Search

Some information is factual (e.g., "Paris is in France"), whereas other information is probabilistic (e.g., "the coin flip will be a [Heads/Tails]."). We believe that good Language Models (LMs) should understand and reflect this nuance. Our work investigates this by testing if LMs' output probabilities are calibrated to their textual contexts. We define model "calibration" as the degree to which the output probabilities of candidate tokens are aligned with the relative likelihood that should be inferred from the given context. For example, if the context concerns two equally likely options (e.g., heads or tails for a fair coin), the output probabilities should reflect this. Likewise, context that concerns non-uniformly likely events (e.g., rolling a six with a die) should also be appropriately captured with proportionate output probabilities. We find that even in simple settings the best LMs (1) are poorly calibrated, and (2) have systematic biases (e.g., preferred colors and sensitivities to word orderings). For example, gpt-4o-mini often picks the first of two options presented in the prompt regardless of the options' implied likelihood, whereas Llama-3.1-8B picks the second. Our other consistent finding is mode-collapse: Instruction-tuned models often over-allocate probability mass on a single option. These systematic biases introduce non-intuitive model behavior, making models harder for users to understand., Comment: 10 pages (main), 24 pages (appendix), under review

This research tackles the challenge of speckle noise in Synthetic Aperture Radar (SAR) space data, a prevalent issue that hampers the clarity and utility of SAR images. The study presents a comparative analysis of six distinct speckle noise reduction techniques: Lee Filtering, Frost Filtering, Kuan Filtering, Gaussian Filtering, Median Filtering, and Bilateral Filtering. These methods, selected for their unique approaches to noise reduction and image preservation, were applied to SAR datasets sourced from the Alaska Satellite Facility (ASF). The performance of each technique was evaluated using a comprehensive set of metrics, including Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), Structural Similarity Index (SSIM), Equivalent Number of Looks (ENL), and Speckle Suppression Index (SSI). The study concludes that both the Lee and Kuan Filters are effective, with the choice of filter depending on the specific application requirements for image quality and noise suppression. This work provides valuable insights into optimizing SAR image processing, with significant implications for remote sensing, environmental monitoring, and geological surveying.

Training large language models (LLMs) for pretraining or adapting to new tasks and domains has become increasingly critical as their applications expand. However, as the model and the data sizes grow, the training process presents significant memory challenges, often requiring a prohibitive amount of GPU memory that may not be readily available. Existing methods such as low-rank adaptation (LoRA) add trainable low-rank matrix factorizations, altering the training dynamics and limiting the model's parameter search to a low-rank subspace. GaLore, a more recent method, employs Gradient Low-Rank Projection to reduce the memory footprint, in the full parameter training setting. However GaLore can only be applied to a subset of the LLM layers that satisfy the "reversibility" property, thus limiting their applicability. In response to these challenges, we introduce BlockLLM, an approach inspired by block coordinate descent. Our method carefully selects and updates a very small subset of the trainable parameters without altering any part of its architecture and training procedure. BlockLLM achieves state-of-the-art performance in both finetuning and pretraining tasks, while reducing the memory footprint of the underlying optimization process. Our experiments demonstrate that fine-tuning with only less than 5% of the parameters, BlockLLM achieves state-of-the-art perplexity scores on the GLUE benchmarks. On Llama model pretrained on C4 dataset, BlockLLM is able to train with significantly less memory than the state-of-the-art, while still maintaining competitive performance., Comment: 16 pages, 7 figures

FActScore has gained popularity as a metric to estimate the factuality of long-form texts generated by Large Language Models (LLMs) in English. However, there has not been any work in studying the behavior of FActScore in other languages. This paper studies the limitations of each component in the four-component pipeline of FActScore in the multilingual setting. We introduce a new dataset for FActScore on texts generated by strong multilingual LLMs. Our evaluation shows that LLMs exhibit distinct behaviors in both fact extraction and fact scoring tasks. No LLM produces consistent and reliable FActScore across languages with varying levels of resources. We also find that the knowledge source plays an important role in the quality of the estimated FActScore. Using Wikipedia as the knowledge source may hinder the true FActScore of long-form text due to its limited coverage in medium- and low-resource languages. We also incorporate three mitigations to our knowledge source that ultimately improve FActScore estimation across all languages.

