Search

Oracle character recognition-an analysis of ancient Chinese inscriptions found on oracle bones-has become a pivotal field intersecting archaeology, paleography, and historical cultural studies. Traditional methods of oracle character recognition have relied heavily on manual interpretation by experts, which is not only labor-intensive but also limits broader accessibility to the general public. With recent breakthroughs in pattern recognition and deep learning, there is a growing movement towards the automation of oracle character recognition (OrCR), showing considerable promise in tackling the challenges inherent to these ancient scripts. However, a comprehensive understanding of OrCR still remains elusive. Therefore, this paper presents a systematic and structured survey of the current landscape of OrCR research. We commence by identifying and analyzing the key challenges of OrCR. Then, we provide an overview of the primary benchmark datasets and digital resources available for OrCR. A review of contemporary research methodologies follows, in which their respective efficacies, limitations, and applicability to the complex nature of oracle characters are critically highlighted and examined. Additionally, our review extends to ancillary tasks associated with OrCR across diverse disciplines, providing a broad-spectrum analysis of its applications. We conclude with a forward-looking perspective, proposing potential avenues for future investigations that could yield significant advancements in the field.

Automatic summarization has consistently attracted attention, due to its versatility and wide application in various downstream tasks. Despite its popularity, we find that annotation efforts have largely been disjointed, and have lacked common terminology. Consequently, it is challenging to discover existing resources or identify coherent research directions. To address this, we survey a large body of work spanning 133 datasets in over 100 languages, creating a novel ontology covering sample properties, collection methods and distribution. With this ontology we make key observations, including the lack in accessible high-quality datasets for low-resource languages, and the field's over-reliance on the news domain and on automatically collected distant supervision. Finally, we make available a web interface that allows users to interact and explore our ontology and dataset collection, as well as a template for a summarization data card, which can be used to streamline future research into a more coherent body of work.

Misinformation, defined as false or inaccurate information, can result in significant societal harm when it is spread with malicious or even innocuous intent. The rapid online information exchange necessitates advanced detection mechanisms to mitigate misinformation-induced harm. Existing research, however, has predominantly focused on assessing veracity, overlooking the legal implications and social consequences of misinformation. In this work, we take a novel angle to consolidate the definition of misinformation detection using legal issues as a measurement of societal ramifications, aiming to bring interdisciplinary efforts to tackle misinformation and its consequence. We introduce a new task: Misinformation with Legal Consequence (MisLC), which leverages definitions from a wide range of legal domains covering 4 broader legal topics and 11 fine-grained legal issues, including hate speech, election laws, and privacy regulations. For this task, we advocate a two-step dataset curation approach that utilizes crowd-sourced checkworthiness and expert evaluations of misinformation. We provide insights about the MisLC task through empirical evidence, from the problem definition to experiments and expert involvement. While the latest large language models and retrieval-augmented generation are effective baselines for the task, we find they are still far from replicating expert performance., Comment: 8.5 pages of main body, 20 pages total; Accepted to Findings of EMNLP 2024

Generative AI models, such as the GPT and Llama series, have significant potential to assist laypeople in answering legal questions. However, little prior work focuses on the data sourcing, inference, and evaluation of these models in the context of laypersons. To this end, we propose a human-centric legal NLP pipeline, covering data sourcing, inference, and evaluation. We introduce and release a dataset, LegalQA, with real and specific legal questions spanning from employment law to criminal law, corresponding answers written by legal experts, and citations for each answer. We develop an automatic evaluation protocol for this dataset, then show that retrieval-augmented generation from only 850 citations in the train set can match or outperform internet-wide retrieval, despite containing 9 orders of magnitude less data. Finally, we propose future directions for open-sourced efforts, which fall behind closed-sourced models., Comment: Accepted into GenLaw '24 (ICML 2024 workshop)

