We introduce CHARM, the first benchmark for comprehensively and in-depth evaluating the commonsense reasoning ability of large language models (LLMs) in Chinese, which covers both globally known and Chinese-specific commonsense. We evaluated 7 English and 12 Chinese-oriented LLMs on CHARM, employing 5 representative prompt strategies for improving LLMs' reasoning ability, such as Chain-of-Thought. Our findings indicate that the LLM's language orientation and the task's domain influence the effectiveness of the prompt strategy, which enriches previous research findings. We built closely-interconnected reasoning and memorization tasks, and found that some LLMs struggle with memorizing Chinese commonsense, affecting their reasoning ability, while others show differences in reasoning despite similar memorization performance. We also evaluated the LLMs' memorization-independent reasoning abilities and analyzed the typical errors. Our study precisely identified the LLMs' strengths and weaknesses, providing the clear direction for optimization. It can also serve as a reference for studies in other fields. We will release CHARM at //github.com/opendatalab/CHARM .
相關內容
We warn against a common but incomplete understanding of empirical research in machine learning (ML) that leads to non-replicable results, makes findings unreliable, and threatens to undermine progress in the field. To overcome this alarming situation, we call for more awareness of the plurality of ways of gaining knowledge experimentally but also of some epistemic limitations. In particular, we argue most current empirical ML research is fashioned as confirmatory research while it should rather be considered exploratory.
Journals and conferences worry that peer reviews assisted by artificial intelligence (AI), in particular, large language models (LLMs), may negatively influence the validity and fairness of the peer-review system, a cornerstone of modern science. In this work, we address this concern with a quasi-experimental study of the prevalence and impact of AI-assisted peer reviews in the context of the 2024 International Conference on Learning Representations (ICLR), a large and prestigious machine-learning conference. Our contributions are threefold. Firstly, we obtain a lower bound for the prevalence of AI-assisted reviews at ICLR 2024 using the GPTZero LLM detector, estimating that at least $15.8\%$ of reviews were written with AI assistance. Secondly, we estimate the impact of AI-assisted reviews on submission scores. Considering pairs of reviews with different scores assigned to the same paper, we find that in $53.4\%$ of pairs the AI-assisted review scores higher than the human review ($p = 0.002$; relative difference in probability of scoring higher: $+14.4\%$ in favor of AI-assisted reviews). Thirdly, we assess the impact of receiving an AI-assisted peer review on submission acceptance. In a matched study, submissions near the acceptance threshold that received an AI-assisted peer review were $4.9$ percentage points ($p = 0.024$) more likely to be accepted than submissions that did not. Overall, we show that AI-assisted reviews are consequential to the peer-review process and offer a discussion on future implications of current trends
Mission critical communication (MCC) involves the exchange of information and data among emergency services, including the police, fire brigade, and other first responders, particularly during emergencies, disasters, or critical incidents. The widely-adopted TETRA (Terrestrial Trunked Radio)-based communication for mission critical services faces challenges including limited data capacity, coverage limitations, spectrum congestion, and security concerns. Therefore, as an alternative, mission critical communication over cellular networks (4G and 5G) has emerged. While cellular-based MCC enables features like real-time video streaming and high-speed data transmission, the involvement of network operators and application service providers in the MCC architecture raises privacy concerns for mission critical users and services. For instance, the disclosure of a policeman's location details to the network operator raises privacy concerns. To the best of our knowledge, no existing work considers the privacy issues in mission critical system with respect to 5G and upcoming technologies. Therefore, in this paper, we analyse the 3GPP standardised MCC architecture within the context of 5G core network concepts and assess the privacy implications for MC users, network entities, and MC servers. The privacy analysis adheres to the deployment strategies in the standard for MCC. Additionally, we explore emerging 6G technologies, such as off-network communications, joint communication and sensing, and non-3GPP communications, to identify privacy challenges in MCC architecture. Finally, we propose privacy controls to establish a next-generation privacy-preserving MCC architecture.
Retrieval-augmented Generation (RAG) systems have been actively studied and deployed across various industries to query on domain-specific knowledge base. However, evaluating these systems presents unique challenges due to the scarcity of domain-specific queries and corresponding ground truths, as well as a lack of systematic approaches to diagnosing the cause of failure cases -- whether they stem from knowledge deficits or issues related to system robustness. To address these challenges, we introduce GRAMMAR (GRounded And Modular Methodology for Assessment of RAG), an evaluation framework comprising two key elements: 1) a data generation process that leverages relational databases and LLMs to efficiently produce scalable query-answer pairs. This method facilitates the separation of query logic from linguistic variations for enhanced debugging capabilities; and 2) an evaluation framework that differentiates knowledge gaps from robustness and enables the identification of defective modules. Our empirical results underscore the limitations of current reference-free evaluation approaches and the reliability of GRAMMAR to accurately identify model vulnerabilities.
We introduce a new regression framework designed to deal with large-scale, complex data that lies around a low-dimensional manifold with noises. Our approach first constructs a graph representation, referred to as the skeleton, to capture the underlying geometric structure. We then define metrics on the skeleton graph and apply nonparametric regression techniques, along with feature transformations based on the graph, to estimate the regression function. We also discuss the limitations of some nonparametric regressors with respect to the general metric space such as the skeleton graph. The proposed regression framework suggests a novel way to deal with data with underlying geometric structures and provides additional advantages in handling the union of multiple manifolds, additive noises, and noisy observations. We provide statistical guarantees for the proposed method and demonstrate its effectiveness through simulations and real data examples.
