By locally encoding raw data into intermediate features, collaborative inference enables end users to leverage powerful deep learning models without exposure of sensitive raw data to cloud servers. However, recent studies have revealed that these intermediate features may not sufficiently preserve privacy, as information can be leaked and raw data can be reconstructed via model inversion attacks (MIAs). Obfuscation-based methods, such as noise corruption, adversarial representation learning, and information filters, enhance the inversion robustness by obfuscating the task-irrelevant redundancy empirically. However, methods for quantifying such redundancy remain elusive, and the explicit mathematical relation between this redundancy minimization and inversion robustness enhancement has not yet been established. To address that, this work first theoretically proves that the conditional entropy of inputs given intermediate features provides a guaranteed lower bound on the reconstruction mean square error (MSE) under any MIA. Then, we derive a differentiable and solvable measure for bounding this conditional entropy based on the Gaussian mixture estimation and propose a conditional entropy maximization (CEM) algorithm to enhance the inversion robustness. Experimental results on four datasets demonstrate the effectiveness and adaptability of our proposed CEM; without compromising feature utility and computing efficiency, plugging the proposed CEM into obfuscation-based defense mechanisms consistently boosts their inversion robustness, achieving average gains ranging from 12.9\% to 48.2\%. Code is available at \href{https://github.com/xiasong0501/CEM}{https://github.com/xiasong0501/CEM}.
Learning informative representations of phylogenetic tree structures is essential for analyzing evolutionary relationships. Classical distance-based methods have been widely used to project phylogenetic trees into Euclidean space, but they are often sensitive to the choice of distance metric and may lack sufficient resolution. In this paper, we introduce phylogenetic variational autoencoders (PhyloVAEs), an unsupervised learning framework designed for representation learning and generative modeling of tree topologies. Leveraging an efficient encoding mechanism inspired by autoregressive tree topology generation, we develop a deep latent-variable generative model that facilitates fast, parallelized topology generation. PhyloVAE combines this generative model with a collaborative inference model based on learnable topological features, allowing for high-resolution representations of phylogenetic tree samples. Extensive experiments demonstrate PhyloVAE's robust representation learning capabilities and fast generation of phylogenetic tree topologies.
Large language models have achieved remarkable success in various tasks but suffer from high computational costs during inference, limiting their deployment in resource-constrained applications. To address this issue, we propose a novel Collaborative Inference with Token-lEvel Routing (CITER) framework that enables efficient collaboration between small and large language models (SLMs \& LLMs) through a token-level routing strategy. Specifically, CITER routes non-critical tokens to an SLM for efficiency and routes critical tokens to an LLM for generalization quality. We formulate router training as a policy optimization, where the router receives rewards based on both the quality of predictions and the inference costs of generation. This allows the router to learn to predict token-level routing scores and make routing decisions based on both the current token and the future impact of its decisions. To further accelerate the reward evaluation process, we introduce a shortcut which significantly reduces the costs of the reward estimation and improving the practicality of our approach. Extensive experiments on five benchmark datasets demonstrate that CITER reduces the inference costs while preserving high-quality generation, offering a promising solution for real-time and resource-constrained applications. Our data and code are available at https://github.com/aiming-lab/CITER.
Mixture-of-Agents (MoA) has recently been proposed as a method to enhance performance of large language models (LLMs), enabling multiple individual LLMs to work together for collaborative inference. This collaborative approach results in improved responses to user prompts compared to relying on a single LLM. In this paper, we consider such an MoA architecture in a distributed setting, where LLMs operate on individual edge devices, each uniquely associated with a user and equipped with its own distributed computing power. These devices exchange information using decentralized gossip algorithms, allowing different device nodes to talk without the supervision of a centralized server. In the considered setup, different users have their own LLM models to address user prompts. Additionally, the devices gossip either their own user-specific prompts or augmented prompts to generate more refined answers to certain queries. User prompts are temporarily stored in the device queues when their corresponding LLMs are busy. Given the memory limitations of edge devices, it is crucial to ensure that the average queue sizes in the system remain bounded. In this paper, we address this by theoretically calculating the queuing stability conditions for the device queues under reasonable assumptions, which we validate experimentally as well. Further, we demonstrate through experiments, leveraging open-source LLMs for the implementation of distributed MoA, that certain MoA configurations produce higher-quality responses compared to others, as evaluated on AlpacaEval 2.0 benchmark. The implementation is available at: https://github.com/purbeshmitra/distributed_moa.
