Foundation models demand advanced data processing for their vast, multimodal datasets.However, traditional frameworks struggle with the unique complexities of multimodal data.In response, we present Data-Juicer 2.0, a data processing system backed by 100+ data processing operators spanning text, image, video, and audio modalities, supporting more critical tasks including data analysis, synthesis, annotation, and foundation model post-training.With seamless compatibility and dedicated optimization for popular dataset hubs like Hugging Face and computing engines like Ray, it improves upon its predecessor in terms of usability, efficiency, and programmability.It features an easily accessible user interface layer that supports decoupled Python interactions, RESTful APIs, and conversational commands. Its new runtime layer offers adaptive execution across diverse scales and environments, abstracting away system complexities.Extensive empirical evaluations demonstrate Data-Juicer 2.0's remarkable performance and scalability, highlighting its capability to efficiently process TB-level data with 10k+ CPU cores. The system is publicly available and has been widely adopted in diverse research fields and real-world products such as Alibaba Cloud PAI. We actively maintain the system and share practical insights to foster research and applications of next-generation foundation models.

We propose two simple, principled and practical algorithms that enjoy provable scaling laws for the test-time compute of large language models (LLMs). The first one is a two-stage knockout-style algorithm: given an input problem, it first generates multiple candidate solutions, and then aggregate them via a knockout tournament for the final output. Assuming that the LLM can generate a correct solution with non-zero probability and do better than a random guess in comparing a pair of correct and incorrect solutions, we prove theoretically that the failure probability of this algorithm decays to zero exponentially or by a power law (depending on the specific way of scaling) as its test-time compute grows. The second one is a two-stage league-style algorithm, where each candidate is evaluated by its average win rate against multiple opponents, rather than eliminated upon loss to a single opponent. Under analogous but more robust assumptions, we prove that its failure probability also decays to zero exponentially with more test-time compute. Both algorithms require a black-box LLM and nothing else (e.g., no verifier or reward model) for a minimalistic implementation, which makes them appealing for practical applications and easy to adapt for different tasks. Through extensive experiments with diverse models and datasets, we validate the proposed theories and demonstrate the outstanding scaling properties of both algorithms.

We present EE-LLM, a framework for large-scale training and inference of early-exit large language models (LLMs). While recent works have shown preliminary evidence for the efficacy of early exiting in accelerating LLM inference, EE-LLM makes a foundational step towards scaling up early-exit LLMs by supporting their training and inference with massive 3D parallelism. Built upon Megatron-LM, EE-LLM implements a variety of algorithmic innovations and performance optimizations tailored to early exiting, including a lightweight method that facilitates backpropagation for the early-exit training objective with pipeline parallelism, techniques of leveraging idle resources in the original pipeline schedule for computation related to early-exit layers, and two approaches of early-exit inference that are compatible with KV caching for autoregressive generation. Our analytical and empirical study shows that EE-LLM achieves great training efficiency with negligible computational overhead compared to standard LLM training, as well as outstanding inference speedup without compromising output quality. To facilitate further research and adoption, we release EE-LLM at https://github.com/pan-x-c/EE-LLM.

Large language models (LLMs) have demonstrated great capabilities in various natural language understanding and generation tasks.These pre-trained LLMs can be further improved for specific downstream tasks by fine-tuning.However, the adoption of LLM in real-world applications can be hindered by privacy concerns and the resource-intensive nature of model training and fine-tuning.When multiple entities have similar interested tasks but cannot directly share their local data due to privacy regulations, federated learning (FL) is a mainstream solution to leverage the data of different entities.Besides avoiding direct data sharing, FL can also achieve rigorous data privacy protection, model intelligent property protection, and model customization via composition with different techniques.Despite the aforementioned advantages of FL, fine-tuning LLMs in FL settings still lacks adequate support from the existing frameworks and, therefore, faces challenges in optimizing the consumption of significant communication and computational resources, preparing various data for different tasks, and satisfying diverse information protection demands.

