We investigate the CLIP image encoder by analyzing how individual model components affect the final representation. We decompose the image representation as a sum across individual image patches, model layers, and attention heads, and use CLIP's text representation to interpret the summands. Interpreting the attention heads, we characterize each head's role by automatically finding text representations that span its output space, which reveals property-specific roles for many heads (e.g. location or shape). Next, interpreting the image patches, we uncover an emergent spatial localization within CLIP. Finally, we use this understanding to remove spurious features from CLIP and to create a strong zero-shot image segmenter. Our results indicate that a scalable understanding of transformer models is attainable and can be used to repair and improve models.

Charting the Landscape of Nefarious Applications of Generative Artificial Intelligence and Large Language Models Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs) are marvels of technology; celebrated for their prowess in natural language processing and multimodal content generation, they promise a transformative future. But as with all powerful tools, they come with their shadows. Picture living in a world where deepfakes are indistinguishable from reality, where synthetic identities orchestrate malicious campaigns, and where targeted misinformation or scams are crafted with unparalleled precision. Welcome to the darker side of GenAI applications. This article is not just a journey through the meanders of potential misuse of GenAI and LLMs, but also a call to recognize the urgency of the challenges ahead. As we navigate the seas of misinformation campaigns, malicious content generation, and the eerie creation of sophisticated malware, we'll uncover the societal implications that ripple through the GenAI revolution we are witnessing. From AI-powered botnets on social media platforms to the unnerving potential of AI to generate fabricated identities, or alibis made of synthetic realities, the stakes have never been higher. The lines between the virtual and the real worlds are blurring, and the consequences of potential GenAI's nefarious applications impact us all. This article serves both as a synthesis of rigorous research presented on the risks of GenAI and misuse of LLMs and as a thought-provoking vision of the different types of harmful GenAI applications we might encounter in the near future, and some ways we can prepare for them. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. INTRODUCTION In March 2019, a UK-based energy firm's CEO was duped out of $243,000.

Detecting fake news requires both a delicate sense of diverse clues and a profound understanding of the real-world background, which remains challenging for detectors based on small language models (SLMs) due to their knowledge and capability limitations. Recent advances in large language models (LLMs) have shown remarkable performance in various tasks, but whether and how LLMs could help with fake news detection remains underexplored. In this paper, we investigate the potential of LLMs in fake news detection. First, we conduct an empirical study and find that a sophisticated LLM such as GPT 3.5 could generally expose fake news and provide desirable multi-perspective rationales but still underperforms the basic SLM, fine-tuned BERT. Our subsequent analysis attributes such a gap to the LLM's inability to select and integrate rationales properly to conclude. Based on these findings, we propose that current LLMs may not substitute fine-tuned SLMs in fake news detection but can be a good advisor for SLMs by providing multi-perspective instructive rationales. To instantiate this proposal, we design an adaptive rationale guidance network for fake news detection (ARG), in which SLMs selectively acquire insights on news analysis from the LLMs' rationales. We further derive a rationale-free version of ARG by distillation, namely ARG-D, which services cost-sensitive scenarios without querying LLMs. Experiments on two real-world datasets demonstrate that ARG and ARG-D outperform three types of baseline methods, including SLM-based, LLM-based, and combinations of small and large language models.

Robot dog uses sensors, cameras and artificial intelligence to perceive and navigate surroundings. Agnieszka Pilat is not your typical artist. She doesn't use brushes, pencils or even her own hands to create her artwork. Pilat, who was born in Poland and now lives in the U.S., spent months teaching three of these four-legged machines named Basia, Vanya and Bunny to hold a paintbrush in their "mouths" and move them across a large canvas, turning the paint into abstract art. They use sensors, cameras and artificial intelligence to perceive and navigate their surroundings.

Natalie Nasatka, who says her Apple Watch saved her life, joined'Fox & Friends' to discuss the incident after the near-fatal poisoning. As Apple brings the Apple Vision Pro to market, its first major product category in nine years, the absence of key streaming apps and unconventional design choices cast shadows over its debut. Apple Inc.'s upcoming mixed-reality headset, the Apple Vision Pro, is poised to make waves. It lets you see and interact with virtual worlds using artificial intelligence, voice and gesture control, plus spatial audio. The world's leading video and music streaming services, including Google's YouTube, Spotify and Netflix, are noticeably absent from the device's lineup of supported applications.

