Video-text retrieval has seen significant advancements, yet the ability of models to discern subtle differences in captions still requires verification. In this paper, we introduce a new approach for fine-grained evaluation. Our approach can be applied to existing datasets by automatically generating hard negative test captions with subtle single-word variations across nouns, verbs, adjectives, adverbs, and prepositions. We perform comprehensive experiments using four state-of-the-art models across two standard benchmarks (MSR-VTT and VATEX) and two specially curated datasets enriched with detailed descriptions (VLN-UVO and VLN-OOPS), resulting in a number of novel insights: 1) our analyses show that the current evaluation benchmarks fall short in detecting a model's ability to perceive subtle single-word differences, 2) our fine-grained evaluation highlights the difficulty models face in distinguishing such subtle variations. To enhance fine-grained understanding, we propose a new baseline that can be easily combined with current methods. Experiments on our fine-grained evaluations demonstrate that this approach enhances a model's ability to understand fine-grained differences.

We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user's image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models. All videos from this paper are available at https://go.fb.me/MovieGenResearchVideos.

This study investigates how personal differences (digital self-efficacy, technical knowledge, belief in equality, political ideology) and demographic factors (age, education, and income) are associated with perceptions of artificial intelligence (AI) outcomes exhibiting gender and racial bias and with general attitudes towards AI. Analyses of a large-scale experiment dataset (N = 1,206) indicate that digital self-efficacy and technical knowledge are positively associated with attitudes toward AI, while liberal ideologies are negatively associated with outcome trust, higher negative emotion, and greater skepticism. Furthermore, age and income are closely connected to cognitive gaps in understanding discriminatory AI outcomes. These findings highlight the importance of promoting digital literacy skills and enhancing digital self-efficacy to maintain trust in AI and beliefs in AI usefulness and safety. The findings also suggest that the disparities in understanding problematic AI outcomes may be aligned with economic inequalities and generational gaps in society. Overall, this study sheds light on the socio-technological system in which complex interactions occur between social hierarchies, divisions, and machines that reflect and exacerbate the disparities.

Alignment of large language models (LLMs) involves training models on preference-contrastive output pairs to adjust their responses according to human preferences. To obtain such contrastive pairs, traditional methods like RLHF and RLAIF rely on limited contrasting patterns, such as varying model variants or decoding temperatures. This singularity leads to two issues: (1) alignment is not comprehensive; and thereby (2) models are susceptible to jailbreaking attacks. To address these issues, we investigate how to construct more comprehensive and diversified contrasting patterns to enhance preference data (RQ1) and verify the impact of the diversification of contrasting patterns on model alignment (RQ2). For RQ1, we propose PopAlign, a framework that integrates diversified contrasting patterns across the prompt, model, and pipeline levels, introducing six contrasting strategies that do not require additional feedback labeling procedures. Regarding RQ2, we conduct thorough experiments demonstrating that PopAlign significantly outperforms existing methods, leading to more comprehensive alignment.

The well-groomed young man dressed in a crisp, blue shirt speaking with a soft American accent seems an unlikely supporter of the junta leader of the west African state of Burkina Faso. "We must support … President Ibrahim Traoré … Homeland or death we shall overcome!" he says in a video that began circulating in early 2023 on Telegram. It was just a few months after the dictator had come to power via a military coup. Other videos fronted by different people, with a similar professional-looking appearance and repeating the exact same script in front of the Burkina Faso flag, cropped up around the same time. On a verified account on X a few days later the same young man, in the same blue shirt, claimed to be Archie, the chief executive of a new cryptocurrency platform. They were generated with artificial intelligence (AI) developed by a startup based in east London.

A lover of magazines may find a few good reasons to pay attention to AFM, a new publication about sex and relationships. It's also the latest in a long line of magazines to exist only because of the largesse of a tech company. AFM stands for both "A Fucking Magazine" and "A Feeld Magazine"--that second one a reference to the dating app that is funding the enterprise. Feeld started its life in 2014 specifically to facilitate threesomes. It was originally called 3nder, pronounced "Thrinder," which quickly led the company to receive a trademark-infringement complaint from Tinder.

There's a Japanese word, komorebi, that describes beams of light and dappled shadows that result when the sun shines through trees. When I take my dog on walks around my leafy neighborhood in Washington, D.C., komorebi is what most often catches my eye, especially in this autumnal moment when dense, green summer foliage is starting to thin and turn golden. As the sun sets and the shadows grow long on the edge of a precipitous valley near my apartment, the foliage creates fluttering patterns of warm and cool colors. I try to photograph these apparitions with my iPhone camera, but I'm always disappointed in the results: the device's automated image processing treats contrast as a problem to be solved, aggressively darkening the highlights and lightening up the shadows to achieve a bland flatness. Little of the lambent atmosphere I see in real life survives in the image.

Catch up on your favourite BBC radio show from your favourite DJ right here, whenever you like. Listen without limits with BBC Sounds.

It's a very Imogen Heap way to say hello: "I've got to show you this thing – it's going to change your life!" She beams at me, showing off a mysterious black device. The musician and technologist is an electric, eccentric presence even on video call, talking passionately and changing thoughts like a rally driver turns corners. She whirls me from her kitchen floor to her living room in her family home in Havering near London, familiar to thousands of fans (AKA Heapsters) who tune in to watch her improvise, via livestream, on a grand piano. She points to a glamorous white tent on the edge of a well-kept lawn: "That's my tent I've been sleeping in, by the way," she laughs, enjoying the surprise.

With the rapid advancements of various machine learning models, there is a significant demand for model-agnostic explanation techniques, which can explain these models across different architectures. Mainstream model-agnostic explanation techniques generate local explanations based on basic features (e.g., words for text models and (super-)pixels for image models). However, these explanations often do not align with the decision-making processes of the target models and end-users, resulting in explanations that are unfaithful and difficult for users to understand. On the other hand, concept-based techniques provide explanations based on high-level features (e.g., topics for text models and objects for image models), but most are model-specific or require additional pre-defined external concept knowledge. To address this limitation, we propose \toolname, a general framework to provide concept-based local explanations for any machine learning models. Our key insight is that we can automatically extract high-level concepts from large pre-trained models, and uniformly extend existing local model-agnostic techniques to provide unified concept-based explanations. We have instantiated \toolname on four different types of explanation techniques: LIME, Kernel SHAP, Anchor, and LORE, and applied these techniques to text and image models. Our evaluation results demonstrate that 1) compared to the vanilla versions, \toolname offers more faithful explanations and makes them more understandable to users, and 2) by offering multiple forms of explanations, \toolname outperforms state-of-the-art concept-based explanation techniques specifically designed for text and image models, respectively.