Evaluating large vision-language models (LVLMs) is very expensive, due to high computational cost and the wide variety of tasks. The good news is that if we already have some observed performance scores, we may be able to infer unknown ones. In this study, we propose a new framework for predicting unknown performance scores based on observed ones from other LVLMs or tasks. We first formulate the performance prediction as a matrix completion task. Specifically, we construct a sparse performance matrix $\boldsymbol{R}$, where each entry $R_{mn}$ represents the performance score of the $m$-th model on the $n$-th dataset. By applying probabilistic matrix factorization (PMF) with Markov chain Monte Carlo (MCMC), we can complete the performance matrix, i.e., predict unknown scores. Additionally, we estimate the uncertainty of performance prediction based on MCMC. Practitioners can evaluate their models on untested tasks with higher uncertainty first, which quickly reduces the prediction errors. We further introduce several improvements to enhance PMF for scenarios with sparse observed performance scores. Our experiments demonstrate the accuracy of PMF in predicting unknown scores, the reliability of uncertainty estimates in ordering evaluations, and the effectiveness of our enhancements for handling sparse data. Our code is available at https://github.com/Qinyu-Allen-Zhao/CrossPred-LVLM.

The emergence of generative artificial intelligence has profoundly reshaped the landscape of creative production, but it has also raised critical questions regarding authorship, copyright, and ethical accountability. This paper engages with the growing discourse on the legal status of AI-generated content, emphasizing the challenges it poses to existing copyright frameworks and the rights of human creators. Our research proposes a quantitative, multi-modal framework designed to assess the influence of artists whose works are potentially reflected in AI-generated art. Drawing from contemporary discussion in digital ethics, IP law, and AI governance, this work advocates for the development of enforceable mechanisms to ensure transparency, accountability, and attribution in creative AI systems. It also underscores the pressing need for international legal harmonization and policy intervention to safeguard the labor, legacy, and moral rights of artists. The study contributes to the broader movement toward responsible and equitable innovation, where the protection of human intellectual contribution remains central to the evolution of AI-generated cultural production.

Feature shaping refers to a family of methods that exhibit state-of-the-art performance for out-of-distribution (OOD) detection. These approaches manipulate the feature representation, typically from the penultimate layer of a pre-trained deep learning model, so as to better differentiate between in-distribution (ID) and OOD samples. However, existing feature-shaping methods usually employ rules manually designed for specific model architectures and OOD datasets, which consequently limit their generalization ability. To address this gap, we first formulate an abstract optimization framework for studying feature-shaping methods. We then propose a concrete reduction of the framework with a simple piecewise constant shaping function and show that existing feature-shaping methods approximate the optimal solution to the concrete optimization problem. Further, assuming that OOD data is inaccessible, we propose a formulation that yields a closed-form solution for the piecewise constant shaping function, utilizing solely the ID data. Through extensive experiments, we show that the feature-shaping function optimized by our method improves the generalization ability of OOD detection across a large variety of datasets and model architectures. Our code is available at https://github.com/Qinyu-Allen-Zhao/OptFSOOD.

Quantized networks use less computational and memory resources and are suitable for deployment on edge devices. While quantization-aware training (QAT) is a well-studied approach to quantize the networks at low precision, most research focuses on over-parameterized networks for classification with limited studies on popular and edge device friendly single-shot object detection and semantic segmentation methods like YOLO. Moreover, majority of QAT methods rely on Straight Through Estimator (STE) approximation which suffers from an oscillation phenomenon resulting in sub-optimal network quantization. In this paper, we show that it is difficult to achieve extremely low precision (4-bit and lower) for efficient YOLO models even with SOTA QAT methods due to oscillation issue and existing methods to overcome this problem are not effective on these models. To mitigate the effect of oscillation, we first propose Exponentially Moving Average (EMA) based update to the QAT model. Further, we propose a simple QAT correction method, namely QC, that takes only a single epoch of training after standard Quantization-Aware Training (QAT) procedure to correct the error induced by oscillating weights and activations resulting in a more accurate quantized model. With extensive evaluation on COCO dataset using various YOLO5 and YOLO7 variants, we show that our correction method improves quantized YOLO networks consistently on both object detection and segmentation tasks at low-precision (4-bit and 3-bit).

Neural network quantization has become increasingly popular due to efficient memory consumption and faster computation resulting from bitwise operations on the quantized networks. Even though they exhibit excellent generalization capabilities, their robustness properties are not well-understood. In this work, we systematically study the robustness of quantized networks against gradient based adversarial attacks and demonstrate that these quantized models suffer from gradient vanishing issues and show a fake sense of robustness. By attributing gradient vanishing to poor forward-backward signal propagation in the trained network, we introduce a simple temperature scaling approach to mitigate this issue while preserving the decision boundary. Despite being a simple modification to existing gradient based adversarial attacks, experiments on multiple image classification datasets with multiple network architectures demonstrate that our temperature scaled attacks obtain near-perfect success rate on quantized networks while outperforming original attacks on adversarially trained models as well as floating-point networks.

The Sustainable Development Goals (SDGs) are the framework adopted by the global community to encourage taking actions on the multiple challenges facing the world today to ensure environmental protection, health and well-being, and economic prosperity. This framework provides a detailed list of indicators that are interconnected and cover a holistic view on sustainable development. The goals were defined by the United Nations General Assembly in 2015 and expected to be achieved by 2030. Since the release of this agenda, the research community has begun to intensify work in these areas, yet these efforts seem to be relatively limited. This is especially true about the employment of data and artificial intelligence (AI), which are not widely engaged in SDGs related topics. The AI-based research on SDGs and further developments depends heavily on the availability and accessibility of related real-world data collected by the community. However, there is no central, structured, and holistic database of datasets and metadata associated with the SDGs, which prevents large-scale collaboration on these topics. In this paper, we present the SDG Data Catalog, a global open-source database indexing SDG-related datasets, associated metadata, and research networks. We describe the construction of this catalog, which relies on state-of-the-art natural language processing models with human supervision. The catalog breaks down data silos and helps sustainability researchers navigate the data sea to initiate effective collaborations.

