Search

We focus on the problem of fusing two or more heterogeneous large language models (LLMs) to facilitate their complementary strengths. One of the challenges on model fusion is high computational load, i.e. to fine-tune or to align vocabularies via combinatorial optimization. To this end, we propose \emph{Cool-Fusion}, a simple yet effective approach that fuses the knowledge of heterogeneous source LLMs to leverage their complementary strengths. \emph{Cool-Fusion} is the first method that does not require any type of training like the ensemble approaches. But unlike ensemble methods, it is applicable to any set of source LLMs that have different vocabularies. The basic idea is to have each source LLM individually generate tokens until the tokens can be decoded into a text segment that ends at word boundaries common to all source LLMs. Then, the source LLMs jointly rerank the generated text segment and select the best one, which is the fused text generation in one step. Extensive experiments are conducted across a variety of benchmark datasets. On \emph{GSM8K}, \emph{Cool-Fusion} increases accuracy from three strong source LLMs by a significant 8\%-17.8\%.

Large Language Models (LLMs) have made tremendous strides in code generation, but existing research fails to account for the dynamic nature of software development, marked by frequent library updates. This gap significantly limits LLMs' deployment in realistic settings. In this paper, we propose two novel tasks aimed at bridging this gap: version-specific code completion (VSCC) and version-aware code migration (VACM). In conjunction, we introduce VersiCode, a comprehensive Python dataset specifically designed to evaluate LLMs on these two tasks, together with a novel evaluation metric, Critical Diff Check (CDC@1), which assesses code generation against evolving API requirements. We conduct an extensive evaluation on VersiCode, which reveals that version-controllable code generation is indeed a significant challenge, even for GPT-4o and other strong frontier models. We believe the novel tasks, dataset, and metric open up a new, important research direction that will further enhance LLMs' real-world applicability. The code and resources can be found at https://github.com/wutong8023/VersiCode.

Data heterogeneity across clients is one of the key challenges in Federated Learning (FL), which may slow down the global model convergence and even weaken global model performance. Most existing approaches tackle the heterogeneity by constraining local model updates through reference to global information provided by the server. This can alleviate the performance degradation on the aggregated global model. Different from existing methods, we focus the information exchange between clients, which could also enhance the effectiveness of local training and lead to generate a high-performance global model. Concretely, we propose a novel FL framework named FedCME by client matching and classifier exchanging. In FedCME, clients with large differences in data distribution will be matched in pairs, and then the corresponding pair of clients will exchange their classifiers at the stage of local training in an intermediate moment. Since the local data determines the local model training direction, our method can correct update direction of classifiers and effectively alleviate local update divergence. Besides, we propose feature alignment to enhance the training of the feature extractor. Experimental results demonstrate that FedCME performs better than FedAvg, FedProx, MOON and FedRS on popular federated learning benchmarks including FMNIST and CIFAR10, in the case where data are heterogeneous.

Federated learning (FL) is an emerging distributed machine learning paradigm that avoids data sharing among training nodes so as to protect data privacy. Under coordination of the FL server, each client conducts model training using its own computing resource and private data set. The global model can be created by aggregating the training results of clients. To cope with highly non-IID data distributions, personalized federated learning (PFL) has been proposed to improve overall performance by allowing each client to learn a personalized model. However, one major drawback of a personalized model is the loss of generalization. To achieve model personalization while maintaining generalization, in this paper, we propose a new approach, named PFL-MoE, which mixes outputs of the personalized model and global model via the MoE architecture. PFL-MoE is a generic approach and can be instantiated by integrating existing PFL algorithms. Particularly, we propose the PFL-MF algorithm which is an instance of PFL-MoE based on the freeze-base PFL algorithm. We further improve PFL-MF by enhancing the decision-making ability of MoE gating network and propose a variant algorithm PFL-MFE. We demonstrate the effectiveness of PFL-MoE by training the LeNet-5 and VGG-16 models on the Fashion-MNIST and CIFAR-10 datasets with non-IID partitions., Comment: 10 pages, 5 figures, conference

