We propose an explainable approach for relation extraction that mitigates the tension between generalization and explainability by jointly training for the two goals. Our approach uses a multi-task learning architecture, which jointly trains a classifier for relation extraction, and a sequence model that labels words in the context of the relation that explain the decisions of the relation classifier. We also convert the model outputs to rules to bring global explanations to this approach. This sequence model is trained using a hybrid strategy: supervised, when supervision from pre-existing patterns is available, and semi-supervised otherwise. In the latter situation, we treat the sequence model's labels as latent variables, and learn the best assignment that maximizes the performance of the relation classifier. We evaluate the proposed approach on the two datasets and show that the sequence model provides labels that serve as accurate explanations for the relation classifier's decisions, and, importantly, that the joint training generally improves the performance of the relation classifier. We also evaluate the performance of the generated rules and show that the new rules are great add-on to the manual rules and bring the rule-based system much closer to the neural models.

Inspired by the success of self-supervised autoregressive representation learning in natural language (GPT and its variants), and advances in recent visual architecture design with Vision Transformers (ViTs), in this paper, we explore the effect various design choices have on the success of applying such training strategies for visual feature learning. Specifically, we introduce a novel strategy that we call Random Segments with Autoregressive Coding (RandSAC). In RandSAC, we group patch representations (image tokens) into hierarchically arranged segments; within each segment, tokens are predicted in parallel, similar to BERT, while across segment predictions are sequential, similar to GPT. We illustrate that randomized serialization of the segments significantly improves the performance and results in distribution over spatially-long (across-segments) and -short (within-segment) predictions which are effective for feature learning. We illustrate the pertinence of these design choices and explore alternatives on a number of datasets (e.g., CIFAR10, CIFAR100, ImageNet). While our pre-training strategy works with vanilla Transformer, we also propose a conceptually simple, but highly effective, addition to the decoder that allows learnable skip-connections to encoder's feature layers, which further improves the performance. Deep learning has powered enormous successes in Computer Vision and NLP over the past 10, or so, years. It has lead to significant improvements in object detection (Redmon et al., 2016), segmentation (He et al., 2017), as well as higher-level cognition tasks (e.g., Visual Question Answering (Antol et al., 2015), Visual Navigation (Mayo et al., 2021), etc.). These successes have been enabled by both advances in parallel hardware (GPUs) and, perhaps more importantly, large-scale task-specific labeled datasets that allow supervised learning. This appetite for large data has, until very recently, stagnated progress, particularly in building general-purpose visual architectures. These types of considerations date back to the early days of machine learning, and deep learning in particular, where it has long been postulated that unsupervised, or self-supervised, learning could allow learning of robust and general feature representations that can then be readily used (or finetuned) to target tasks. Self-supervised learning has been explored in computer vision in various forms: denoising autoencoders (Pathak et al., 2016; Vincent et al., 2008), colorization (Zhang et al., 2016) or jigsaw puzzle (Doersch et al., 2015; Noroozi & Favaro, 2016) proxy objectives. However, the success of such self-supervised pre-training was somewhat limited. In contrast, the success of similar self-supervised ideas in NLP has been much more dominant with GPT (Brown et al., 2020) and BERT (Devlin et al., 2018) architectures, and their variants.

Misinformation emerges in times of uncertainty when credible information is limited. This is challenging for NLP-based fact-checking as it relies on counter-evidence, which may not yet be available. Despite increasing interest in automatic fact-checking, it is still unclear if automated approaches can realistically refute harmful real-world misinformation. Here, we contrast and compare NLP fact-checking with how professional fact-checkers combat misinformation in the absence of counter-evidence. In our analysis, we show that, by design, existing NLP task definitions for fact-checking cannot refute misinformation as professional fact-checkers do for the majority of claims. We then define two requirements that the evidence in datasets must fulfill for realistic fact-checking: It must be (1) sufficient to refute the claim and (2) not leaked from existing fact-checking articles. We survey existing fact-checking datasets and find that all of them fail to satisfy both criteria. Finally, we perform experiments to demonstrate that models trained on a large-scale fact-checking dataset rely on leaked evidence, which makes them unsuitable in real-world scenarios. Taken together, we show that current NLP fact-checking cannot realistically combat real-world misinformation because it depends on unrealistic assumptions about counter-evidence in the data.