This paper explores the intricate relationship between capitalism, racial injustice, and artificial intelligence (AI), arguing that AI acts as a contemporary vehicle for age-old forms of exploitation. By linking historical patterns of racial and economic oppression with current AI practices, this study illustrates how modern technology perpetuates and deepens societal inequalities. It specifically examines how AI is implicated in the exploitation of marginalized communities through underpaid labor in the gig economy, the perpetuation of biases in algorithmic decision-making, and the reinforcement of systemic barriers that prevent these groups from benefiting equitably from technological advances. Furthermore, the paper discusses the role of AI in extending and intensifying the social, economic, and psychological burdens faced by these communities, highlighting the problematic use of AI in surveillance, law enforcement, and mental health contexts. The analysis concludes with a call for transformative changes in how AI is developed and deployed. Advocating for a reevaluation of the values driving AI innovation, the paper promotes an approach that integrates social justice and equity into the core of technological design and policy. This shift is crucial for ensuring that AI serves as a tool for societal improvement, fostering empowerment and healing rather than deepening existing divides.
Existing committee-based Byzantine state machine replication (SMR) protocols, typically deployed in production blockchains, face a clear trade-off: (1) they either achieve linear communication cost in the happy path, but sacrifice liveness during periods of asynchrony, or (2) they are robust (progress with probability one) but pay quadratic communication cost. We believe this trade-off is unwarranted since existing linear protocols still have asymptotic quadratic cost in the worst case. We design Ditto, a Byzantine SMR protocol that enjoys the best of both worlds: optimal communication on and off the happy path (linear and quadratic, respectively) and progress guarantee under asynchrony and DDoS attacks. We achieve this by replacing the view-synchronization of partially synchronous protocols with an asynchronous fallback mechanism at no extra asymptotic cost. Specifically, we start from HotStuff, a state-of-the-art linear protocol, and gradually build Ditto. As a separate contribution and an intermediate step, we design a 2-chain version of HotStuff, Jolteon, which leverages a quadratic view-change mechanism to reduce the latency of the standard 3-chain HotStuff. We implement and experimentally evaluate all our systems. Notably, Jolteon's commit latency outperforms HotStuff by 200-300ms with varying system size. Additionally, Ditto adapts to the network and provides better performance than Jolteon under faulty conditions and better performance than VABA (a state-of-the-art asynchronous protocol) under faultless conditions. This proves our case that breaking the robustness-efficiency trade-off is in the realm of practicality.
The advent of large language models marks a revolutionary breakthrough in artificial intelligence. With the unprecedented scale of training and model parameters, the capability of large language models has been dramatically improved, leading to human-like performances in understanding, language synthesizing, and common-sense reasoning, etc. Such a major leap-forward in general AI capacity will change the pattern of how personalization is conducted. For one thing, it will reform the way of interaction between humans and personalization systems. Instead of being a passive medium of information filtering, large language models present the foundation for active user engagement. On top of such a new foundation, user requests can be proactively explored, and user's required information can be delivered in a natural and explainable way. For another thing, it will also considerably expand the scope of personalization, making it grow from the sole function of collecting personalized information to the compound function of providing personalized services. By leveraging large language models as general-purpose interface, the personalization systems may compile user requests into plans, calls the functions of external tools to execute the plans, and integrate the tools' outputs to complete the end-to-end personalization tasks. Today, large language models are still being developed, whereas the application in personalization is largely unexplored. Therefore, we consider it to be the right time to review the challenges in personalization and the opportunities to address them with LLMs. In particular, we dedicate this perspective paper to the discussion of the following aspects: the development and challenges for the existing personalization system, the newly emerged capabilities of large language models, and the potential ways of making use of large language models for personalization.
More than one hundred benchmarks have been developed to test the commonsense knowledge and commonsense reasoning abilities of artificial intelligence (AI) systems. However, these benchmarks are often flawed and many aspects of common sense remain untested. Consequently, we do not currently have any reliable way of measuring to what extent existing AI systems have achieved these abilities. This paper surveys the development and uses of AI commonsense benchmarks. We discuss the nature of common sense; the role of common sense in AI; the goals served by constructing commonsense benchmarks; and desirable features of commonsense benchmarks. We analyze the common flaws in benchmarks, and we argue that it is worthwhile to invest the work needed ensure that benchmark examples are consistently high quality. We survey the various methods of constructing commonsense benchmarks. We enumerate 139 commonsense benchmarks that have been developed: 102 text-based, 18 image-based, 12 video based, and 7 simulated physical environments. We discuss the gaps in the existing benchmarks and aspects of commonsense reasoning that are not addressed in any existing benchmark. We conclude with a number of recommendations for future development of commonsense AI benchmarks.
Small data challenges have emerged in many learning problems, since the success of deep neural networks often relies on the availability of a huge amount of labeled data that is expensive to collect. To address it, many efforts have been made on training complex models with small data in an unsupervised and semi-supervised fashion. In this paper, we will review the recent progresses on these two major categories of methods. A wide spectrum of small data models will be categorized in a big picture, where we will show how they interplay with each other to motivate explorations of new ideas. We will review the criteria of learning the transformation equivariant, disentangled, self-supervised and semi-supervised representations, which underpin the foundations of recent developments. Many instantiations of unsupervised and semi-supervised generative models have been developed on the basis of these criteria, greatly expanding the territory of existing autoencoders, generative adversarial nets (GANs) and other deep networks by exploring the distribution of unlabeled data for more powerful representations. While we focus on the unsupervised and semi-supervised methods, we will also provide a broader review of other emerging topics, from unsupervised and semi-supervised domain adaptation to the fundamental roles of transformation equivariance and invariance in training a wide spectrum of deep networks. It is impossible for us to write an exclusive encyclopedia to include all related works. Instead, we aim at exploring the main ideas, principles and methods in this area to reveal where we are heading on the journey towards addressing the small data challenges in this big data era.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191