As large language models become increasingly integrated into daily life, audio has emerged as a key interface for human-AI interaction. However, this convenience also introduces new vulnerabilities, making audio a potential attack surface for adversaries. Our research introduces WhisperInject, a two-stage adversarial audio attack framework that can manipulate state-of-the-art audio language models to generate harmful content. Our method uses imperceptible perturbations in audio inputs that remain benign to human listeners. The first stage uses a novel reward-based optimization method, Reinforcement Learning with Projected Gradient Descent (RL-PGD), to guide the target model to circumvent its own safety protocols and generate harmful native responses. This native harmful response then serves as the target for Stage 2, Payload Injection, where we use Projected Gradient Descent (PGD) to optimize subtle perturbations that are embedded into benign audio carriers, such as weather queries or greeting messages. Validated under the rigorous StrongREJECT, LlamaGuard, as well as Human Evaluation safety evaluation framework, our experiments demonstrate a success rate exceeding 86% across Qwen2.5-Omni-3B, Qwen2.5-Omni-7B, and Phi-4-Multimodal. Our work demonstrates a new class of practical, audio-native threats, moving beyond theoretical exploits to reveal a feasible and covert method for manipulating AI behavior.

Micro-blogging sources such as the Twitter social network provide valuable real-time data for market prediction models. Investors' opinions in this network follow the fluctuations of the stock markets and often include educated speculations on market opportunities that may have impact on the actions of other investors. In view of this, we propose a novel system to detect positive predictions in tweets, a type of financial emotions which we term "opportunities" that are akin to "anticipation" in Plutchik's theory. Specifically, we seek a high detection precision to present a financial operator a substantial amount of such tweets while differentiating them from the rest of financial emotions in our system. We achieve it with a three-layer stacked Machine Learning classification system with sophisticated features that result from applying Natural Language Processing techniques to extract valuable linguistic information. Experimental results on a dataset that has been manually annotated with financial emotion and ticker occurrence tags demonstrate that our system yields satisfactory and competitive performance in financial opportunity detection, with precision values up to 83%. This promising outcome endorses the usability of our system to support investors' decision making.

A knowledge graph is a powerful representation of real-world entities and their relations. The vast majority of these relations are defined as positive statements, but the importance of negative statements is increasingly recognized, especially under an Open World Assumption. Explicitly considering negative statements has been shown to improve performance on tasks such as entity summarization and question answering or domain-specific tasks such as protein function prediction. However, no attention has been given to the exploration of negative statements by knowledge graph embedding approaches despite the potential of negative statements to produce more accurate representations of entities in a knowledge graph. We propose a novel approach, TrueWalks, to incorporate negative statements into the knowledge graph representation learning process. In particular, we present a novel walk-generation method that is able to not only differentiate between positive and negative statements but also take into account the semantic implications of negation in ontology-rich knowledge graphs. This is of particular importance for applications in the biomedical domain, where the inadequacy of embedding approaches regarding negative statements at the ontology level has been identified as a crucial limitation. We evaluate TrueWalks in ontology-rich biomedical knowledge graphs in two different predictive tasks based on KG embeddings: protein-protein interaction prediction and gene-disease association prediction. We conduct an extensive analysis over established benchmarks and demonstrate that our method is able to improve the performance of knowledge graph embeddings on all tasks.

Knowledge graphs represent facts about real-world entities. Most of these facts are defined as positive statements. The negative statements are scarce but highly relevant under the open-world assumption. Furthermore, they have been demonstrated to improve the performance of several applications, namely in the biomedical domain. However, no benchmark dataset supports the evaluation of the methods that consider these negative statements. We present a collection of datasets for three relation prediction tasks - protein-protein interaction prediction, gene-disease association prediction and disease prediction - that aim at circumventing the difficulties in building benchmarks for knowledge graphs with negative statements. These datasets include data from two successful biomedical ontologies, Gene Ontology and Human Phenotype Ontology, enriched with negative statements. We also generate knowledge graph embeddings for each dataset with two popular path-based methods and evaluate the performance in each task. The results show that the negative statements can improve the performance of knowledge graph embeddings.

Knowledge bases (KBs), pragmatic collections of knowledge about notable entities, are an important asset in applications such as search, question answering and dialogue. Rooted in a long tradition in knowledge representation, all popular KBs only store positive information, while they abstain from taking any stance towards statements not contained in them. In this paper, we make the case for explicitly stating interesting statements which are not true. Negative statements would be important to overcome current limitations of question answering, yet due to their potential abundance, any effort towards compiling them needs a tight coupling with ranking. We introduce two approaches towards compiling negative statements. (i) In peer-based statistical inferences, we compare entities with highly related entities in order to derive potential negative statements, which we then rank using supervised and unsupervised features. (ii) In query-log-based text extraction, we use a pattern-based approach for harvesting search engine query logs. Experimental results show that both approaches hold promising and complementary potential. Along with this paper, we publish the first datasets on interesting negative information, containing over 1.1M statements for 100K popular Wikidata entities.

In one study, Schuster and team showed that machine learning-taught fact-checking systems struggled to handle negative statements ("Greg never said his car wasn't blue") even when they would know the positive statement was true ("Greg says his car is blue"). The problem, say the researchers, is that the database is filled with human bias. The people who created FEVER tended to write their false entries as negative statements and their true statements as positive statements -- so the computers learned to rate sentences with negative statements as false. That means the systems were solving a much easier problem than detecting fake news. "If you create for yourself an easy target, you can win at that target," said MIT professor Regina Barzilay.

The research was conducted by MIT doctoral student Tal Schuster in two separate studies. While Schuster found that computers are adept and identifying ML-generated text, they have a hard time flagging what's false from what's true. Schuster says the problem lies with the database used to train computers to spot fake news. That database is called Fact Extraction and Verification (FEVER). Schuster found that ML-taught computers struggled to interpret negative statements about a subject even when the computers could easily interpret positive statements.

Chatbots offer a useful way to leverage the power of AI, and are now accessible for any size of application. The back-and-forth written nature of chatting is conducive to utilizing existing chatbot frameworks and AI models to automate interactions which would have required a whole team of agents just a short time ago. To demonstrate how easy it is, we'll use a chatbot framework and a sentiment analysis model from the Serverless AI Cloud -- both of which have free trial tiers. It only takes a few minutes, and it's free to get started. After signing up, click the "New Bot" button and pick "Blank Project": Note the column on the left called "All Topics".