Recent work on dictionary learning with set-atoms has shown benefits in anomaly detection. Instead of viewing an atom as a single vector, these methods allow building sparse representations with atoms taken from a set around a central vector; the set can be a cone or may have a probability distribution associated to it. We propose a method for adaptively adjusting the size of set-atoms in Gaussian and cone dictionary learning. The purpose of the algorithm is to match the atom sizes with their contribution in representing the signals. The proposed algorithm not only decreases the representation error, but also improves anomaly detection, for a class of anomalies called `dependency'. We obtain better detection performance than state-of-the-art methods.

Latvia and Romania, NATO members who are allies of Ukraine, have said that Russian drones violated their airspace. Romania said a Russian drone entered its airspace during nighttime attacks across the Danube River in neighbouring Ukraine in the early hours of Sunday, while Latvia said one crashed in the eastern part of the country a day earlier. Romania's Ministry of National Defence said that Bucharest deployed F-16 fighter jets to monitor its airspace and that a search for the weapon's debris was under way at a potential crash site near the border. There were no immediate reports of casualties or damage. Bucharest strongly condemned the "renewed violation" brought on by Moscow's "illegal attacks".

We introduce Instruct-SkillMix, an automated approach for creating diverse, high quality SFT data. The Instruct-SkillMix pipeline involves two stages, each leveraging an existing powerful LLM: (1) Skill extraction: uses the LLM to extract core "skills" for instruction-following, either from existing datasets, or by directly prompting the model; (2) Data generation: uses the powerful LLM to generate (instruction, response) data that exhibit a randomly chosen pair of these skills. Here, the use of random skill combinations promotes diversity and difficulty. Vanilla SFT (i.e., no PPO, DPO, or RL methods) on data generated from Instruct-SkillMix leads to strong gains on instruction following benchmarks such as AlpacaEval 2.0, MT-Bench, and WildBench. With just $4$K examples, LLaMA-3-8B-Base achieves 42.76% length-controlled win rate on AlpacaEval 2.0. To our knowledge, this achieves state-of-the-art performance among all models that have only undergone SFT (no RL methods) and competes with proprietary models such as Claude 3 Opus and LLaMA-3.1-405B-Instruct. Ablation studies also suggest plausible reasons for why creating open instruction-tuning datasets via naive crowd-sourcing has proved difficult. Introducing low quality answers ("shirkers") in $20\%$ of Instruct-SkillMix examples causes performance to plummet, sometimes catastrophically. The Instruct-SkillMix pipeline is flexible and is adaptable to other settings.

Given an image query, visually prompted keyword localisation (VPKL) aims to find occurrences of the depicted word in a speech collection. This can be useful when transcriptions are not available for a low-resource language (e.g. if it is unwritten). Previous work showed that VPKL can be performed with a visually grounded speech model trained on paired images and unlabelled speech. But all experiments were done on English. Moreover, transcriptions were used to get positive and negative pairs for the contrastive loss. This paper introduces a few-shot learning scheme to mine pairs automatically without transcriptions. On English, this results in only a small drop in performance. We also - for the first time - consider VPKL on a real low-resource language, Yoruba. While scores are reasonable, here we see a bigger drop in performance compared to using ground truth pairs because the mining is less accurate in Yoruba.

This study explores the performance of large language models (LLMs) in solving competitive programming problems from the Romanian Informatics Olympiad at the county level. Romania, a leading nation in computer science competitions, provides an ideal environment for evaluating LLM capabilities due to its rich history and stringent competition standards. We collected and analyzed a dataset comprising 304 challenges from 2002 to 2023, focusing on solutions written by LLMs in C++ and Python for these problems. Our primary goal is to understand why LLMs perform well or poorly on different tasks. We evaluated various models, including closed-source models like GPT-4 and open-weight models such as CodeLlama and RoMistral, using a standardized process involving multiple attempts and feedback rounds. The analysis revealed significant variations in LLM performance across different grades and problem types. Notably, GPT-4 showed strong performance, indicating its potential use as an educational tool for middle school students. We also observed differences in code quality and style across various LLMs

In this paper, we demonstrate that attacks in the latest ASVspoof5 dataset -- a de facto standard in the field of voice authenticity and deepfake detection -- can be identified with surprising accuracy using a small subset of very simplistic features. These are derived from the openSMILE library, and are scalar-valued, easy to compute, and human interpretable. For example, attack A10`s unvoiced segments have a mean length of 0.09 +- 0.02, while bona fide instances have a mean length of 0.18 +- 0.07. Using this feature alone, a threshold classifier achieves an Equal Error Rate (EER) of 10.3% for attack A10. Similarly, across all attacks, we achieve up to 0.8% EER, with an overall EER of 15.7 +- 6.0%. We explore the generalization capabilities of these features and find that some of them transfer effectively between attacks, primarily when the attacks originate from similar Text-to-Speech (TTS) architectures. This finding may indicate that voice anti-spoofing is, in part, a problem of identifying and remembering signatures or fingerprints of individual TTS systems. This allows to better understand anti-spoofing models and their challenges in real-world application.