Pictionary, the popular sketch-based guessing game, provides an opportunity to analyze shared goal cooperative game play in restricted communication settings. However, some players occasionally draw atypical sketch content. While such content is occasionally relevant in the game context, it sometimes represents a rule violation and impairs the game experience. To address such situations in a timely and scalable manner, we introduce DrawMon, a novel distributed framework for automatic detection of atypical sketch content in concurrently occurring Pictionary game sessions. We build specialized online interfaces to collect game session data and annotate atypical sketch content, resulting in AtyPict, the first ever atypical sketch content dataset. We use AtyPict to train CanvasNet, a deep neural atypical content detection network. We utilize CanvasNet as a core component of DrawMon. Our analysis of post deployment game session data indicates DrawMon's effectiveness for scalable monitoring and atypical sketch content detection. Beyond Pictionary, our contributions also serve as a design guide for customized atypical content response systems involving shared and interactive whiteboards. Code and datasets are available at https://drawm0n.github.io.

Quantizing large Neural Networks (NN) while maintaining the performance is highly desirable for resource-limited devices due to reduced memory and time complexity. It is usually formulated as a constrained optimization problem and optimized via a modified version of gradient descent. In this work, by interpreting the continuous parameters (unconstrained) as the dual of the quantized ones, we introduce a Mirror Descent (MD) framework for NN quantization. Specifically, we provide conditions on the projections (i.e., mapping from continuous to quantized ones) which would enable us to derive valid mirror maps and in turn the respective MD updates. Furthermore, we present a numerically stable implementation of MD that requires storing an additional set of auxiliary variables (unconstrained), and show that it is strikingly analogous to the Straight Through Estimator (STE) based method which is typically viewed as a “trick” to avoid vanishing gradients issue. Our experiments on CIFAR-10/100, TinyImageNet, and ImageNet classification datasets with VGG-16, ResNet-18, and MobileNetV2 architectures show that our MD variants yield state-of-the-art performance.

Grasping a simple object from the side is easy - unless the object is almost as big as the hand or space constraints require positioning the robot hand awkwardly with respect to the object. We show that humans - when faced with this challenge - adopt coordinated finger movements which enable them to successfully grasp objects even from these awkward poses. We also show that it is relatively straight forward to implement these strategies autonomously. Our human-studies approach asks participants to perform grasping task by either “puppetteering” a robotic manipulator that is identical (geometrically and kinematically) to a popular underactuated robotic manipulator (the Barrett hand), or using sliders to control the original Barrett hand. Unlike previous studies, this enables us to directly capture and compare human manipulation strategies with robotic ones. Our observation is that, while humans employ underactuation, how they use it is fundamentally different (and more effective) than that found in existing hardware.

Recommender Systems are an integral part of music sharing platforms. Often the aim of these systems is to increase the time, the user spends on the platform and hence having a high commercial value. The systems which aim at increasing the average time a user spends on the platform often need to recommend songs which the user might want to listen to next at each point in time. This is different from recommendation systems which try to predict the item which might be of interest to the user at some point in the user lifetime but not necessarily in the very near future. Prediction of next song the user might like requires some kind of modeling of the user interests at the given point of time. Attentive neural networks have been exploiting the sequence in which the items were selected by the user to model the implicit short-term interests of the user for the task of next item prediction, however we feel that features of the songs occurring in the sequence could also convey some important information about the short-term user interest which only the items cannot. In this direction we propose a novel attentive neural architecture which in addition to the sequence of items selected by the user, uses the features of these items to better learn the user short-term preferences and recommend next song to the user.

In this paper, we consider the problem of Robust Matrix Completion (RMC) where the goal is to recover a low-rank matrix by observing a small number of its entries out of which a few can be arbitrarily corrupted. We propose a simple projected gradient descent-based method to estimate the low-rank matrix that alternately performs a projected gradient descent step and cleans up a few of the corrupted entries using hard-thresholding. Our algorithm solves RMC using nearly optimal number of observations while tolerating a nearly optimal number of corruptions. Our result also implies significant improvement over the existing time complexity bounds for the low-rank matrix completion problem. Finally, an application of our result to the robust PCA problem (low-rank+sparse matrix separation) leads to nearly linear time (in matrix dimensions) algorithm for the same; existing state-of-the-art methods require quadratic time. Our empirical results corroborate our theoretical results and show that even for moderate sized problems, our method for robust PCA is an order of magnitude faster than the existing methods.

Summary Lexical states in JavaCC provide a powerful mechanism to scan regular expressions in a context sensitive manner. But lexical states also make it hard to reason about the correctness of the grammar. We first categorize the related correctness issues into two classes: errors and warnings. We then extend the traditional context sensitive and a context insensitive analysis to identify errors and warnings in context‐free grammars. We have implemented these analyses as a standalone tool (LSA), the first of its kind, to identify errors and warnings in JavaCC grammars. The LSA tool outputs a graph that depicts the grammar and the error transitions. Importantly, it can also generate counter example strings that can be used to establish the errors. We have used LSA to analyze a host of open‐source JavaCC grammar files to good effect. Copyright © 2015 John Wiley & Sons, Ltd.