arXiv subjects

Wuhui Chen

Publications and source records attributed to Wuhui Chen.

14 recordsLinked to original sources

arXiv2026

Observation-Grounded Self-Predictive Reinforcement Learning for Visual Continuous Control

Sample-efficient policy learning from pixels is a long-standing challenge in reinforcement learning (RL). Recent dynamics-based representation learning methods have significantly improved the sample efficiency of model-free visual RL by learning dynamics-aware representations through auxiliary prediction performed either in latent space (self-prediction) or observation space (observation prediction). However, state-of-the-art methods from both categories still struggle on challenging visual control tasks when training data is limited. We posit that relying on either predictive objective alone may be insufficient. In contrast, observation prediction grounds learned representations in observation-level dynamics, but does not directly regularize the temporal predictability of latent representations over extended horizons. In this paper, we propose Observation-Grounded Self-Predictive Representations (OG-SPR), a model-free visual RL algorithm for continuous control that learns representations that are both temporally predictive in latent space and grounded in observation-level dynamics. OG-SPR incorporates two core auxiliary objectives: multi-step latent self-prediction and next-observation prediction. We empirically show that directly imposing latent self-prediction on the shared representation may over-constrain it and does not necessarily improve performance. To address this issue, OG-SPR introduces two lightweight adapters for latent self-prediction, allowing the shared representation to benefit from temporally predictive signals without being forced to directly satisfy the self-prediction objective. Experiments on 28 visual control tasks from the DeepMind Control Suite show that OG-SPR improves aggregate performance over state-of-the-art self-predictive and observation-predictive RL methods, with particularly pronounced gains in challenging domains such as dog and humanoid.

cs.LG↗

arXiv2026

ProDVI: Programmatic Dynamics Priors for Value Network Initialization

Deep Reinforcement Learning (RL) is notoriously sample inefficient. One contributing factor is that RL agents are typically initialized from scratch, forcing them to acquire task-relevant knowledge through online interaction. Existing approaches obtain informative initializations through pre-collected datasets, high-fidelity simulators, or meta-learning over related tasks, but these prerequisites may be difficult to access or even unavailable. In this paper, we propose Programmatic Dynamics Priors for Value Network Initialization (ProDVI), a framework that leverages the commonsense and domain knowledge encoded in large language models to initialize RL agents without relying on these resources. Specifically, ProDVI prompts a code-generating language model to produce executable Python functions that encode coarse hypotheses about environment dynamics. These functions are then used to generate synthetic transitions. Based on these transitions, we construct an auxiliary dynamics prediction objective to pretrain the state-action encoder of the value network in an actor-critic framework. The learned representation provides dynamics-aware inductive biases before online RL begins. Importantly, the generated programs are used only for representation pretraining and are not required to faithfully simulate the target environment. While the generated programs may be inaccurate, their induced initialization can be corrected through online learning from real transitions and rewards. Experiments on OpenAI Gym and DeepMind Control Suite tasks show that ProDVI can effectively improve the sample efficiency of model-free RL algorithms.

cs.LG↗

arXiv2026

Hawk: Harnessing Hardware-Aware Knowledge for High-Performance NPU Kernel Generation

Developing high-performance kernels for Neural Processing Units (NPUs) is a critical industry bottleneck, requiring developers to manually navigate implicit hardware constraints and strict memory hierarchies. While large language models offer immense automation potential, they fail catastrophically on NPUs due to a fundamental lack of hardware-specific priors. Naively transplanting code snippets from similar NPU kernels may pass the compiler, but it consistently triggers runtime crashes and performance degradation by blindly violating underlying hardware constraints. To overcome this, we introduce Hawk, a training-free framework that harnesses hardware-aware knowledge through three core modules: (1) Run-Time Knowledge Synthesis Module, which employs a Triple-Part Executable Knowledge Representation to inherently couple the error context with executable semantics; (2) Bottleneck-Aware Knowledge Retrieval Module, which implements a 2D-Retrieval paradigm to project queries into orthogonal syntactic and hardware-aligned semantic spaces; and (3) Effect-Driven Knowledge Distillation Module, which leverages LLM-driven semantic arbitration to continuously distill the knowledge by pruning errors and consolidating redundancies based on the empirical execution feedback. Extensive evaluations on real-world NPU workloads demonstrate that Hawk elevates generation accuracy from 49.4% to 80.0%, while achieving up to a 2.2x execution speedup over state-of-the-art baselines.