This paper explores the application of the Condorcet Jury theorem to the domain of sentiment analysis, specifically examining the performance of various large language models (LLMs) compared to simpler natural language processing (NLP) models. The theorem posits that a majority vote classifier should enhance predictive accuracy, provided that individual classifiers' decisions are independent. Our empirical study tests this theoretical framework by implementing a majority vote mechanism across different models, including advanced LLMs such as ChatGPT 4. Contrary to expectations, the results reveal only marginal improvements in performance when incorporating larger models, suggesting a lack of independence among them. This finding aligns with the hypothesis that despite their complexity, LLMs do not significantly outperform simpler models in reasoning tasks within sentiment analysis, showing the practical limits of model independence in the context of advanced NLP tasks.

Transformer models have achieved state-of-the-art results, with Large Language Models (LLMs), an evolution of first-generation transformers (1stTR), being considered the cutting edge in several NLP tasks. However, the literature has yet to conclusively demonstrate that LLMs consistently outperform 1stTRs across all NLP tasks. This study compares three 1stTRs (BERT, RoBERTa, and BART) with two open LLMs (Llama 2 and Bloom) across 11 sentiment analysis datasets. The results indicate that open LLMs may moderately outperform or match 1stTRs in 8 out of 11 datasets but only when fine-tuned. Given this substantial cost for only moderate gains, the practical applicability of these models in cost-sensitive scenarios is questionable. In this context, a confidence-based strategy that seamlessly integrates 1stTRs with open LLMs based on prediction certainty is proposed. High-confidence documents are classified by the more cost-effective 1stTRs, while uncertain cases are handled by LLMs in zero-shot or few-shot modes, at a much lower cost than fine-tuned versions. Experiments in sentiment analysis demonstrate that our solution not only outperforms 1stTRs, zero-shot, and few-shot LLMs but also competes closely with fine-tuned LLMs at a fraction of the cost.

Approximate Nearest Neighbor Search (ANNS) is a fundamental and critical component in many applications, including recommendation systems and large language model-based applications. With the advancement of multimodal neural models, which transform data from different modalities into a shared high-dimensional space as feature vectors, cross-modal ANNS aims to use the data vector from one modality (e.g., texts) as the query to retrieve the most similar items from another (e.g., images or videos). However, there is an inherent distribution gap between embeddings from different modalities, and cross-modal queries become Out-of-Distribution (OOD) to the base data. Consequently, state-of-the-art ANNS approaches suffer poor performance for OOD workloads. In this paper, we quantitatively analyze the properties of the OOD workloads to gain an understanding of their ANNS efficiency. Unlike single-modal workloads, we reveal OOD queries spatially deviate from base data, and the k-nearest neighbors of an OOD query are distant from each other in the embedding space. The property breaks the assumptions of existing ANNS approaches and mismatches their design for efficient search. With insights from the OOD workloads, we propose pRojected bipartite Graph (RoarGraph), an efficient ANNS graph index built under the guidance of query distribution. Extensive experiments show that RoarGraph significantly outperforms state-of-the-art approaches on modern cross-modal datasets, achieving up to 3.56x faster search speed at a 90% recall rate for OOD queries.

In this work, we survey recent studies on masked image modeling (MIM), an approach that emerged as a powerful self-supervised learning technique in computer vision. The MIM task involves masking some information, e.g. pixels, patches, or even latent representations, and training a model, usually an autoencoder, to predicting the missing information by using the context available in the visible part of the input. We identify and formalize two categories of approaches on how to implement MIM as a pretext task, one based on reconstruction and one based on contrastive learning. Then, we construct a taxonomy and review the most prominent papers in recent years. We complement the manually constructed taxonomy with a dendrogram obtained by applying a hierarchical clustering algorithm. We further identify relevant clusters via manually inspecting the resulting dendrogram. Our review also includes datasets that are commonly used in MIM research. We aggregate the performance results of various masked image modeling methods on the most popular datasets, to facilitate the comparison of competing methods. Finally, we identify research gaps and propose several interesting directions of future work.