Large Language Models (LLMs) have achieved excellent performances in various tasks. However, fine-tuning an LLM requires extensive supervision. Human, on the other hand, may improve their reasoning abilities by self-thinking without external inputs. In this work, we demonstrate that an LLM is also capable of self-improving with only unlabeled datasets. We use a pre-trained LLM to generate "high-confidence" rationale-augmented answers for unlabeled questions using Chain-of-Thought prompting and self-consistency, and fine-tune the LLM using those self-generated solutions as target outputs. We show that our approach improves the general reasoning ability of a 540B-parameter LLM (74.4% 82.1% on GSM8K, 78.2% 83.0% on DROP, 90.0% 94.4% on OpenBookQA, and 63.4% 67.9% on ANLI-A3) and achieves state-of-the-art-level performance, without any ground truth label. We conduct ablation studies and show that finetuning on reasoning is critical for self-improvement. Scaling has enabled Large Language ...

Cybereason's mission is to'protect it all' – delivering unparalleled prevention, detection, investigation, and response for all endpoints: workstations, laptops, mobile devices and more. Our cyber-defence solutions combine machine learning and AI to analyze threats, connecting huge volumes of data to reveal cyber-attacks and shut them down, as well as block intrusion of known and unknown threats. Since entering the Japan market in 2016, we have seen tremendous growth, now holding #1 market share. We are constantly evolving and hope to expand our team with daring individuals that never give up! Starting this year, we are focusing on reversing the adversaries advantage with the establishment of a new team.

Tehran, Iran – Iran and the West are clashing over Tehran's alleged drone sales to Russia for the war in Ukraine, an issue now being linked to a UN resolution backing the country's nuclear deal with world powers. UN Security Council Resolution 2231 was unanimously adopted in 2015 to endorse the Joint Comprehensive Plan of Action (JCPOA) – the accord that Iran signed with China, Russia, United States, United Kingdom, France and Germany to get sanctions relief in exchange for curbs on its nuclear programme. The US unilaterally abandoned the accord in 2018 and imposed harsh sanctions that remain in place today. Efforts since April 2021 to restore the deal have stalled. European powers are now trying to use a periodic reporting mechanism in the resolution.

Considering how powerful AI systems are, and the roles they increasingly play in helping to make high-stakes decisions about our lives, homes, and societies, they receive surprisingly little formal scrutiny. That's starting to change, thanks to the blossoming field of AI audits. When they work well, these audits allow us to reliably check how well a system is working and figure out how to mitigate any possible bias or harm. Famously, a 2018 audit of commercial facial recognition systems by AI researchers Joy Buolamwini and Timnit Gebru found that the system didn't recognize darker-skinned people as well as white people. For dark-skinned women, the error rate was up to 34%. As AI researcher Abeba Birhane points out in a new essay in Nature, the audit "instigated a body of critical work that has exposed the bias, discrimination, and oppressive nature of facial-analysis algorithms."

The company's previous demo had involved marching a human out in a robot-like body suit, so when Optimus walked slowly around the stage, it was met with delight from the cheering crowd. Despite the show's futuristic framing, robotics experts were mostly underwhelmed by the reveal. Optimus' clunky attempts at something like a dance seemed less advanced than other humanoid robots, such as Honda's Asimo, which played soccer with former President Barack Obama back in 2014. Tesla engineers boasted that Optimus' hand had as many as 11 degrees of freedom (that's to say, all the ways in which robotic parts can bend). In comparison, a robotic hand designed by a Japanese engineer back in 1963 had 27. What is it about Optimus that makes us feel threatened?

Former Israeli Prime Minister Benjamin Netanyahu says it's time for everyone to take a stand against the Iranian regime on'One Nation with Brian Kilmeade.' Russia is "probably" paying for Iranian kamikaze drones by assisting Iran's nuclear research programs, Ukrainian President Volodymyr Zelenskyy said Monday. Zelenskyy made the statement during an address to the Haaretz Democracy Conference on Monday, saying Russia has purchased at least 2,000 Shahed-136 drones and has used them to bombard Ukraine. Iranian instructors have been spotted in Belarus teaching Russian forces to coordinate drone strikes with the Iranian-made drone system, leading to further fears that Belarus' role in the conflict in Ukraine may soon escalate. "Russia has used almost 4,500 missiles against us, and their stock of missiles is dwindling," Zelenskyy said.

When you picture a scientist, do you see a white coat-clad PhD-holder pipetting away at a lab bench? Or maybe a skygazer with a different day job who goes out on clear nights for a good view of the stars? Historically speaking, both of those examples fit the bill. German-British astronomer William Herschel was originally an amateur who observed the night sky using homemade telescopes. He discovered Uranus in 1781, working alongside his sister, Caroline Herschel, who made multiple discoveries herself.