Large Language Models (LLMs) have achieved unprecedented breakthroughs in various natural language processing domains. However, the enigmatic ``black-box'' nature of LLMs remains a significant challenge for interpretability, hampering transparent and accountable applications. While past approaches, such as attention visualization, pivotal subnetwork extraction, and concept-based analyses, offer some insight, they often focus on either local or global explanations within a single dimension, occasionally falling short in providing comprehensive clarity. In response, we propose a novel methodology anchored in sparsity-guided techniques, aiming to provide a holistic interpretation of LLMs. Our framework, termed SparseCBM, innovatively integrates sparsity to elucidate three intertwined layers of interpretation: input, subnetwork, and concept levels. In addition, the newly introduced dimension of interpretable inference-time intervention facilitates dynamic adjustments to the model during deployment. Through rigorous empirical evaluations on real-world datasets, we demonstrate that SparseCBM delivers a profound understanding of LLM behaviors, setting it apart in both interpreting and ameliorating model inaccuracies. Codes are provided in supplements.

Through iterative, cross-disciplinary discussions, we define and propose next-steps for Human-centered Generative AI (HGAI). We contribute a comprehensive research agenda that lays out future directions of Generative AI spanning three levels: aligning with human values; assimilating human intents; and augmenting human abilities. By identifying these next-steps, we intend to draw interdisciplinary research teams to pursue a coherent set of emergent ideas in HGAI, focusing on their interested topics while maintaining a coherent big picture of the future work landscape.

Automatic melody-to-lyric generation is a task in which song lyrics are generated to go with a given melody. It is of significant practical interest and more challenging than unconstrained lyric generation as the music imposes additional constraints onto the lyrics. The training data is limited as most songs are copyrighted, resulting in models that underfit the complicated cross-modal relationship between melody and lyrics. In this work, we propose a method for generating high-quality lyrics without training on any aligned melody-lyric data. Specifically, we design a hierarchical lyric generation framework that first generates a song outline and second the complete lyrics. The framework enables disentanglement of training (based purely on text) from inference (melody-guided text generation) to circumvent the shortage of parallel data. We leverage the segmentation and rhythm alignment between melody and lyrics to compile the given melody into decoding constraints as guidance during inference. The two-step hierarchical design also enables content control via the lyric outline, a much-desired feature for democratizing collaborative song creation. Experimental results show that our model can generate high-quality lyrics that are more on-topic, singable, intelligible, and coherent than strong baselines, for example SongMASS, a SOTA model trained on a parallel dataset, with a 24% relative overall quality improvement based on human ratings.

Last week, OpenAI and the German media conglomerate Axel Springer signed a multi-year licensing agreement. It allows OpenAI to incorporate articles from Axel Springer–owned outlets like Business Insider and Politico into its products, including ChatGPT. Although the deal centers on using journalistic work, reporters whose stories will be shared as part of the agreement were not consulted about the deal beforehand. Four Business Insider employees told WIRED that they found out about the AI deal at the same time it was announced publicly. PEN Guild, the US union which represents around 280 workers at Politico and E&E News, another Axel Springer publication, says it was "not consulted or informed about the decision to have robots summarize our work."

Put bluntly, picking the best movies of 2023 was tough. The double-whammy of Barbie and Oppenheimer gave the box office a long-overdue, post-Covid-19 jolt, only to be followed by a pair of months-long strikes in Hollywood that shut down production on nearly all the films in the works for 2024 and beyond. Even now, with the strikes over, the industry is scratching its head at what happened and what's to come. Still, amidst all the noise, 2023 provided a wealth of quietly beautiful films. Even as Hollywood fretted over the possibility of artificial intelligence upending filmmaking and giving writing and acting gigs to bots, it's impossible to watch the movies on this list and not feel such a possibility is faintly ridiculous.

However, localized editing in complex the other hand, mask-free methods that utilize attention injection scenarios has not been well-studied in the literature, despite mechanisms such as Prompt-to-Prompt (P2P) [10] its growing real-world demands. Existing mask-based and Plug-and-Play (PnP) [28] can preserve the original image's inpainting methods fall short of retaining the underlying structure and layout. Nevertheless, they struggle to structure within the edit region. Meanwhile, mask-free precisely align the local editing region with the intended attention-based methods often exhibit editing leakage and text in intricate scenarios, largely due to their reliance on the misalignment in more complex compositions. In this work, text prompts' localization capabilities. As a result, editing we develop MAG-Edit, a training-free, inference-stage optimization effects often extend beyond the intended area and impact method, which enables localized image editing in incorrect regions, as shown in the fourth column of Figure 1.

