Fake news poses a significant threat to the integrity of information ecosystems and public trust. The advent of Large Language Models (LLMs) holds considerable promise for transforming the battle against fake news. Generally, LLMs represent a double-edged sword in this struggle. One major concern is that LLMs can be readily used to craft and disseminate misleading information on a large scale. This raises the pressing questions: Can LLMs easily generate biased fake news? Do all LLMs have this capability? Conversely, LLMs offer valuable prospects for countering fake news, thanks to their extensive knowledge of the world and robust reasoning capabilities. This leads to other critical inquiries: Can we use LLMs to detect fake news, and do they outperform typical detection models? In this paper, we aim to address these pivotal questions by exploring the performance of various LLMs. Our objective is to explore the capability of various LLMs in effectively combating fake news, marking this as the first investigation to analyze seven such models. Our results reveal that while some models adhere strictly to safety protocols, refusing to generate biased or misleading content, other models can readily produce fake news across a spectrum of biases. Additionally, our results show that larger models generally exhibit superior detection abilities and that LLM-generated fake news are less likely to be detected than human-written ones. Finally, our findings demonstrate that users can benefit from LLM-generated explanations in identifying fake news.

Generative AI holds immense promise in addressing global healthcare access challenges, with numerous innovative applications now ready for use across various healthcare domains. However, a significant barrier to the widespread adoption of these domain-specific AI solutions is the lack of robust safety mechanisms to effectively manage issues such as hallucination, misinformation, and ensuring truthfulness. Left unchecked, these risks can compromise patient safety and erode trust in healthcare AI systems. While general-purpose frameworks like Llama Guard are useful for filtering toxicity and harmful content, they do not fully address the stringent requirements for truthfulness and safety in healthcare contexts. This paper examines the unique safety and security challenges inherent to healthcare AI, particularly the risk of hallucinations, the spread of misinformation, and the need for factual accuracy in clinical settings. I propose enhancements to existing guardrails frameworks, such as Nvidia NeMo Guardrails, to better suit healthcare-specific needs. By strengthening these safeguards, I aim to ensure the secure, reliable, and accurate use of AI in healthcare, mitigating misinformation risks and improving patient safety.

The subjective quality of natural signals can be approximated with objective perceptual metrics. Designed to approximate the perceptual behaviour of human observers, perceptual metrics often reflect structures found in natural signals and neurological pathways. Models trained with perceptual metrics as loss functions can capture perceptually meaningful features from the structures held within these metrics. We demonstrate that using features extracted from autoencoders trained with perceptual losses can improve performance on music understanding tasks, i.e. genre classification, over using these metrics directly as distances when learning a classifier. This result suggests improved generalisation to novel signals when using perceptual metrics as loss functions for representation learning.

This paper examines the development of the Artificial Intelligence (AI) meta-debate in Sweden before and after the release of ChatGPT. From the perspective of agenda-setting theory, we propose that it is an elite outside of party politics that is leading the debate -- i.e. that the politicians are relatively silent when it comes to this rapid development. We also suggest that the debate has become more substantive and risk-oriented in recent years. To investigate this claim, we draw on an original dataset of elite-level documents from the early 2010s to the present, using op-eds published in a number of leading Swedish newspapers. By conducting a qualitative content analysis of these materials, our preliminary findings lend support to the expectation that an academic, rather than a political elite is steering the debate.

The climate crisis is a salient issue in online discussions, and hypocrisy accusations are a central rhetorical element in these debates. However, for large-scale text analysis, hypocrisy accusation detection is an understudied tool, most often defined as a smaller subtask of fallacious argument detection. In this paper, we define hypocrisy accusation detection as an independent task in NLP, and identify different relevant subtypes of hypocrisy accusations. Our Climate Hypocrisy Accusation Corpus (CHAC) consists of 420 Reddit climate debate comments, expert-annotated into two different types of hypocrisy accusations: personal versus political hypocrisy. We evaluate few-shot in-context learning with 6 shots and 3 instruction-tuned Large Language Models (LLMs) for detecting hypocrisy accusations in this dataset. Results indicate that the GPT-4o and Llama-3 models in particular show promise in detecting hypocrisy accusations (F1 reaching 0.68, while previous work shows F1 of 0.44). However, context matters for a complex semantic concept such as hypocrisy accusations, and we find models struggle especially at identifying political hypocrisy accusations compared to personal moral hypocrisy. Our study contributes new insights in hypocrisy detection and climate change discourse, and is a stepping stone for large-scale analysis of hypocrisy accusation in online climate debates.