The financial domain frequently deals with large numbers of long documents that are essential for daily operations. Significant effort is put towards automating financial data analysis. However, a persistent challenge, not limited to the finance domain, is the scarcity of datasets that accurately reflect real-world tasks for model evaluation. Existing datasets are often constrained by size, context, or relevance to practical applications. Moreover, LLMs are currently trained on trillions of tokens of text, limiting access to novel data or documents that models have not encountered during training for unbiased evaluation. We propose SEC-QA, a continuous dataset generation framework with two key features: 1) the semi-automatic generation of Question-Answer (QA) pairs spanning multiple long context financial documents, which better represent real-world financial scenarios; 2) the ability to continually refresh the dataset using the most recent public document collections, not yet ingested by LLMs. Our experiments show that current retrieval augmented generation methods systematically fail to answer these challenging multi-document questions. In response, we introduce a QA system based on program-of-thought that improves the ability to perform complex information retrieval and quantitative reasoning pipelines, thereby increasing QA accuracy.

The detection of cardiovascular diseases (CVD) using machine learning techniques represents a significant advancement in medical diagnostics, aiming to enhance early detection, accuracy, and efficiency. This study explores a comparative analysis of various machine learning algorithms, including Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and XGBoost. By utilising a structured workflow encompassing data collection, preprocessing, model selection and hyperparameter tuning, training, evaluation, and choice of the optimal model, this research addresses the critical need for improved diagnostic tools. The findings highlight the efficacy of ensemble methods and advanced algorithms in providing reliable predictions, thereby offering a comprehensive framework for CVD detection that can be readily implemented and adapted in clinical settings.

In the contemporary landscape of technological advancements, the automation of manual processes is crucial, compelling the demand for huge datasets to effectively train and test machines. This research paper is dedicated to the exploration and implementation of an automated approach to generate test cases specifically using Large Language Models. The methodology integrates the use of Open AI to enhance the efficiency and effectiveness of test case generation for training and evaluating Large Language Models. This formalized approach with LLMs simplifies the testing process, making it more efficient and comprehensive. Leveraging natural language understanding, LLMs can intelligently formulate test cases that cover a broad range of REST API properties, ensuring comprehensive testing. The model that is developed during the research is trained using manually collected postman test cases or instances for various Rest APIs. LLMs enhance the creation of Postman test cases by automating the generation of varied and intricate test scenarios. Postman test cases offer streamlined automation, collaboration, and dynamic data handling, providing a user-friendly and efficient approach to API testing compared to traditional test cases. Thus, the model developed not only conforms to current technological standards but also holds the promise of evolving into an idea of substantial importance in future technological advancements.

This paper develops a novel car-following control method to reduce voluntary driver interventions and improve traffic stability in Automated Vehicles (AVs). Through a combination of experimental and empirical analysis, we show how voluntary driver interventions can instigate substantial traffic disturbances that are amplified along the traffic upstream. Motivated by these findings, we present a framework for driver intervention based on evidence accumulation (EA), which describes the evolution of the driver's distrust in automation, ultimately resulting in intervention. Informed through the EA framework, we propose a deep reinforcement learning (DRL)-based car-following control for AVs that is strategically designed to mitigate unnecessary driver intervention and improve traffic stability. Numerical experiments are conducted to demonstrate the effectiveness of the proposed control model.

Tokenization is a foundational step in natural language processing (NLP) tasks, bridging raw text and language models. Existing tokenization approaches like Byte-Pair Encoding (BPE) originate from the field of data compression, and it has been suggested that the effectiveness of BPE stems from its ability to condense text into a relatively small number of tokens. We test the hypothesis that fewer tokens lead to better downstream performance by introducing PathPiece, a new tokenizer that segments a document's text into the minimum number of tokens for a given vocabulary. Through extensive experimentation we find this hypothesis not to be the case, casting doubt on the understanding of the reasons for effective tokenization. To examine which other factors play a role, we evaluate design decisions across all three phases of tokenization: pre-tokenization, vocabulary construction, and segmentation, offering new insights into the design of effective tokenizers. Specifically, we illustrate the importance of pre-tokenization and the benefits of using BPE to initialize vocabulary construction. We train 64 language models with varying tokenization, ranging in size from 350M to 2.4B parameters, all of which are made publicly available., Comment: EMNLP 2024

For large language models (LLMs) to be effective in the financial domain -- where each decision can have a significant impact -- it is necessary to investigate realistic tasks and data. Financial professionals often interact with documents that are hundreds of pages long, but most financial research datasets only deal with short excerpts from these documents. To address this, we introduce a long-document financial QA task. We augment 7,437 questions from the existing FinQA dataset with the full-document context, extending the average context length from under 700 words in FinQA to 123k words in DocFinQA. We conduct extensive experiments over retrieval-based QA pipelines and long-context language models. DocFinQA proves a significant challenge for even state-of-the-art systems. We also provide a case-study on the longest documents in DocFinQA and find that models particularly struggle on these documents. Addressing these challenges may have a wide reaching impact across applications where specificity and long-range contexts are critical, like gene sequences and legal document contract analysis., Comment: 13 pages

