Kirk Lippold discusses the reported three U.S. strikes against Houthis in Yemen on'Your World.' Sailors aboard the aircraft carrier USS Dwight D. Eisenhower and its accompanying warships have spent four months straight at sea defending against ballistic missiles and flying attack drones fired by Iranian-backed Houthis, and are now more regularly also defending against a new threat -- fast unmanned vessels that are fired at them through the water. While the Houthis have launched unmanned surface vessels, or USVs, in the past against Saudi coalition forces that have intervened in Yemen's civil war, they were used for the first time against U.S. military and commercial vessels in the Red Sea on Jan. 4. In the weeks since, the Navy has had to intercept and destroy multiple USVs. It's "more of an unknown threat that we don't have a lot of intel on, that could be extremely lethal -- an unmanned surface vessel," said Rear Adm. Marc Miguez, commander of Carrier Strike Group Two, of which the Eisenhower is the flagship. The Houthis "have ways of obviously controlling them just like they do the (unmanned aerial vehicles), and we have very little little fidelity as to all the stockpiles of what they have USV-wise," Miguez said.

The dominant paradigm in generative modeling consists of two steps: i) pre-training on a large-scale but unsafe dataset, ii) aligning the pre-trained model with human values via fine-tuning. This practice is considered safe, as no current method can recover the unsafe, pre-fine-tuning model weights. In this paper, we demonstrate that this assumption is often false. Concretely, we present Spectral DeTuning, a method that can recover the weights of the pre-fine-tuning model using a few low-rank (LoRA) fine-tuned models. In contrast to previous attacks that attempt to recover pre-fine-tuning capabilities, our method aims to recover the exact pre-fine-tuning weights. Our approach exploits this new vulnerability against large-scale models such as a personalized Stable Diffusion and an aligned Mistral.

Mussel farming is one of the most important aquaculture industries. The main risk to mussel farming is harmful algal blooms (HABs), which pose a risk to human consumption. In Galicia, the Spanish main producer of cultivated mussels, the opening and closing of the production areas is controlled by a monitoring program. In addition to the closures resulting from the presence of toxicity exceeding the legal threshold, in the absence of a confirmatory sampling and the existence of risk factors, precautionary closures may be applied. These decisions are made by experts without the support or formalisation of the experience on which they are based. Therefore, this work proposes a predictive model capable of supporting the application of precautionary closures. Achieving sensitivity, accuracy and kappa index values of 97.34%, 91.83% and 0.75 respectively, the kNN algorithm has provided the best results. This allows the creation of a system capable of helping in complex situations where forecast errors are more common.

Houthi militants in Yemen are using Iranian-supplied missiles and drones to attack civilian and military targets across the Middle East, analysis from the Defense Intelligence Agency (DIA) shows. The report, "Iran: Enabling Houthi Attacks Across the Middle East," aims to provide more insight into the relationship between Iran and the Houthis. The militant group, stationed in Yemen, has for months been striking commercial vessels traveling through the Red Sea in protest of Palestinian civilians killed during Israel's ongoing offensive against Hamas members in Gaza. Houthi fighters stage a rally in support of the Palestinians in the Gaza Strip and against the U.S.-led airstrikes on Yemen, in Sanaa, Yemen, Monday, Jan. 29, 2024. Most recently, Houthi rebels fired ballistic missiles at two ships traveling through Middle East waters.

Combining discrete and continuous data is an important capability for generative models. We present Discrete Flow Models (DFMs), a new flow-based model of discrete data that provides the missing link in enabling flow-based generative models to be applied to multimodal continuous and discrete data problems. Our key insight is that the discrete equivalent of continuous space flow matching can be realized using Continuous Time Markov Chains. DFMs benefit from a simple derivation that includes discrete diffusion models as a specific instance while allowing improved performance over existing diffusion-based approaches. We utilize our DFMs method to build a multimodal flow-based modeling framework. We apply this capability to the task of protein co-design, wherein we learn a model for jointly generating protein structure and sequence. Our approach achieves state-of-the-art co-design performance while allowing the same multimodal model to be used for flexible generation of the sequence or structure.