Federated Learning (FL) aims to train high-quality models in collaboration with distributed clients while not uploading their local data, which attracts increasing attention in both academia and industry. However, there is still a considerable gap between the flourishing FL research and real-world scenarios, mainly caused by the characteristics of heterogeneous devices and its scales. Most existing works conduct evaluations with homogeneous devices, which are mismatched with the diversity and variability of heterogeneous devices in real-world scenarios. Moreover, it is challenging to conduct research and development at scale with heterogeneous devices due to limited resources and complex software stacks. These two key factors are important yet underexplored in FL research as they directly impact the FL training dynamics and final performance, making the effectiveness and usability of FL algorithms unclear. To bridge the gap, in this paper, we propose an efficient and scalable prototyping system for real-world cross-device FL, FS-REAL. It supports heterogeneous device runtime, contains parallelism and robustness enhanced FL server, and provides implementations and extensibility for advanced FL utility features such as personalization, communication compression and asynchronous aggregation. To demonstrate the usability and efficiency of FS-REAL, we conduct extensive experiments with various device distributions, quantify and analyze the effect of the heterogeneous device and various scales, and further provide insights and open discussions about real-world FL scenarios. Our system is released to help to pave the way for further real-world FL research and broad applications involving diverse devices and scales.

Federated learning (FL) is a general distributed machine learning paradigm that provides solutions for tasks where data cannot be shared directly. Due to the difficulties in communication management and heterogeneity of distributed data and devices, initiating and using an FL algorithm for real-world cross-device scenarios requires significant repetitive effort but may not be transferable to similar projects. To reduce the effort required for developing and deploying FL algorithms, we present FS-Real, an open-source FL platform designed to address the need of a general and efficient infrastructure for real-world cross-device FL. In this paper, we introduce the key components of FS-Real and demonstrate that FS-Real has the following capabilities: 1) reducing the programming burden of FL algorithm development with plug-and-play and adaptable runtimes on Android and other Internet of Things (IoT) devices; 2) handling a large number of heterogeneous devices efficiently and robustly with our communication management components; 3) supporting a wide range of advanced FL algorithms with flexible configuration and extension; 4) alleviating the costs and efforts for deployment, evaluation, simulation, and performance optimization of FL algorithms with automatized tool kits.

The increasing concerns on data security limit the sharing of data distributedly stored at multiple data owners and impede the scale of spatial queries over big urban data. In response, data federation systems have emerged to perform secure queries across multiple data owners leveraging secure multi-party computation. However, existing systems are designed for relational data. They are highly inefficient on spatial queries and limited in usability. In this demonstration, we introduce Hu-Fu, the first data federation system for secure spatial queries with high efficiency and usability. Hu-Fu is designed from the perspectives of the query user and the data owner for high usability and decomposes a spatial query into as many plaintext operators and as few secure operators as possible for high efficiency. We demonstrate the deployment and usage of Hu-Fu via cross-company taxi-calling, a popular smart city application.

Data isolation has become an obstacle to scale up query processing over big data, since sharing raw data among data owners is often prohibitive due to security concerns. A promising solution is to perform secure queries over a federation of multiple data owners leveraging secure multi-party computation (SMC) techniques, as evidenced by recent federation work over relational data. However, existing solutions are highly inefficient on spatial queries due to excessive secure distance operations for query processing and their usage of general-purpose SMC libraries for secure operation implementation. In this paper, we propose Hu-Fu, the first system for efficient and secure spatial query processing on a data federation. The idea is to decompose the secure processing of a spatial query into as many plaintext operations and as few secure operations as possible, where fewer secure operators are involved and all secure operators are implemented dedicatedly. As a working system, Hu-Fu supports not only query input in native SQL, but also heterogeneous spatial databases ( e.g. , PostGIS, Simba, GeoMesa, and SpatialHadoop) at the backend. Extensive experiments show that Hu-Fu usually outperforms the state-of-the-arts in running time and communication cost while guaranteeing security.