In the evolving world, we require more additionally the young era to flourish and evolve into developed land. Most of the population all around the world are unaware of the complications involved in the routine they follow while they are pregnant and how hospital facilities affect maternal health. Maternal Mortality is the death of a pregnant woman due to intricacies correlated to pregnancy, underlying circumstances exacerbated by the pregnancy or management of these situations. It is crucial to consider the Maternal Mortality Rate (MMR) in diverse locations and determine which human routines and hospital facilities diminish the Maternal Mortality Rate (MMR). This research aims to examine and discover the countries which are keeping more lavish threats of MMR and countries alike in MMR encountered. Data is examined and collected for various countries, data consists of the earlier years' observation. From the perspective of Machine Learning, Unsupervised Machine Learning is implemented to perform Cluster Analysis. Therefore the pairs of countries with similar MMR as well as the extreme opposite pair concerning the MMR are found., Comment: 14 pages, 6 figures

Answering questions within business and finance requires reasoning, precision, and a wide-breadth of technical knowledge. Together, these requirements make this domain difficult for large language models (LLMs). We introduce BizBench, a benchmark for evaluating models' ability to reason about realistic financial problems. BizBench comprises eight quantitative reasoning tasks, focusing on question-answering (QA) over financial data via program synthesis. We include three financially-themed code-generation tasks from newly collected and augmented QA data. Additionally, we isolate the reasoning capabilities required for financial QA: reading comprehension of financial text and tables for extracting intermediate values, and understanding financial concepts and formulas needed to calculate complex solutions. Collectively, these tasks evaluate a model's financial background knowledge, ability to parse financial documents, and capacity to solve problems with code. We conduct an in-depth evaluation of open-source and commercial LLMs, comparing and contrasting the behavior of code-focused and language-focused models. We demonstrate that the current bottleneck in performance is due to LLMs' limited business and financial understanding, highlighting the value of a challenging benchmark for quantitative reasoning within this domain., Comment: Work in progress

Automated Essay Score (AES) is proven to be one of the cutting-edge technologies. Scoring techniques are used for various purposes. Reliable scores are calculated based on influential variables. Such variables can be computed by different methods based on the domain. The research is concentrated on the user's understanding of a given topic. The analysis is based on a scoring index by using Large Language Models. The user can then compare and contrast the understanding of a topic that they recently learned. The results are then contributed towards learning analytics and progression is made for enhancing the learning ability. In this research, the focus is on summarizing a PDF document and gauging a user's understanding of its content. The process involves utilizing a Langchain tool to summarize the PDF and extract the essential information. By employing this technique, the research aims to determine how well the user comprehends the summarized content.

Transformer based large language models with emergent capabilities are becoming increasingly ubiquitous in society. However, the task of understanding and interpreting their internal workings, in the context of adversarial attacks, remains largely unsolved. Gradient-based universal adversarial attacks have been shown to be highly effective on large language models and potentially dangerous due to their input-agnostic nature. This work presents a novel geometric perspective explaining universal adversarial attacks on large language models. By attacking the 117M parameter GPT-2 model, we find evidence indicating that universal adversarial triggers could be embedding vectors which merely approximate the semantic information in their adversarial training region. This hypothesis is supported by white-box model analysis comprising dimensionality reduction and similarity measurement of hidden representations. We believe this new geometric perspective on the underlying mechanism driving universal attacks could help us gain deeper insight into the internal workings and failure modes of LLMs, thus enabling their mitigation., Comment: 2nd AdvML Frontiers Workshop at 40th International Conference on Machine Learning, Honolulu, Hawaii, USA, 2023

One of the most intriguing debates regarding a novel task is the development of search engines and recommendation-based systems in the music industry. Studies have shown a drastic depression in the search engine fields, due to concerning factors such as speed, accuracy and the format of data given for querying. Often people face difficulty in searching for a song solely based on the title, hence a solution is proposed to complete a search analysis through a single query input and is matched with the lyrics of the songs present in the database. Hence it is essential to incorporate cutting-edge technology tools for developing a user-friendly search engine. Jina AI is an MLOps framework for building neural search engines that are utilized, in order for the user to obtain accurate results. Jina AI effectively helps to maintain and enhance the quality of performance for the search engine for the query given. An effective search engine and a recommendation system for the music industry, built with JinaAI.