This study focuses on relay transport carriers (RTCs) that contract with hub providers to lease hub capacity and employ relay transportation via hubs. It enables long-haul freight shipments to be transported by multiple short-haul drivers commuting between fixed-base hubs, promoting a driver-friendly approach. Inspired by Physical Internet, our paper addresses the multi-period capacity planning of logistic hubs within relay networks, accounting for uncertainty in demand and travel times. We model the problem as a two-stage stochastic optimization to determine the dynamic logistic hub throughput capacities for each planning period, ensuring the fulfillment of logistic demand while simultaneously minimizing both hub and transportation costs. This optimization problem falls within the NP-hard complexity class. To alleviate the inherent challenges in solving this problem, we employ a scenario reduction algorithm based on the fast forward selection (FFS) method to reduce computational effort while preserving approximation quality. Experiments with an automotive-delivery RTC in the Southeastern US demonstrate that our capacity planning model enables RTCs to proactively respond to dynamic circumstances, curtail avoidable expenditures, and enhance overall logistical efficiency.

Clinical adoption of deep learning models has been hindered, in part, because the black-box nature of neural networks leads to concerns regarding their trustworthiness and reliability. These concerns are particularly relevant in the field of neuroimaging due to the complex brain phenotypes and inter-subject heterogeneity often encountered. The challenge can be addressed by interpretable deep learning (iDL) methods that enable the visualisation and interpretation of the inner workings of deep learning models. This study systematically reviewed the literature on neuroimaging applications of iDL methods and critically analysed how iDL explanation properties were evaluated. Seventy-five studies were included, and ten categories of iDL methods were identified. We also reviewed five properties of iDL explanations that were analysed in the included studies: biological validity, robustness, continuity, selectivity, and downstream task performance. We found that the most popular iDL approaches used in the literature may be sub-optimal for neuroimaging data, and we discussed possible future directions for the field.

In this paper, we investigate the challenging framework of Byzantine-robust training in distributed machine learning (ML) systems, focusing on enhancing both efficiency and practicality. As distributed ML systems become integral for complex ML tasks, ensuring resilience against Byzantine failures-where workers may contribute incorrect updates due to malice or error-gains paramount importance. Our first contribution is the introduction of the Centered Trimmed Meta Aggregator (CTMA), an efficient meta-aggregator that upgrades baseline aggregators to optimal performance levels, while requiring low computational demands. Additionally, we propose harnessing a recently developed gradient estimation technique based on a double-momentum strategy within the Byzantine context. Our paper highlights its theoretical and practical advantages for Byzantine-robust training, especially in simplifying the tuning process and reducing the reliance on numerous hyperparameters. The effectiveness of this technique is supported by theoretical insights within the stochastic convex optimization (SCO) framework and corroborated by empirical evidence.

Despite the remarkable success of deep learning in medical imaging analysis, medical image segmentation remains challenging due to the scarcity of high-quality labeled images for supervision. Further, the significant domain gap between natural and medical images in general and ultrasound images in particular hinders fine-tuning models trained on natural images to the task at hand. In this work, we address the performance degradation of segmentation models in low-data regimes and propose a prompt-less segmentation method harnessing the ability of segmentation foundation models to segment abstract shapes. We do that via our novel prompt point generation algorithm which uses coarse semantic segmentation masks as input and a zero-shot prompt-able foundation model as an optimization target. We demonstrate our method on a segmentation findings task (pathologic anomalies) in ultrasound images. Our method's advantages are brought to light in varying degrees of low-data regime experiments on a small-scale musculoskeletal ultrasound images dataset, yielding a larger performance gain as the training set size decreases.

This study evaluates the performance of general-purpose AI, like ChatGPT, in legal question-answering tasks, highlighting significant risks to legal professionals and clients. It suggests leveraging foundational models enhanced by domain-specific knowledge to overcome these issues. The paper advocates for creating open-source legal AI systems to improve accuracy, transparency, and narrative diversity, addressing general AI's shortcomings in legal contexts.