Deep learning-based (DL) models in recommender systems (RecSys) have gained significant recognition for their remarkable accuracy in predicting user preferences. However, their performance often lacks a comprehensive evaluation from a human-centric perspective, which encompasses various dimensions beyond simple interest matching. In this work, we have developed a robust human-centric evaluation framework that incorporates seven diverse metrics to assess the quality of recommendations generated by five recent open-sourced DL models. Our evaluation datasets consist of both offline benchmark data and personalized online recommendation feedback collected from 445 real users. We find that (1) different DL models have different pros and cons in the multi-dimensional metrics that we test with; (2) users generally want a combination of accuracy with at least one another human values in the recommendation; (3) the degree of combination of different values needs to be carefully experimented to user preferred level.

We review research on generating visual data from text from the angle of "cross-modal generation." This point of view allows us to draw parallels between various methods geared towards working on input text and producing visual output, without limiting the analysis to narrow sub-areas. It also results in the identification of common templates in the field, which are then compared and contrasted both within pools of similar methods and across lines of research. We provide a breakdown of text-to-image generation into various flavors of image-from-text methods, video-from-text methods, image editing, self-supervised and graph-based approaches. In this discussion, we focus on research papers published at 8 leading machine learning conferences in the years 2016-2022, also incorporating a number of relevant papers not matching the outlined search criteria. The conducted review suggests a significant increase in the number of papers published in the area and highlights research gaps and potential lines of investigation. To our knowledge, this is the first review to systematically look at text-to-image generation from the perspective of "cross-modal generation."

The rapid and widespread dissemination of misinformation through social networks is a growing concern in today's digital age. This study focused on modeling fake news diffusion, discovering the spreading dynamics, and designing control strategies. A common approach for modeling the misinformation dynamics is SIR-based models. Our approach is an extension of a model called 'SBFC' which is a SIR-based model. This model has three states, Susceptible, Believer, and Fact-Checker. The dynamics and transition between states are based on neighbors' beliefs, hoax credibility, spreading rate, probability of verifying the news, and probability of forgetting the current state. Our contribution is to push this model to real social networks by considering different classes of agents with their characteristics. We proposed two main strategies for confronting misinformation diffusion. First, we can educate a minor class, like scholars or influencers, to improve their ability to verify the news or remember their state longer. The second strategy is adding fact-checker bots to the network to spread the facts and influence their neighbors' states. Our result shows that both of these approaches can effectively control the misinformation spread.

It has long been recognized that it is not enough for a Recommender System (RS) to provide recommendations based only on their relevance to users. Among many other criteria, the set of recommendations may need to be diverse in order to handle uncertainty and offer a meaningful choice. The literature reports many ways of measuring diversity and ways of improving the diversity of a set of recommendations, most notably by re-ranking and selecting from a larger set of candidate recommendations. Driven by promising insights from the literature on how to incorporate versatile Large Language Models (LLMs) into the RS pipeline, in this paper, we show how LLMs can be used for diversity re-ranking. We begin with an informal study that verifies that LLMs can be used for re-ranking tasks and do have some understanding of the concept of diversity. Then, we design a more rigorous methodology where LLMs are prompted to generate a diverse ranking from a candidate ranking using various prompt templates with different re-ranking instructions in a zero-shot fashion. We conduct comprehensive experiments testing state-of-the-art conversational LLMs from the GPT and Llama families. We compare their re-ranking capabilities with random re-ranking and various traditional re-ranking methods from the literature (MMR, xQuAD and RxQuAD). We find that LLM-based re-ranking outperforms random re-ranking across all the metrics that we use but does not perform as well as the traditional re-ranking methods. We gain insight into prompt design for this task (e.g.\ on the whole, it is better to prompt for diversity rather than a balance of diversity and relevance). Given that no special knowledge engineering is needed, we conclude that LLM-based re-ranking is a promising approach, and we highlight directions for future research. We open-source the code of our experiments for reproducibility.

The rise of singing voice synthesis presents critical challenges to artists and industry stakeholders over unauthorized voice usage. Unlike synthesized speech, synthesized singing voices are typically released in songs containing strong background music that may hide synthesis artifacts. Additionally, singing voices present different acoustic and linguistic characteristics from speech utterances. These unique properties make singing voice deepfake detection a relevant but significantly different problem from synthetic speech detection. In this work, we propose the singing voice deepfake detection task. We first present SingFake, the first curated in-the-wild dataset consisting of 28.93 hours of bonafide and 29.40 hours of deepfake song clips in five languages from 40 singers. We provide a train/validation/test split where the test sets include various scenarios. We then use SingFake to evaluate four state-of-the-art speech countermeasure systems trained on speech utterances. We find these systems lag significantly behind their performance on speech test data. When trained on SingFake, either using separated vocal tracks or song mixtures, these systems show substantial improvement. However, our evaluations also identify challenges associated with unseen singers, communication codecs, languages, and musical contexts, calling for dedicated research into singing voice deepfake detection. The SingFake dataset and related resources are available at https://www.singfake.org/.