Can Transformers predict new syllogisms by composing established ones? More generally, what type of targets can be learned by such models from scratch? Recent works show that Transformers can be Turing-complete in terms of expressivity, but this does not address the learnability objective. This paper puts forward the notion of 'distribution locality' to capture when weak learning is efficiently achievable by regular Transformers, where the locality measures the least number of tokens required in addition to the tokens histogram to correlate nontrivially with the target. As shown experimentally and theoretically under additional assumptions, distributions with high locality cannot be learned efficiently. In particular, syllogisms cannot be composed on long chains. Furthermore, we show that (i) an agnostic scratchpad cannot help to break the locality barrier, (ii) an educated scratchpad can help if it breaks the locality at each step, (iii) a notion of 'inductive scratchpad' can both break the locality and improve the out-of-distribution generalization, e.g., generalizing to almost double input size for some arithmetic tasks.

Woman critically injured in ride'malfunction' 10 hours agoShareBBCOn Sunday tarpaulin was seen around one of the rides, although it is not clear which ride suffered the malfunction A woman in her 40s is in hospital with life-threatening injuries after a funfair ride malfunctioned at a country show in south London, the Met Police has said. The incident happened during Lambeth Country Show in Brockwell Park at about 18:20 BST on Saturday. A man in his 40s is also being treated for "potentially life-threatening injuries", the force said. Lambeth Council said the investigation would "determine the cause of the malfunction". Two other people, a man in his 50s and an 11-year-old girl, were injured in the incident and have since been discharged from hospital, the Met said.

Visual Question Answering (VQA) is an important task in multimodal AI, and it is often used to test the ability of vision-language models to understand and reason on knowledge present in both visual and textual data. However, most of the current VQA models use datasets that are primarily focused on English and a few major world languages, with images that are typically Western-centric. While recent efforts have tried to increase the number of languages covered on VQA datasets, they still lack diversity in low-resource languages. More importantly, although these datasets often extend their linguistic range via translation or some other approaches, they usually keep images the same, resulting in narrow cultural representation. To address these limitations, we construct CVQA, a new Culturally-diverse multilingual Visual Question Answering benchmark, designed to cover a rich set of languages and cultures, where we engage native speakers and cultural experts in the data collection process. As a result, CVQA includes culturally-driven images and questions from across 28 countries on four continents, covering 26 languages with 11 scripts, providing a total of 9k questions. We then benchmark several Multimodal Large Language Models (MLLMs) on CVQA, and show that the dataset is challenging for the current state-of-the-art models. This benchmark can serve as a probing evaluation suite for assessing the cultural capability and bias of multimodal models and hopefully encourage more research efforts toward increasing cultural awareness and linguistic diversity in this field.

Land Use Land Cover (LULC) analysis on satellite images using deep learning-based methods is significantly helpful in understanding the geography, socio-economic conditions, poverty levels, and urban sprawl in developing countries. Recent works involve segmentation with LULC classes such as farmland, built-up areas, forests, meadows, water bodies, etc. Training deep learning methods on satellite images requires large sets of images annotated with LULC classes. However, annotated data for developing countries are scarce due to a lack of funding, absence of dedicated residential/industrial/economic zones, a large population, and diverse building materials. BD-SAT provides a high-resolution dataset that includes pixel-by-pixel LULC annotations for Dhaka metropolitan city and surrounding rural/urban areas. Using a strict and standardized procedure, the ground truth is created using Bing satellite imagery with a ground spatial distance of 2.22 meters per pixel. A three-stage, well-defined annotation process has been followed with support from GIS experts to ensure the reliability of the annotations. We performed several experiments to establish benchmark results. The results show that the annotated BD-SAT is sufficient to train large deep learning models with adequate accuracy for five major LULC classes: forest, farmland, built-up areas, water bodies, and meadows.

The widespread use of large language models has brought up essential questions about the potential biases these models might learn. This led to the development of several metrics aimed at evaluating and mitigating these biases. In this paper, we first demonstrate that prompt-based fairness metrics exhibit poor agreement, as measured by correlation, raising important questions about the reliability of fairness assessment using prompts. Then, we outline six relevant reasons why such a low correlation is observed across existing metrics. Based on these insights, we propose a method called Correlated Fairness Output (CAIRO) to enhance the correlation between fairness metrics. CAIRO augments the original prompts of a given fairness metric by using several pre-trained language models and then selects the combination of the augmented prompts that achieves the highest correlation across metrics. We show a significant improvement in Pearson correlation from 0.3 and 0.18 to 0.90 and 0.98 across metrics for gender and religion biases, respectively. Our code is available at https://github.com/chandar-lab/CAIRO.