cs.AI↗

arXiv2026

NASDAQ: Normalized Observation Space Dynamics-Augmented Q-Learning

Augmenting model-free reinforcement learning (RL) with representations learned through observation dynamics prediction (observation-predictive RL) can improve sample efficiency and performance, with minor modifications and limited additional computation. However, this approach still struggles in challenging tasks with low-dimensional observations. In this paper, we identify a key factor behind this problem: unbalanced reconstruction losses across observation dimensions, where dimensions with larger value ranges dominate the loss. This encourages the agent to neglect dimensions with relatively small ranges, leading to degraded performance. To address this issue, we propose a novel normalization method tailored to online RL, which normalizes low-dimensional observations and balances the resulting losses and gradients. Beyond balancing reconstruction losses, observation normalization enables dynamics prediction to be performed in a normalized observation space, thereby providing a unified treatment of low- and high-dimensional inputs (e.g., physical states and images). Building on this idea, we further introduce Normalized Observation Space Dynamics-Augmented Q-learning (NASDAQ), a framework for observation-predictive RL applicable across diverse domains. NASDAQ learns state-action representations by coupling value learning with two auxiliary tasks: short-term value prediction and next normalized observation prediction. Extensive experiments demonstrate that NASDAQ achieves competitive or superior performance compared with state-of-the-art model-based and self-predictive RL methods, while requiring significantly less training wall-time.

cs.LG↗

arXiv2026

DyMoE: Dynamic Expert Orchestration with Mixed-Precision Quantization for Efficient MoE Inference on Edge

Despite the computational efficiency of MoE models, the excessive memory footprint and I/O overhead inherent in multi-expert architectures pose formidable challenges for real-time inference on resource-constrained edge platforms. While existing static methods struggle with a rigid latency-accuracy trade-off, we observe that expert importance is highly skewed and depth-dependent. Motivated by these insights, we propose DyMoE, a dynamic mixed-precision quantization framework designed for high-performance edge inference. Leveraging insights into expert importance skewness and depth-dependent sensitivity, DyMoE introduces: (1) importance-aware prioritization to dynamically quantize experts at runtime; (2) depth-adaptive scheduling to preserve semantic integrity in critical layers; and (3) look-ahead prefetching to overlap I/O stalls. Experimental results on commercial edge hardware show that DyMoE reduces Time-to-First-Token (TTFT) by 3.44x-22.7x and up to a 14.58x speedup in Time-Per-Output-Token (TPOT) compared to state-of-the-art offloading baselines, enabling real-time, accuracy-preserving MoE inference on resource-constrained edge devices.

cs.LG↗

arXiv2025

Krul: Efficient State Restoration for Multi-turn Conversations with Dynamic Cross-layer KV Sharing

Efficient state restoration in multi-turn conversations with large language models (LLMs) remains a critical challenge, primarily due to the overhead of recomputing or loading full key-value (KV) caches for all historical tokens. To address this, existing approaches compress KV caches across adjacent layers with highly similar attention patterns. However, these methods often apply a fixed compression scheme across all conversations, selecting the same layer pairs for compression without considering conversation-specific attention dynamics. This static strategy overlooks variability in attention pattern similarity across different conversations, which can lead to noticeable accuracy degradation. We present Krul, a multi-turn LLM inference system that enables accurate and efficient KV cache restoration. Krul dynamically selects compression strategies based on attention similarity across layer pairs and uses a recomputation-loading pipeline to restore the KV cache. It introduces three key innovations: 1) a preemptive compression strategy selector to preserve critical context for future conversation turns and selects a customized strategy for the conversation; 2) a token-wise heterogeneous attention similarity estimator to mitigate the attention similarity computation and storage overhead during model generation; 3) a bubble-free restoration scheduler to reduce potential bubbles brought by the imbalance of recomputing and loading stream due to compressed KV caches. Empirical evaluations on real-world tasks demonstrate that Krul achieves a 1.5x-2.68x reduction in time-to-first-token (TTFT) and a 1.33x-2.35x reduction in KV cache storage compared to state-of-the-art methods without compromising generation quality.

cs.CL↗

arXiv2025

Fate: Fast Edge Inference of Mixture-of-Experts Models via Cross-Layer Gate

Large Language Models (LLMs) have demonstrated impressive performance across various tasks, and their application in edge scenarios has attracted significant attention. However, sparse-activated Mixture-of-Experts (MoE) models, which are well suited for edge scenarios, have received relatively little attention due to their high memory demands. Offload-based methods have been proposed to address this challenge, but they face difficulties with expert prediction. Inaccurate expert predictions can result in prolonged inference delays. To promote the application of MoE models in edge scenarios, we propose Fate, an offloading system designed for MoE models to enable efficient inference in resource-constrained environments. The key insight behind Fate is that gate inputs from adjacent layers can be effectively used for expert prefetching, achieving high prediction accuracy without additional GPU overhead. Furthermore, Fate employs a shallow-favoring expert caching strategy that increases the expert hit rate to 99\%. Additionally, Fate integrates tailored quantization strategies for cache optimization and IO efficiency. Experimental results show that, compared to Load on Demand and Expert Activation Path-based method, Fate achieves up to 4.5x and 1.9x speedups in prefill speed and up to 4.1x and 2.2x speedups in decoding speed, respectively, while maintaining inference quality. Moreover, Fate's performance improvements are scalable across different memory budgets.