Anomaly detection in decision-making sequences is a challenging problem due to the complexity of normality representation learning and the sequential nature of the task. Most existing methods based on Reinforcement Learning (RL) are difficult to implement in the real world due to unrealistic assumptions, such as having access to environment dynamics, reward signals, and online interactions with the environment. To address these limitations, we propose an unsupervised method named Offline Imitation Learning based Anomaly Detection (OIL-AD), which detects anomalies in decision-making sequences using two extracted behaviour features: action optimality and sequential association. Our offline learning model is an adaptation of behavioural cloning with a transformer policy network, where we modify the training process to learn a Q function and a state value function from normal trajectories. We propose that the Q function and the state value function can provide sufficient information about agents' behavioural data, from which we derive two features for anomaly detection. The intuition behind our method is that the action optimality feature derived from the Q function can differentiate the optimal action from others at each local state, and the sequential association feature derived from the state value function has the potential to maintain the temporal correlations between decisions (state-action pairs). Our experiments show that OIL-AD can achieve outstanding online anomaly detection performance with up to 34.8% improvement in F1 score over comparable baselines.

Knowledge hallucination have raised widespread concerns for the security and reliability of deployed LLMs. Previous efforts in detecting hallucinations have been employed at logit-level uncertainty estimation or language-level self-consistency evaluation, where the semantic information is inevitably lost during the tokendecoding procedure. Thus, we propose to explore the dense semantic information retained within LLMs' INternal States for hallucInation DEtection (INSIDE). In particular, a simple yet effective EigenScore metric is proposed to better evaluate responses' self-consistency, which exploits the eigenvalues of responses' covariance matrix to measure the semantic consistency/diversity in the dense embedding space. Furthermore, from the perspective of self-consistent hallucination detection, a test time feature clipping approach is explored to truncate extreme activations in the internal states, which reduces overconfident generations and potentially benefits the detection of overconfident hallucinations. Extensive experiments and ablation studies are performed on several popular LLMs and questionanswering (QA) benchmarks, showing the effectiveness of our proposal. Large Language Models (LLMs) have recently achieved a milestone breakthrough and demonstrated impressive abilities in various applications (Ouyang et al., 2022; OpenAI, 2023). However, it has been widely observed that even the state-of-the-art LLMs often make factually incorrect or nonsense generations (Cohen et al., 2023; Ren et al., 2022; Kuhn et al., 2022), which is also known as knowledge hallucination (Ji et al., 2023). The potentially unreliable generations make it risky to deploy LLMs in practical scenarios.

Four months into the war between Israel and Hamas, the combatants, their allies, and their neighbors are closer than ever to reaching a cease-fire or even a settlement of their disputes--and are also equally close to seeing it spin out of control into a widening regional conflict. They are tracing this thin line between negotiated peace and escalating mayhem along every front of the Middle East's hot spots, which are intensifying, enlarging, and mingling with one another--a fact that makes it harder but also potentially more manageable to douse the flames. On Friday, U.S. combat planes fired 125 precision-guided missiles and drones at 85 targets into seven facilities--command-control and intelligence centers, supply lines and storage sites for rockets, missiles, and drones, as well as other military targets--all run by Iranian-backed militias in Iraq and Syria. The attack was in retaliation to a Jan. 28 drone strike launched by one of those militias in Iraq that killed three U.S. soldiers at a base in northeastern Jordan, near the Iraqi and Syrian borders. Militias had fired 165 drones or missiles at U.S. forces in the region since Hamas' Oct. 7 attack, but this was the first strike that killed Americans.

Former Acting Defense Secretary Chris Miller joined'Fox & Friends' to discuss the latest on the escalation in the Middle East as the U.S. continues to strike Iranian proxies. A drone attack late Sunday evening that struck a military base in eastern Syria, where U.S. troops are stationed, left at least six allied Kurdish soldiers dead, officials said. The attack hit a training ground at al-Omar base in Syria's eastern province of Deir el-Zour, the U.S.-backed, Kurdish-led Syrian Democratic Forces (SDF) said in a statement Monday. According to the statement, the drone attack struck an area where the forces' commando units were being trained. No U.S. troops were killed or injured in the attack, they said.

We present the DURel tool that implements the annotation of semantic proximity between uses of words into an online, open source interface. The tool supports standardized human annotation as well as computational annotation, building on recent advances with Word-in-Context models. Annotator judgments are clustered with automatic graph clustering techniques and visualized for analysis. This allows to measure word senses with simple and intuitive micro-task judgments between use pairs, requiring minimal preparation efforts. The tool offers additional functionalities to compare the agreement between annotators to guarantee the inter-subjectivity of the obtained judgments and to calculate summary statistics giving insights into sense frequency distributions, semantic variation or changes of senses over time.