Role-playing systems powered by large language models (LLMs) have become increasingly influential in emotional communication applications. However, these systems are susceptible to character hallucinations, where the model deviates from predefined character roles and generates responses that are inconsistent with the intended persona. This paper presents the first systematic analysis of character hallucination from an attack perspective, introducing the RoleBreak framework. Our framework identifies two core mechanisms-query sparsity and role-query conflict-as key factors driving character hallucination. Leveraging these insights, we construct a novel dataset, RoleBreakEval, to evaluate existing hallucination mitigation techniques. Our experiments reveal that even enhanced models trained to minimize hallucination remain vulnerable to attacks. To address these vulnerabilities, we propose a novel defence strategy, the Narrator Mode, which generates supplemental context through narration to mitigate role-query conflicts and improve query generalization. Experimental results demonstrate that Narrator Mode significantly outperforms traditional refusal-based strategies by reducing hallucinations, enhancing fidelity to character roles and queries, and improving overall narrative coherence.

This paper proposes a novel prompt engineering technique called Judgment of Thought (JoT) that is specifically tailored for binary logical reasoning tasks. JoT employs three roles$\unicode{x2014}$lawyer, prosecutor, and judge$\unicode{x2014}$to facilitate more reliable and accurate reasoning by the model. In this framework, the judge utilizes a high$\unicode{x2010}$level model, while the lawyer and prosecutor utilize low$\unicode{x2010}$level models. This structure helps the judge better understand the responses from both the lawyer and prosecutor, enabling a more accurate judgment. Experimental results on large language model (LLM) benchmark datasets, such as BigBenchHard and Winogrande, demonstrate that JoT outperforms existing methods, including Chain of Thought (CoT) and Self$\unicode{x2010}$Consistency (SC), in binary logical reasoning tasks. Additionally, in real$\unicode{x2010}$world tasks, such as Fake News Detection and SMS Spam Detection, JoT shows comparable or improved performance compared to existing techniques. JoT significantly enhances the accuracy and reliability of models in binary reasoning tasks and show potential for practical applicability across various domains. Future research should aim to further broaden the applicability of JoT and optimize its implementation for real$\unicode{x2010}$world problem$\unicode{x2010}$solving.

The rapid spread of diverse information on online social platforms has prompted both academia and industry to realize the importance of predicting content popularity, which could benefit a wide range of applications, such as recommendation systems and strategic decision-making. Recent works mainly focused on extracting spatiotemporal patterns inherent in the information diffusion process within a given observation period so as to predict its popularity over a future period of time. However, these works often overlook the future popularity trend, as future popularity could either increase exponentially or stagnate, introducing uncertainties to the prediction performance. Additionally, how to transfer the preceding-term dynamics learned from the observed diffusion process into future-term trends remains an unexplored challenge. Against this background, we propose CasFT, which leverages observed information Cascades and dynamic cues extracted via neural ODEs as conditions to guide the generation of Future popularity-increasing Trends through a diffusion model. These generated trends are then combined with the spatiotemporal patterns in the observed information cascade to make the final popularity prediction. Extensive experiments conducted on three real-world datasets demonstrate that CasFT significantly improves the prediction accuracy, compared to state-of-the-art approaches, yielding 2.2%-19.3% improvement across different datasets.

LLM-based applications are helping people write, and LLM-generated text is making its way into social media, journalism, and our classrooms. However, the differences between LLM-generated and human-written text remain unclear. To explore this, we hired professional writers to edit paragraphs in several creative domains. We first found these writers agree on undesirable idiosyncrasies in LLM-generated text, formalizing it into a seven-category taxonomy (e.g. cliches, unnecessary exposition). Second, we curated the LAMP corpus: 1,057 LLM-generated paragraphs edited by professional writers according to our taxonomy. Analysis of LAMP reveals that none of the LLMs used in our study (GPT4o, Claude-3.5-Sonnet, Llama-3.1-70b) outperform each other in terms of writing quality, revealing common limitations across model families. Third, we explored automatic editing methods to improve LLM-generated text. A large-scale preference annotation confirms that although experts largely prefer text edited by other experts, automatic editing methods show promise in improving alignment between LLM-generated and human-written text.

This paper advances the understanding of how the size of a machine learning model affects its vulnerability to poisoning, despite state-of-the-art defenses. Given isotropic random honest feature vectors and the geometric median (or clipped mean) as the robust gradient aggregator rule, we essentially prove that, perhaps surprisingly, linear and logistic regressions with $D \geq 169 H^2/P^2$ parameters are subject to arbitrary model manipulation by poisoners, where $H$ and $P$ are the numbers of honestly labeled and poisoned data points used for training. Our experiments go on exposing a fundamental tradeoff between augmenting model expressivity and increasing the poisoners' attack surface, on both synthetic data, and on MNIST & FashionMNIST data for linear classifiers with random features. We also discuss potential implications for source-based learning and neural nets.