cs.AI↗

arXiv2025

Klotski: Efficient Mixture-of-Expert Inference via Expert-Aware Multi-Batch Pipeline

Mixture of Experts (MoE), with its distinctive sparse structure, enables the scaling of language models up to trillions of parameters without significantly increasing computational costs. However, the substantial parameter size presents a challenge for inference, as the expansion in GPU memory cannot keep pace with the growth in parameters. Although offloading techniques utilise memory from the CPU and disk and parallelise the I/O and computation for efficiency, the computation for each expert in MoE models is often less than the I/O, resulting in numerous bubbles in the pipeline. Therefore, we propose Klotski, an efficient MoE inference engine that significantly reduces pipeline bubbles through a novel expert-aware multi-batch pipeline paradigm. The proposed paradigm uses batch processing to extend the computation time of the current layer to overlap with the loading time of the next layer. Although this idea has been effectively applied to dense models, more batches may activate more experts in the MoE, leading to longer loading times and more bubbles. Thus, unlike traditional approaches, we balance computation and I/O time and minimise bubbles by orchestrating their inference orders based on their heterogeneous computation and I/O requirements and activation patterns under different batch numbers. Moreover, to adapt to different hardware environments and models, we design a constraint-sensitive I/O-compute planner and a correlation-aware expert prefetcher for a schedule that minimises pipeline bubbles. Experimental results demonstrate that Klotski achieves a superior throughput-latency trade-off compared to state-of-the-art techniques, with throughput improvements of up to 85.12x.

cs.LG↗

arXiv2024

GriDB: Scaling Blockchain Database via Sharding and Off-Chain Cross-Shard Mechanism

Blockchain databases have attracted widespread attention but suffer from poor scalability due to underlying non-scalable blockchains. While blockchain sharding is necessary for a scalable blockchain database, it poses a new challenge named on-chain cross-shard database services. Each cross-shard database service (e.g., cross-shard queries or inter-shard load balancing) involves massive cross-shard data exchanges, while the existing cross-shard mechanisms need to process each cross-shard data exchange via the consensus of all nodes in the related shards (i.e., on-chain) to resist a Byzantine environment of blockchain, which eliminates sharding benefits. To tackle the challenge, this paper presents GriDB, the first scalable blockchain database, by designing a novel off-chain cross-shard mechanism for efficient cross-shard database services. Borrowing the idea of off-chain payments, GriDB delegates massive cross-shard data exchange to a few nodes, each of which is randomly picked from a different shard. Considering the Byzantine environment, the untrusted delegates cooperate to generate succinct proof for cross-shard data exchanges, while the consensus is only responsible for the low-cost proof verification. However, different from payments, the database services' verification has more requirements (e.g., completeness, correctness, freshness, and availability); thus, we introduce several new authenticated data structures (ADS). Particularly, we utilize consensus to extend the threat model and reduce the complexity of traditional accumulator-based ADS for verifiable cross-shard queries with a rich set of relational operators. Moreover, we study the necessity of inter-shard load balancing for a scalable blockchain database and design an off-chain and live approach for both efficiency and availability during balancing.

cs.DB↗

arXiv2024

Training and Serving System of Foundation Models: A Comprehensive Survey

Foundation models (e.g., ChatGPT, DALL-E, PengCheng Mind, PanGu-$Σ$) have demonstrated extraordinary performance in key technological areas, such as natural language processing and visual recognition, and have become the mainstream trend of artificial general intelligence. This has led more and more major technology giants to dedicate significant human and financial resources to actively develop their foundation model systems, which drives continuous growth of these models' parameters. As a result, the training and serving of these models have posed significant challenges, including substantial computing power, memory consumption, bandwidth demands, etc. Therefore, employing efficient training and serving strategies becomes particularly crucial. Many researchers have actively explored and proposed effective methods. So, a comprehensive survey of them is essential for system developers and researchers. This paper extensively explores the methods employed in training and serving foundation models from various perspectives. It provides a detailed categorization of these state-of-the-art methods, including finer aspects such as network, computing, and storage. Additionally, the paper summarizes the challenges and presents a perspective on the future development direction of foundation model systems. Through comprehensive discussion and analysis, it hopes to provide a solid theoretical basis and practical guidance for future research and applications, promoting continuous innovation and development in foundation model systems.