The widespread absence of diacritical marks in Arabic text poses a significant challenge for Arabic natural language processing (NLP). This paper explores instances of naturally occurring diacritics, referred to as "diacritics in the wild," to unveil patterns and latent information across six diverse genres: news articles, novels, children's books, poetry, political documents, and ChatGPT outputs. We present a new annotated dataset that maps real-world partially diacritized words to their maximal full diacritization in context. Additionally, we propose extensions to the analyze-and-disambiguate approach in Arabic NLP to leverage these diacritics, resulting in notable improvements. Our contributions encompass a thorough analysis, valuable datasets, and an extended diacritization algorithm. We release our code and datasets as open source.

Large Language Models (LLMs) often struggle with hallucinations and outdated information. To address this, Information Retrieval (IR) systems can be employed to augment LLMs with up-to-date knowledge. However, existing IR techniques contain deficiencies, posing a performance bottleneck. Given the extensive array of IR systems, combining diverse approaches presents a viable strategy. Nevertheless, prior attempts have yielded restricted efficacy. In this work, we propose an approach that leverages learning-to-rank techniques to combine heterogeneous IR systems. We demonstrate the method on two Retrieval Question Answering (ReQA) tasks. Our empirical findings exhibit a significant performance enhancement, outperforming previous approaches and achieving state-of-the-art results on ReQA SQuAD.

Retrieval-augmented generation (RAG) systems respond to queries by retrieving relevant documents from a knowledge database, then generating an answer by applying an LLM to the retrieved documents. We demonstrate that RAG systems that operate on databases with potentially untrusted content are vulnerable to a new class of denial-of-service attacks we call jamming. An adversary can add a single ``blocker'' document to the database that will be retrieved in response to a specific query and, furthermore, result in the RAG system not answering the query - ostensibly because it lacks the information or because the answer is unsafe. We describe and analyze several methods for generating blocker documents, including a new method based on black-box optimization that does not require the adversary to know the embedding or LLM used by the target RAG system, nor access to an auxiliary LLM to generate blocker documents. We measure the efficacy of the considered methods against several LLMs and embeddings, and demonstrate that the existing safety metrics for LLMs do not capture their vulnerability to jamming. We then discuss defenses against blocker documents.

As language models (LMs) become capable of handling a wide range of tasks, their evaluation is becoming as challenging as their development. Most generation benchmarks currently assess LMs using abstract evaluation criteria like helpfulness and harmlessness, which often lack the flexibility and granularity of human assessment. Additionally, these benchmarks tend to focus disproportionately on specific capabilities such as instruction following, leading to coverage bias. To overcome these limitations, we introduce the BiGGen Bench, a principled generation benchmark designed to thoroughly evaluate nine distinct capabilities of LMs across 77 diverse tasks. A key feature of the BiGGen Bench is its use of instance-specific evaluation criteria, closely mirroring the nuanced discernment of human evaluation. We apply this benchmark to assess 103 frontier LMs using five evaluator LMs. Our code, data, and evaluation results are all publicly available at https://github.com/prometheus-eval/prometheus-eval/tree/main/BiGGen-Bench.

Digital pathology and microscopy image analysis are widely employed in the segmentation of digitally scanned IHC slides, primarily to identify cancer and pinpoint regions of interest (ROI) indicative of tumor presence. However, current ROI segmentation models are either stain-specific or suffer from the issues of stain and scanner variance due to different staining protocols or modalities across multiple labs. Also, tissues like Ductal Carcinoma in Situ (DCIS), acini, etc. are often classified as Tumors due to their structural similarities and color compositions. In this paper, we proposed a novel convolutional neural network (CNN) based Multi-class Tissue Segmentation model for histopathology whole-slide Breast slides which classify tumors and segments other tissue regions such as Ducts, acini, DCIS, Squamous epithelium, Blood Vessels, Necrosis, etc. as a separate class. Our unique pixel-aligned non-linear merge across spatial resolutions empowers models with both local and global fields of view for accurate detection of various classes. Our proposed model is also able to separate bad regions such as folds, artifacts, blurry regions, bubbles, etc. from tissue regions using multi-level context from different resolutions of WSI. Multi-phase iterative training with context-aware augmentation and increasing noise was used to efficiently train a multi-stain generic model with partial and noisy annotations from 513 slides. Our training pipeline used 12 million patches generated using context-aware augmentations which made our model stain and scanner invariant across data sources. To extrapolate stain and scanner invariance, our model was evaluated on 23000 patches which were for a completely new stain (Hematoxylin and Eosin) from a completely new scanner (Motic) from a different lab. The mean IOU was 0.72 which is on par with model performance on other data sources and scanners.