Colorectal cancer is one of the most common cancers in the world. While colonoscopy is an effective screening technique, navigating an endoscope through the colon to detect polyps is challenging. A 3D map of the observed surfaces could enhance the identification of unscreened colon tissue and serve as a training platform. However, reconstructing the colon from video footage remains difficult. Learning-based approaches hold promise as robust alternatives, but necessitate extensive datasets. Establishing a benchmark dataset, the 2022 EndoVis sub-challenge SimCol3D aimed to facilitate data-driven depth and pose prediction during colonoscopy. The challenge was hosted as part of MICCAI 2022 in Singapore. Six teams from around the world and representatives from academia and industry participated in the three sub-challenges: synthetic depth prediction, synthetic pose prediction, and real pose prediction. This paper describes the challenge, the submitted methods, and their results. We show that depth prediction from synthetic colonoscopy images is robustly solvable, while pose estimation remains an open research question.

We consider minimizing a smooth function subject to a summation constraint over its variables. By exploiting a connection between the greedy 2-coordinate update for this problem and equality-constrained steepest descent in the 1-norm, we give a convergence rate for greedy selection under a proximal Polyak-Lojasiewicz assumption that is faster than random selection and independent of the problem dimension $n$. We then consider minimizing with both a summation constraint and bound constraints, as arises in the support vector machine dual problem. Existing greedy rules for this setting either guarantee trivial progress only or require $O(n^2)$ time to compute. We show that bound- and summation-constrained steepest descent in the L1-norm guarantees more progress per iteration than previous rules and can be computed in only $O(n \log n)$ time.

The extremum graph is a succinct representation of the Morse decomposition of a scalar field. It has increasingly become a useful data structure that supports topological feature directed visualization of 2D / 3D scalar fields, and enables dimensionality reduction together with exploratory analysis of high dimensional scalar fields. Current methods that employ the extremum graph compute it either using a simple sequential algorithm for computing the Morse decomposition or by computing the more detailed Morse-Smale complex. Both approaches are typically limited to two and three dimensional scalar fields. We describe a GPU-CPU hybrid parallel algorithm for computing the extremum graph of scalar fields in all dimensions. The proposed shared memory algorithm utilizes both fine grained parallelism and task parallelism to achieve efficiency. An open source software library, TACHYON, that implements the algorithm exhibits superior performance and good scaling behavior., Comment: Appeared in Computer Graphics Forum, 2023

In this paper we propose a mask-conditional synthetic image generation model for creating synthetic satellite imagery datasets. Given a dataset of real high-resolution images and accompanying land cover masks, we show that it is possible to train an upstream conditional synthetic imagery generator, use that generator to create synthetic imagery with the land cover masks, then train a downstream model on the synthetic imagery and land cover masks that achieves similar test performance to a model that was trained with the real imagery. Further, we find that incorporating a mixture of real and synthetic imagery acts as a data augmentation method, producing better models than using only real imagery (0.5834 vs. 0.5235 mIoU). Finally, we find that encouraging diversity of outputs in the upstream model is a necessary component for improved downstream task performance. We have released code for reproducing our work on GitHub, see https://github.com/ms-synthetic-satellite-image/synthetic-satellite-imagery .

Pretrained large language models (LLMs) are becoming increasingly powerful and ubiquitous in mainstream applications such as being a personal assistant, a dialogue model, etc. As these models become proficient in deducing user preferences and offering tailored assistance, there is an increasing concern about the ability of these models to influence, modify and in the extreme case manipulate user preference adversarially. The issue of lack of interpretability in these models in adversarial settings remains largely unsolved. This work tries to study adversarial behavior in user preferences from the lens of attention probing, red teaming and white-box analysis. Specifically, it provides a bird's eye view of existing literature, offers red teaming samples for dialogue models like ChatGPT and GODEL and probes the attention mechanism in the latter for non-adversarial and adversarial settings.

Interpretability provides a means for humans to verify aspects of machine learning (ML) models and empower human+ML teaming in situations where the task cannot be fully automated. Different contexts require explanations with different properties. For example, the kind of explanation required to determine if an early cardiac arrest warning system is ready to be integrated into a care setting is very different from the type of explanation required for a loan applicant to help determine the actions they might need to take to make their application successful. Unfortunately, there is a lack of standardization when it comes to properties of explanations: different papers may use the same term to mean different quantities, and different terms to mean the same quantity. This lack of a standardized terminology and categorization of the properties of ML explanations prevents us from both rigorously comparing interpretable machine learning methods and identifying what properties are needed in what contexts. In this work, we survey properties defined in interpretable machine learning papers, synthesize them based on what they actually measure, and describe the trade-offs between different formulations of these properties. In doing so, we enable more informed selection of task-appropriate formulations of explanation properties as well as standardization for future work in interpretable machine learning., Comment: Short version accepted at NeurIPS 2022 workshops on Progress and Challenges in Building Trustworthy Embodied AI and Trustworthy and Socially Responsible Machine Learning