cs.AI↗

arXiv2023

Front-running Attack in Sharded Blockchains and Fair Cross-shard Consensus

Sharding is a prominent technique for scaling blockchains. By dividing the network into smaller components known as shards, a sharded blockchain can process transactions in parallel without introducing inconsistencies through the coordination of intra-shard and cross-shard consensus protocols. However, we observe a critical security issue with sharded systems: transaction ordering manipulations can occur when coordinating intra-shard and cross-shard consensus protocols, leaving the system vulnerable to attack. Specifically, we identify a novel security issue known as finalization fairness, which can be exploited through a front-running attack. This attack allows an attacker to manipulate the execution order of transactions, even if the victim's transaction has already been processed and added to the blockchain by a fair intra-shard consensus. To address the issue, we offer Haechi, a novel cross-shard protocol that is immune to front-running attacks. Haechi introduces an ordering phase between transaction processing and execution, ensuring that the execution order of transactions is the same as the processing order and achieving finalization fairness. To accommodate different consensus speeds among shards, Haechi incorporates a finalization fairness algorithm to achieve a globally fair order with minimal performance loss. By providing a global order, Haechi ensures strong consistency among shards, enabling better parallelism in handling conflicting transactions across shards. These features make Haechi a promising solution for supporting popular smart contracts in the real world. To evaluate Haechi's performance, we implemented the protocol using Tendermint and conducted extensive experiments on a geo-distributed AWS environment. Our results demonstrate that Haechi achieves finalization fairness with little performance sacrifice compared to existing cross-shard consensus protocols.

cs.CR↗

arXiv2023

Intelligence-Endogenous Management Platform for Computing and Network Convergence

Massive emerging applications are driving demand for the ubiquitous deployment of computing power today. This trend not only spurs the recent popularity of the \emph{Computing and Network Convergence} (CNC), but also introduces an urgent need for the intelligentization of a management platform to coordinate changing resources and tasks in the CNC. Therefore, in this article, we present the concept of an intelligence-endogenous management platform for CNCs called \emph{CNC brain} based on artificial intelligence technologies. It aims at efficiently and automatically matching the supply and demand with high heterogeneity in a CNC via four key building blocks, i.e., perception, scheduling, adaptation, and governance, throughout the CNC's life cycle. Their functionalities, goals, and challenges are presented. To examine the effectiveness of the proposed concept and framework, we also implement a prototype for the CNC brain based on a deep reinforcement learning technology. Also, it is evaluated on a CNC testbed that integrates two open-source and popular frameworks (OpenFaas and Kubernetes) and a real-world business dataset provided by Microsoft Azure. The evaluation results prove the proposed method's effectiveness in terms of resource utilization and performance. Finally, we highlight the future research directions of the CNC brain.

cs.AI↗

arXiv2023

Prophet: Conflict-Free Sharding Blockchain via Byzantine-Tolerant Deterministic Ordering

Sharding scales throughput by splitting blockchain nodes into parallel groups. However, different shards' independent and random scheduling for cross-shard transactions results in numerous conflicts and aborts, since cross-shard transactions from different shards may access the same account. A deterministic ordering can eliminate conflicts by determining a global order for transactions before processing, as proved in the database field. Unfortunately, due to the intertwining of the Byzantine environment and information isolation among shards, there is no trusted party able to predetermine such an order for cross-shard transactions. To tackle this challenge, this paper proposes Prophet, a conflict-free sharding blockchain based on Byzantine-tolerant deterministic ordering. It first depends on untrusted self-organizing coalitions of nodes from different shards to pre-execute cross-shard transactions for prerequisite information about ordering. It then determines a trusted global order based on stateless ordering and post-verification for pre-executed results, through shard cooperation. Following the order, the shards thus orderly execute and commit transactions without conflicts. Prophet orchestrates the pre-execution, ordering, and execution processes in the sharding consensus for minimal overhead. We rigorously prove the determinism and serializability of transactions under the Byzantine and sharded environment. An evaluation of our prototype shows that Prophet improves the throughput by $3.11\times$ and achieves nearly no aborts on 1 million Ethereum transactions compared with state-of-the-art sharding.

cs.CR↗

arXiv2019

PIRATE: A Blockchain-based Secure Framework of Distributed Machine Learning in 5G Networks

In the fifth-generation (5G) networks and the beyond, communication latency and network bandwidth will be no more bottleneck to mobile users. Thus, almost every mobile device can participate in the distributed learning. That is, the availability issue of distributed learning can be eliminated. However, the model safety will become a challenge. This is because the distributed learning system is prone to suffering from byzantine attacks during the stages of updating model parameters and aggregating gradients amongst multiple learning participants. Therefore, to provide the byzantine-resilience for distributed learning in 5G era, this article proposes a secure computing framework based on the sharding-technique of blockchain, namely PIRATE. A case-study shows how the proposed PIRATE contributes to the distributed learning. Finally, we also envision some open issues and challenges based on the proposed byzantine-resilient learning framework.

cs.DC↗