We consider a two-person network inspection game, in which a defender positions a limited number of detectors to detect multiple attacks caused by an attacker. We assume that detection is imperfect, and each detector location is associated with a probability of detecting attacks within its set of monitored network components. The objective of the defender (resp. attacker) is to minimize (resp. maximize) the expected number of undetected attacks. To compute Nash Equilibria (NE) for this large-scale zero-sum game, we formulate a linear program with a small number of constraints, which we solve via column generation. We provide an exact mixed-integer program for the pricing problem, which entails computing a defender's pure best response, and leverage its supermodular structure to derive two efficient approaches to obtain approximate NE with theoretical guarantees: A column generation and a multiplicative weights update (MWU) algorithm with approximate best responses. To address the computational challenges posed by combinatorial attacker strategies, each iteration of our MWU algorithm requires computing a projection under the unnormalized relative entropy. We provide a closed-form solution and a linear-time algorithm for the projection problem. Our computational results in real-world gas distribution networks illustrate the performance and scalability of our solution approaches., Comment: 50 pages, 6 figures

The balanced double star on $2n+2$ vertices, denoted $S_{n,n}$, is the tree obtained by joining the centers of two disjoint stars each having $n$ leaves. Let $R_r(G)$ be the smallest integer $N$ such that in every $r$-coloring of the edges of $K_N$ there is a monochromatic copy of $G$, and let $R_r^{\mathrm{bip}}(G)$ be the smallest integer $N$ such that in every $r$-coloring of the edges of $K_{N,N}$ there is a monochromatic copy of $G$. It is known that $R_2(S_{n,n})=3n+2$ and $R_2^{\mathrm{bip}}(S_{n,n})=2n+1$ \cite{HJ}, but very little is known about $R_r(S_{n,n})$ and $R^{\mathrm{bip}}_r(S_{n,n})$ when $r\geq 3$ (other than the bounds which follow from considerations on the number of edges in the majority color class). In this paper we prove the following for all $n\geq 1$ (where the lower bounds are adapted from existing examples): \[(r-1)2n+1\leq R_r(S_{n,n})\leq (r-\frac{1}{2})(2n+2)-1,\]and \[(2r-4)n+1\leq R^{\mathrm{bip}}_r(S_{n,n})\leq (2r-3+\frac{2}{r}+O(\frac{1}{r^2}))n.\] These bounds are similar to the best known bounds on $R_r(P_{2n+2})$ and $R_r^{\mathrm{bip}}(P_{2n+2})$, where $P_{2n+2}$ is a path on $2n+2$ vertices (which is also a balanced tree). We also give an example which improves the lower bound on $R^{\mathrm{bip}}_r(S_{n,n})$ when $r=3$ and $r=5$.

We study a two-player discounted zero-sum stochastic game model for dynamic operational planning in military campaigns. At each stage, the players manage multiple commanders who order military actions on objectives that have an open line of control. When a battle over the control of an objective occurs, its stochastic outcome depends on the actions and the enabling support provided by the control of other objectives. Each player aims to maximize the cumulative number of objectives they control, weighted by their criticality. To solve this large-scale stochastic game, we derive properties of its Markov perfect equilibria by leveraging the logistics and military operational command and control structure. We show the consequential isotonicity of the optimal value function with respect to the partially ordered state space, which in turn leads to a significant reduction of the state and action spaces. We also accelerate Shapley's value iteration algorithm by eliminating dominated actions and investigating pure equilibria of the matrix game solved at each iteration. We demonstrate the computational value of our equilibrium results on a case study that reflects representative operational-level military campaigns with geopolitical implications. Our analysis reveals a complex interplay between the game's parameters and dynamics in equilibrium, resulting in new military insights for campaign analysts.