Open-Vocabulary 3D object affordance grounding aims to anticipate ``action possibilities'' regions on 3D objects with arbitrary instructions, which is crucial for robots to generically perceive real scenarios and respond to operational changes. Existing methods focus on combining images or languages that depict interactions with 3D geometries to introduce external interaction priors. However, they are still vulnerable to a limited semantic space by failing to leverage implied invariant geometries and potential interaction intentions. Normally, humans address complex tasks through multi-step reasoning and respond to diverse situations by leveraging associative and analogical thinking. In light of this, we propose GREAT (GeometRy-intEntion collAboraTive inference) for Open-Vocabulary 3D Object Affordance Grounding, a novel framework that mines the object invariant geometry attributes and performs analogically reason in potential interaction scenarios to form affordance knowledge, fully combining the knowledge with both geometries and visual contents to ground 3D object affordance. Besides, we introduce the Point Image Affordance Dataset v2 (PIADv2), the largest 3D object affordance dataset at present to support the task. Extensive experiments demonstrate the effectiveness and superiority of GREAT. The code and dataset are available at https://yawen-shao.github.io/GREAT/.
We consider collaborative inference at the wireless edge, where each client's model is trained independently on its local dataset. Clients are queried in parallel to make an accurate decision collaboratively. In addition to maximizing the inference accuracy, we also want to ensure the privacy of local models. To this end, we leverage the superposition property of the multiple access channel to implement bandwidth-efficient multi-user inference methods. We propose different methods for ensemble and multi-view classification that exploit over-the-air computation (OAC). We show that these schemes perform better than their orthogonal counterparts with statistically significant differences while using fewer resources and providing privacy guarantees. We also provide experimental results verifying the benefits of the proposed OAC approach to multi-user inference, and perform an ablation study to demonstrate the effectiveness of our design choices. We share the source code of the framework publicly on Github to facilitate further research and reproducibility.
Open-Vocabulary Detection (OVD) is the task of detecting all interesting objects in a given scene without predefined object classes. Extensive work has been done to deal with the OVD for 2D RGB images, but the exploration of 3D OVD is still limited. Intuitively, lidar point clouds provide 3D information, both object level and scene level, to generate trustful detection results. However, previous lidar-based OVD methods only focus on the usage of object-level features, ignoring the essence of scene-level information. In this paper, we propose a Global-Local Collaborative Scheme (GLIS) for the lidar-based OVD task, which contains a local branch to generate object-level detection result and a global branch to obtain scene-level global feature. With the global-local information, a Large Language Model (LLM) is applied for chain-of-thought inference, and the detection result can be refined accordingly. We further propose Reflected Pseudo Labels Generation (RPLG) to generate high-quality pseudo labels for supervision and Background-Aware Object Localization (BAOL) to select precise object proposals. Extensive experiments on ScanNetV2 and SUN RGB-D demonstrate the superiority of our methods. Code is released at https://github.com/GradiusTwinbee/GLIS.
In this paper, we introduce LiveMind, a novel low-latency inference framework for large language model (LLM) inference which enables LLMs to perform inferences with incomplete user input. By reallocating computational processes to the input phase, a substantial reduction in latency is achieved, thereby significantly enhancing the interactive experience for users of LLMs. The framework adeptly manages the visibility of the streaming input to the model, allowing it to infer from incomplete user input or await additional content. Compared with traditional inference methods on complete user input, our approach demonstrates an average reduction in response latency of 84.0% on the MMLU dataset and 71.6% on the MMLU-Pro dataset, while maintaining comparable accuracy. Additionally, our framework facilitates collaborative inference and output across different models. By employing an large LLM for inference and a small LLM for output, we achieve an average 37% reduction in response latency, alongside a 4.30% improvement in accuracy on the MMLU-Pro dataset compared with the baseline. The proposed LiveMind framework advances the field of human-AI interaction by enabling more responsive and efficient communication between users and AI systems.
Generalist foundation models (GFMs) are renowned for their exceptional capability and flexibility in effectively generalizing across diverse tasks and modalities. In the field of medicine, while GFMs exhibit superior generalizability based on their extensive intrinsic knowledge as well as proficiency in instruction following and in-context learning, specialist models excel in precision due to their domain knowledge. In this work, for the first time, we explore the synergy between the GFM and specialist models, to enable precise medical image analysis on a broader scope. Specifically, we propose a cooperative framework, Generalist-Specialist Collaboration (GSCo), which consists of two stages, namely the construction of GFM and specialists, and collaborative inference on downstream tasks. In the construction stage, we develop MedDr, the largest open-source GFM tailored for medicine, showcasing exceptional instruction-following and in-context learning capabilities. Meanwhile, a series of lightweight specialists are crafted for downstream tasks with low computational cost. In the collaborative inference stage, we introduce two cooperative mechanisms, Mixture-of-Expert Diagnosis and Retrieval-Augmented Diagnosis, to harvest the generalist's in-context learning abilities alongside the specialists' domain expertise. For a comprehensive evaluation, we curate a large-scale benchmark featuring 28 datasets and about 250,000 images. Extensive results demonstrate that MedDr consistently outperforms state-of-the-art GFMs on downstream datasets. Furthermore, GSCo exceeds both GFMs and specialists across all out-of-domain disease diagnosis datasets. These findings indicate a significant paradigm shift in the application of GFMs, transitioning from separate models for specific tasks to a collaborative approach between GFMs and specialists, thereby advancing the frontiers of generalizable AI in medicine.
