoption is the lack of reliable evaluation of their trustworthiness, especially in multilingual healthcare settings. Existing LMs are predominantly trained in high-resource languages, making them ill-equipped to handle the complexity and diversity of healthcare queries in mid- and low-resource languages, posing significant challenges for deploying them in global healthcare contexts where linguistic diversity is key. In this work, we present CLINIC, a Comprehensive Multilingual Benchmark to evaluate the trustworthiness of language models in healthcare. CLINIC systematically benchmarks LMs across five key dimensions of trustworthiness: truthfulness, fairness, safety, robustness, and privacy, operationalized through 18 diverse tasks, spanning 15 languages (covering all the major continents), and encompassing a wide array of critical healthcare topics like disease conditions, preventive actions, diagnostic tests, treatments, surgeries, and medications. Our extensive evaluation reveals that LMs struggle with factual correctness, demonstrate bias across demographic and linguistic groups, and are susceptible to privacy breaches and adversarial attacks. By highlighting these shortcomings, CLINIC lays the foundation for enhancing the global reach and safety of LMs in healthcare across diverse languages.

Darth Vecdor: An Open-Source System for Generating Knowledge Graphs Through Large Language Model Queries

Many large language models (LLMs) are trained on a massive body of knowledge present on the Internet. Darth Vecdor (DV) was designed to extract this knowledge into a structured, terminology-mapped, SQL database ("knowledge base" or "knowledge graph"). Knowledge graphs may be useful in many domains, including healthcare. Although one might query an LLM directly rather than a SQL-based knowledge graph, concerns such as cost, speed, safety, and confidence may arise, especially in high-volume operations. These may be mitigated when the information is pre-extracted from the LLM and becomes query-able through a standard database. However, the author found the need to address several issues. These included erroneous, off-topic, free-text, overly general, and inconsistent LLM responses, as well as allowing for multi-element responses. DV was built with features intended to mitigate these issues. To facilitate ease of use, and to allow for prompt engineering by those with domain expertise but little technical background, DV provides a simple, browser-based graphical user interface. DV has been released as free, open-source, extensible software, on an "as is" basis, without warranties or conditions of any kind, either express or implied. Users need to be cognizant of the potential risks and benefits of using DV and its outputs, and users are responsible for ensuring any use is safe and effective. DV should be assumed to have bugs, potentially very serious ones. However, the author hopes that appropriate use of current and future versions of DV and its outputs can help improve healthcare.

Challenges of Evaluating LLM Safety for User Welfare

Safety evaluations of large language models (LLMs) typically focus on universal risks like dangerous capabilities or undesirable propensities. However, millions use LLMs for personal advice on high-stakes topics like finance and health, where harms are context-dependent rather than universal. While frameworks like the OECD's AI classification recognize the need to assess individual risks, user-welfare safety evaluations remain underdeveloped. We argue that developing such evaluations is non-trivial due to fundamental questions about accounting for user context in evaluation design. In this exploratory study, we evaluated advice on finance and health from GPT-5, Claude Sonnet 4, and Gemini 2.5 Pro across user profiles of varying vulnerability. First, we demonstrate that evaluators must have access to rich user context: identical LLM responses were rated significantly safer by context-blind evaluators than by those aware of user circumstances, with safety scores for high-vulnerability users dropping from safe (5/7) to somewhat unsafe (3/7). One might assume this gap could be addressed by creating realistic user prompts containing key contextual information. However, our second study challenges this: we rerun the evaluation on prompts containing context users report they would disclose, finding no significant improvement. Our work establishes that effective user-welfare safety evaluation requires evaluators to assess responses against diverse user profiles, as realistic user context disclosure alone proves insufficient, particularly for vulnerable populations. By demonstrating a methodology for context-aware evaluation, this study provides both a starting point for such assessments and foundational evidence that evaluating individual welfare demands approaches distinct from existing universal-risk frameworks. We publish our code and dataset to aid future developments.