Hyperconnected relay transportation enables using a relay system of short-haul drivers to deliver long-haul shipments collectively, which helps address root causes of trucker shortage issues by transforming working conditions with potentials of daily returning home, accessing consistent schedules, and facilitating load matching. This paper investigates hyperconnected relay transportation as a sustainable solution to trucker shortage issues through a logistics platform. We propose a two-stage programming model to optimize consistent working schedules for short-haul drivers while minimizing transportation costs. The first stage involves opening services and contracting truckers under demand uncertainty, where each service has a service route and approximate service schedules adhering to USA federal short-haul hour-of-service regulations. The second stage assigns hauling capacities to open services and manages commodity shipping or outsourcing given the demand realization. We extend the model formulation to account for various operational patterns (e.g., freight loading and unloading or hauler swapping) and schedule consistency requirements (e.g., weekly or daily consistency). A scenario-based approach is employed to solve the model for a case study of automotive delivery in the Southeast USA region. The experimental results validate the proposed approach, and further explore the impact of stochastic demands, operational patterns, consistent schedules, and hauling capacities on hyperconnected service network design. This research aims to offer practical guidance to practitioners in the trucking industry.

Cortical surface analysis has gained increased prominence, given its potential implications for neurological and developmental disorders. Traditional vision diffusion models, while effective in generating natural images, present limitations in capturing intricate development patterns in neuroimaging due to limited datasets. This is particularly true for generating cortical surfaces where individual variability in cortical morphology is high, leading to an urgent need for better methods to model brain development and diverse variability inherent across different individuals. In this work, we proposed a novel diffusion model for the generation of cortical surface metrics, using modified surface vision transformers as the principal architecture. We validate our method in the developing Human Connectome Project (dHCP), the results suggest our model demonstrates superior performance in capturing the intricate details of evolving cortical surfaces. Furthermore, our model can generate high-quality realistic samples of cortical surfaces conditioned on postmenstrual age(PMA) at scan., Comment: 4 pages

External control arms (ECA) can inform the early clinical development of experimental drugs and provide efficacy evidence for regulatory approval. However, the main challenge in implementing ECA lies in accessing real-world or historical clinical trials data. Indeed, regulations protecting patients' rights by strictly controlling data processing make pooling data from multiple sources in a central server often difficult. To address these limitations, we develop a new method, 'FedECA' that leverages federated learning (FL) to enable inverse probability of treatment weighting (IPTW) for time-to-event outcomes on separate cohorts without needing to pool data. To showcase the potential of FedECA, we apply it in different settings of increasing complexity culminating with a real-world use-case in which FedECA provides evidence for a differential effect between two drugs that would have otherwise gone unnoticed. By sharing our code, we hope FedECA will foster the creation of federated research networks and thus accelerate drug development., Comment: code available at: https://github.com/owkin/fedeca, bug in SMD computation present in v1 and v2 has been fixed, many experiments on real data have been added + fix in YODA experiments using imputed data instead of raw data (v3->v4) as well as affiliations fix + more precise wording for acknowledgments

Histopathology image analysis plays a crucial role in cancer diagnosis. However, training a clinically applicable segmentation algorithm requires pathologists to engage in labour-intensive labelling. In contrast, weakly supervised learning methods, which only require coarse-grained labels at the image level, can significantly reduce the labeling efforts. Unfortunately, while these methods perform reasonably well in slide-level prediction, their ability to locate cancerous regions, which is essential for many clinical applications, remains unsatisfactory. Previously, we proposed CAMEL, which achieves comparable results to those of fully supervised baselines in pixel-level segmentation. However, CAMEL requires 1,280x1,280 image-level binary annotations for positive WSIs. Here, we present CAMEL2, by introducing a threshold of the cancerous ratio for positive bags, it allows us to better utilize the information, consequently enabling us to scale up the image-level setting from 1,280x1,280 to 5,120x5,120 while maintaining the accuracy. Our results with various datasets, demonstrate that CAMEL2, with the help of 5,120x5,120 image-level binary annotations, which are easy to annotate, achieves comparable performance to that of a fully supervised baseline in both instance- and slide-level classifications., Comment: 41 pages, 13 figures, published in Advanced Intelligent Systems