We propose a communication-efficient collaborative inference framework in the domain of edge inference, focusing on the efficient use of vision transformer (ViT) models. The partitioning strategy of conventional collaborative inference fails to reduce communication cost because of the inherent architecture of ViTs maintaining consistent layer dimensions across the entire transformer encoder. Therefore, instead of employing the partitioning strategy, our framework utilizes a lightweight ViT model on the edge device, with the server deploying a complicated ViT model. To enhance communication efficiency and achieve the classification accuracy of the server model, we propose two strategies: 1) attention-aware patch selection and 2) entropy-aware image transmission. Attention-aware patch selection leverages the attention scores generated by the edge device's transformer encoder to identify and select the image patches critical for classification. This strategy enables the edge device to transmit only the essential patches to the server, significantly improving communication efficiency. Entropy-aware image transmission uses min-entropy as a metric to accurately determine whether to depend on the lightweight model on the edge device or to request the inference from the server model. In our framework, the lightweight ViT model on the edge device acts as a semantic encoder, efficiently identifying and selecting the crucial image information required for the classification task. Our experiments demonstrate that the proposed collaborative inference framework can reduce communication overhead by 68% with only a minimal loss in accuracy compared to the server model on the ImageNet dataset.
Quantum processing units (QPUs) are currently exclusively available from cloud vendors. However, with recent advancements, hosting QPUs is soon possible everywhere. Existing work has yet to draw from research in edge computing to explore systems exploiting mobile QPUs, or how hybrid applications can benefit from distributed heterogeneous resources. Hence, this work presents an architecture for Quantum Computing in the edge-cloud continuum. We discuss the necessity, challenges, and solution approaches for extending existing work on classical edge computing to integrate QPUs. We describe how warm-starting allows defining workflows that exploit the hierarchical resources spread across the continuum. Then, we introduce a distributed inference engine with hybrid classical-quantum neural networks (QNNs) to aid system designers in accommodating applications with complex requirements that incur the highest degree of heterogeneity. We propose solutions focusing on classical layer partitioning and quantum circuit cutting to demonstrate the potential of utilizing classical and quantum computation across the continuum. To evaluate the importance and feasibility of our vision, we provide a proof of concept that exemplifies how extending a classical partition method to integrate quantum circuits can improve the solution quality. Specifically, we implement a split neural network with optional hybrid QNN predictors. Our results show that extending classical methods with QNNs is viable and promising for future work.
The cost efficiency of model inference is critical to real-world machine learning (ML) applications, especially for delay-sensitive tasks and resource-limited devices. A typical dilemma is: in order to provide complex intelligent services (e.g. smart city), we need inference results of multiple ML models, but the cost budget (e.g. GPU memory) is not enough to run all of them. In this work, we study underlying relationships among black-box ML models and propose a novel learning task: model linking, which aims to bridge the knowledge of different black-box models by learning mappings (dubbed model links) between their output spaces. We propose the design of model links which supports linking heterogeneous black-box ML models. Also, in order to address the distribution discrepancy challenge, we present adaptation and aggregation methods of model links. Based on our proposed model links, we developed a scheduling algorithm, named MLink. Through collaborative multi-model inference enabled by model links, MLink can improve the accuracy of obtained inference results under the cost budget. We evaluated MLink on a multi-modal dataset with seven different ML models and two real-world video analytics systems with six ML models and 3,264 hours of video. Experimental results show that our proposed model links can be effectively built among various black-box models. Under the budget of GPU memory, MLink can save 66.7% inference computations while preserving 94% inference accuracy, which outperforms multi-task learning, deep reinforcement learning-based scheduler and frame filtering baselines.