Invasive signal crayfish have a detrimental impact on ecosystems. They spread the fungal-type crayfish plague disease (Aphanomyces astaci) that is lethal to the native white clawed crayfish, the only native crayfish species in Britain. Invasive signal crayfish extensively burrow, causing habitat destruction, erosion of river banks and adverse changes in water quality, while also competing with native species for resources and leading to declines in native populations. Moreover, pollution exacerbates the vulnerability of White-clawed crayfish, with their populations declining by over 90% in certain English counties, making them highly susceptible to extinction. To safeguard aquatic ecosystems, it is imperative to address the challenges posed by invasive species and discarded plastics in the United Kingdom's river ecosystem's. The UDEEP platform can play a crucial role in environmental monitoring by performing on-the-fly classification of Signal crayfish and plastic debris while leveraging the efficacy of AI, IoT devices and the power of edge computing (i.e., NJN). By providing accurate data on the presence, spread and abundance of these species, the UDEEP platform can contribute to monitoring efforts and aid in mitigating the spread of invasive species.

Language model attacks typically assume one of two extreme threat models: full white-box access to model weights, or black-box access limited to a text generation API. However, real-world APIs are often more flexible than just text generation: these APIs expose ``gray-box'' access leading to new threat vectors. To explore this, we red-team three new functionalities exposed in the GPT-4 APIs: fine-tuning, function calling and knowledge retrieval. We find that fine-tuning a model on as few as 15 harmful examples or 100 benign examples can remove core safeguards from GPT-4, enabling a range of harmful outputs. Furthermore, we find that GPT-4 Assistants readily divulge the function call schema and can be made to execute arbitrary function calls. Finally, we find that knowledge retrieval can be hijacked by injecting instructions into retrieval documents. These vulnerabilities highlight that any additions to the functionality exposed by an API can create new vulnerabilities.

In the field of audio and speech analysis, the ability to identify emotions from acoustic signals is essential. Human-computer interaction (HCI) and behavioural analysis are only a few of the many areas where the capacity to distinguish emotions from speech signals has an extensive range of applications. Here, we are introducing BanSpEmo, a corpus of emotional speech that only consists of audio recordings and has been created specifically for the Bangla language. This corpus contains 792 audio recordings over a duration of more than 1 hour and 23 minutes. 22 native speakers took part in the recording of two sets of sentences that represent the six desired emotions. The data set consists of 12 Bangla sentences which are uttered in 6 emotions as Disgust, Happy, Sad, Surprised, Anger, and Fear. This corpus is not also gender balanced. Ten individuals who either have experience in related field or have acting experience took part in the assessment of this corpus. It has a balanced number of audio recordings in each emotion class. BanSpEmo can be considered as a useful resource to promote emotion and speech recognition research and related applications in the Bangla language. The dataset can be found here: https://data.mendeley.com/datasets/rdwn4bs5ky and might be employed for academic research.

Current state-of-the-art audio analysis systems rely on pre-trained embedding models, often used off-the-shelf as (frozen) feature extractors. Choosing the best one for a set of tasks is the subject of many recent publications. However, one aspect often overlooked in these works is the influence of the duration of audio input considered to extract an embedding, which we refer to as Temporal Support (TS). In this work, we study the influence of the TS for well-established or emerging pre-trained embeddings, chosen to represent different types of architectures and learning paradigms. We conduct this evaluation using both musical instrument and environmental sound datasets, namely OpenMIC, TAU Urban Acoustic Scenes 2020 Mobile, and ESC-50. We especially highlight that Audio Spectrogram Transformer-based systems (PaSST and BEATs) remain effective with smaller TS, which therefore allows for a drastic reduction in memory and computational cost. Moreover, we show that by choosing the optimal TS we reach competitive results across all tasks. In particular, we improve the state-of-the-art results on OpenMIC, using BEATs and PaSST without any fine-tuning.