Temporary fork is a fundamental phenomenon in many blockchains with proof of work, and the analysis of temporary fork has recently drawn great attention. Different from existing efforts that focus on the blockchain system factors such as block size, network propagation delay or block generation speed, in this paper we explore a new key dimension of computing power from the miners' perspective. Specifically, we first propose a detailed mathematical model to characterize the impact of computing power competition of the mining pools on the temporary fork. We also derive closed-form formula of the probability of temporary fork and the expected mining reward of a mining pool. To reveal the long-term trends on the computing power distributions over the competing mining pools, we then develop an evolutionary game framework based on the temporary fork modeling and accordingly characterize the set of stable evolution equilibriums. Both extensive numerical simulations and realistic blockchain data based evaluation provide evidence to support our theoretical models and discoveries.

To mitigate the scalability problem of decentralized cryptocurrencies such as Bitcoin and Ethereum, the payment channel, which allows two parties to perform secure coin transfers without involving the blockchain, has been proposed. The payment channel increases the transaction throughput of two parties to a level that is only limited by their network bandwidth. Recent proposals focus on extending the two-party payment channel to the N-party payment hub. Unfortunately, none of them can achieve efficiency, flexibility in the absence of a trusted third-party. In this paper, we propose Garou, a secure N-party payment hub that allows multiple parties to perform secure off-chain coin transfers. Except in the case of disputes, participants within the payment hub can make concurrent and direct coin transfers with each other without the involvement of the blockchain or any third-party intermediaries. This allows Garou to achieve both high-performance and flexibility. Garou also guarantees that an honest party always maintains its balance security against strong adversarial capabilities. To demonstrate the feasibility of the Garou protocol, we develop a proof of concept prototype for the Ethereum network. Our evaluation results show that the maximum transaction throughput of Garou is 20 times higher than that of state-of-art payment hubs., Comment: 11 pages, 6 figures

The payment channel, which allows two parties to perform micropayments without involving the blockchain, has become a promising proposal to improve the scalability of decentralized ledgers such as Bitcoin and Ethereum. Payment channels have been extended to the payment network, through which users can utilize existing channels as intermediary links to route coins to others. However, routing payments through multiple channels bears nontrivial overheads. It requires every intermediary channel to lock a portion of its available capacity until the payment is settled. This may lead to deadlock in a concurrent situation. The intermediary nodes in a payment path may also charge fees for routing a payment. The longer the routing path, the more serious the above problems. In this paper, we design and develop a novel off-chain system to shorten the routing path for the payment network. In particular, we propose the channel hub, which is an extension of the payment hub, to allows transferring coins directly from one payment channel to another within the same hub. That is, the channel hub can be viewed as a shortcut device for the underlying payment network. We design a new protocol named Boros to perform secure off-chain cross-channel transfers through the channel hub. We not only present the security definition of the Boros protocol formally but also prove its security using the UC-framework. To demonstrate the feasibility of the Boros protocol, we develop a proof-of-concept prototype running on the Ethereum. Our evaluation shows that our system can effectively shorten the off-chain routing path.

In this article we propose a novel Device-to-Device (D2D) Crowd framework for 5G mobile edge computing, where a massive crowd of devices at the network edge leverage the network-assisted D2D collaboration for computation and communication resource sharing among each other. A key objective of this framework is to achieve energy-efficient collaborative task executions at network-edge for mobile users. Specifically, we first introduce the D2D Crowd system model in details, and then formulate the energy-efficient D2D Crowd task assignment problem by taking into account the necessary constraints. We next propose a graph matching based optimal task assignment policy, and further evaluate its performance through extensive numerical study, which shows a superior performance of more than 50% energy consumption reduction over the case of local task executions. Finally, we also discuss the directions of extending the D2D Crowd framework by taking into variety of application factors., Comment: Xu Chen, Lingjun Pu, Lin Gao, Weigang Wu, and Di Wu, "Exploiting Massive D2D Collaboration for Energy-Efficient Mobile Edge Computing," accepted by IEEE Wireless Communications, 2017