Many NLP tasks benefit from using large language models (LLMs) that often have more than 100 billion parameters. With the release of BLOOM-176B and OPT-175B, everyone can download pretrained models of this scale. Still, using these models requires high-end hardware unavailable to many researchers. In some cases, LLMs can be used more affordably via RAM offloading or hosted APIs. However, these techniques have innate limitations: offloading is too slow for interactive inference, while APIs are not flexible enough for research that requires access to weights, attention or logits. In this work, we propose Petals - a system for inference and fine-tuning of large models collaboratively by joining the resources of multiple parties. We demonstrate that this strategy outperforms offloading for very large models, running inference of BLOOM-176B on consumer GPUs with $\approx$ 1 step per second, which is enough for many interactive LLM applications. Unlike most inference APIs, Petals also natively exposes hidden states of served models, allowing to train and share custom model extensions based on efficient fine-tuning methods.
With the wide application of face recognition systems, there is rising concern that original face images could be exposed to malicious intents and consequently cause personal privacy breaches. This paper presents DuetFace, a novel privacy-preserving face recognition method that employs collaborative inference in the frequency domain. Starting from a counterintuitive discovery that face recognition can achieve surprisingly good performance with only visually indistinguishable high-frequency channels, this method designs a credible split of frequency channels by their cruciality for visualization and operates the server-side model on non-crucial channels. However, the model degrades in its attention to facial features due to the missing visual information. To compensate, the method introduces a plug-in interactive block to allow attention transfer from the client-side by producing a feature mask. The mask is further refined by deriving and overlaying a facial region of interest (ROI). Extensive experiments on multiple datasets validate the effectiveness of the proposed method in protecting face images from undesired visual inspection, reconstruction, and identification while maintaining high task availability and performance. Results show that the proposed method achieves a comparable recognition accuracy and computation cost to the unprotected ArcFace and outperforms the state-of-the-art privacy-preserving methods. The source code is available at https://github.com/Tencent/TFace/tree/master/recognition/tasks/duetface.
The success of deep neural networks (DNNs) is heavily dependent on computational resources. While DNNs are often employed on cloud servers, there is a growing need to operate DNNs on edge devices. Edge devices are typically limited in their computational resources, yet, often multiple edge devices are deployed in the same environment and can reliably communicate with each other. In this work we propose to facilitate the application of DNNs on the edge by allowing multiple users to collaborate during inference to improve their accuracy. Our mechanism, coined {\em edge ensembles}, is based on having diverse predictors at each device, which form an ensemble of models during inference. To mitigate the communication overhead, the users share quantized features, and we propose a method for aggregating multiple decisions into a single inference rule. We analyze the latency induced by edge ensembles, showing that its performance improvement comes at the cost of a minor additional delay under common assumptions on the communication network. Our experiments demonstrate that collaborative inference via edge ensembles equipped with compact DNNs substantially improves the accuracy over having each user infer locally, and can outperform using a single centralized DNN larger than all the networks in the ensemble together.
Recently, deploying deep neural network (DNN) models via collaborative inference, which splits a pre-trained model into two parts and executes them on user equipment (UE) and edge server respectively, becomes attractive. However, the large intermediate feature of DNN impedes flexible decoupling, and existing approaches either focus on the single UE scenario or simply define tasks considering the required CPU cycles, but ignore the indivisibility of a single DNN layer. In this paper, we study the multi-agent collaborative inference scenario, where a single edge server coordinates the inference of multiple UEs. Our goal is to achieve fast and energy-efficient inference for all UEs. To achieve this goal, we first design a lightweight autoencoder-based method to compress the large intermediate feature. Then we define tasks according to the inference overhead of DNNs and formulate the problem as a Markov decision process (MDP). Finally, we propose a multi-agent hybrid proximal policy optimization (MAHPPO) algorithm to solve the optimization problem with a hybrid action space. We conduct extensive experiments with different types of networks, and the results show that our method can reduce up to 56\% of inference latency and save up to 72\% of energy consumption.
When multiple parties that deal with private data aim for a collaborative prediction task such as medical image classification, they are often constrained by data protection regulations and lack of trust among collaborating parties. If done in a privacy-preserving manner, predictive analytics can benefit from the collective prediction capability of multiple parties holding complementary datasets on the same machine learning task. This paper presents PRICURE, a system that combines complementary strengths of secure multi-party computation (SMPC) and differential privacy (DP) to enable privacy-preserving collaborative prediction among multiple model owners. SMPC enables secret-sharing of private models and client inputs with non-colluding secure servers to compute predictions without leaking model parameters and inputs. DP masks true prediction results via noisy aggregation so as to deter a semi-honest client who may mount membership inference attacks. We evaluate PRICURE on neural networks across four datasets including benchmark medical image classification datasets. Our results suggest PRICURE guarantees privacy for tens of model owners and clients with acceptable accuracy loss. We also show that DP reduces membership inference attack exposure without hurting accuracy.