arXiv subjects

Qizhen Weng

Publications and source records attributed to Qizhen Weng.

18 recordsLinked to original sources

arXiv2026

Arachne: Orchestrating Cascades for Efficient Text-to-Video Model Training

The rising demand for AI-generated videos is fueled by advances in large-scale Text-to-Video (T2V) models, trained on extensive datasets of video clips spanning diverse resolutions and durations. To address this data heterogeneity, current training methods often use a bucketing strategy that groups samples into discrete buckets for efficiency. However, this approach struggles to scale with compute and data volumes under static parallelism schemes, such as data and sequence parallelism, leading to significant workload imbalances and hardware under-utilization. In this paper, we present Arachne, a novel training framework for efficient T2V model training at scale. Arachne decomposes the training process into fine-grained computational units, called \textit{cascades}, orchestrating their distributed execution and synchronization across the cluster through coordinated spatial and temporal optimization. Our comprehensive evaluation demonstrates that Arachne reduces iteration time by up to 65\% over leading frameworks, exhibiting a positive scaling trend where its performance advantages amplify as training scale grows.

cs.DC↗

arXiv2026

GN0: Toward a Unified Paradigm for Generation, Evaluation, and Policy Learning in Visual-Language Navigation

Embodied navigation connects intelligent agents with the physical world and is fundamental for general robotic intelligence. Limited availability and quality of navigation data have constrained Vision-and-Language Navigation (VLN) systems' generalization and long-horizon capabilities. To address this, we curate diverse 3D scenes and develop an automated pipeline for large-scale navigation data, resulting in the GN-Matrix dataset. Building on a 3D Gaussian Splatting (3DGS) engine, we introduce a high-fidelity simulation platform supporting interactive roaming and collision-aware navigation. We further propose GN-Bench, the first BEV-based benchmark incorporating dynamic 3DGS avatars for human-robot interaction evaluation. To leverage the simulator, we develop an RL-driven navigation foundation model, Break and Establish (BAE). After supervised learning, DAgger exposes the model to rollout-induced states, breaking narrow expert-centric distributions and enabling downstream RL exploration. This unified VLN paradigm integrates map-based and map-free tasks, including instruction following, human following, and goal navigation. GN-BAE formalizes high-fidelity 3DGS-rendered Bird's Eye View representations as compact memory, unlocking latent spatial reasoning in VLMs. Extensive evaluations on GN-Bench and VLN-CE show that GN0 outperforms state-of-the-art VLN methods. Overall, GN-Matrix offers a unified framework spanning data, simulation, and learning, advancing embodied navigation in research and industrial applications.

cs.RO↗

arXiv2026

Mosaic: Towards Efficient Training of Multimodal Models with Spatial Resource Multiplexing

With the wide adoption of Multimodal Models (MMs) in real-world scenarios, it is significant to efficiently train emerging MMs that exhibit increasingly complex module architectures. For MM deployment, existing works allocate a GPU to only one MM module in a temporal-multiplexing manner; this compromises training efficiency because a single module often fails to achieve high GPU utilization. To improve GPU utilization and enable efficient MM training, we propose deploying MMs in a temporal-spatial multiplexing manner, allowing multiple MM modules to colocate on a GPU with well-controlled resource quotas. In this paper, we propose Apollo, an efficient MM training system that applies temporal-spatial multiplexing. We first develop a flexible and lightweight execution engine that supports MM training with arbitrary resource quotas, and then build a comprehensive and accurate performance model to estimate module execution time under different allocation plans. With the performance model, we further adopt effective heuristics to derive high-quality MM deployment plans efficiently. Testbed experiments confirm that Apollo effectively improves the training efficiency of popular MMs, with a training speedup of up to 1.31x.

cs.DC↗

arXiv2026

PRTS: A Primitive Reasoning and Tasking System via Contrastive Representations

Vision-Language-Action (VLA) models advance robotic control via strong visual-linguistic priors. However, existing VLAs predominantly frame pretraining as supervised behavior cloning, overlooking the fundamental nature of robot learning as a goal-reaching process that requires understanding temporal task progress. We present \textbf{PRTS} (\textbf{P}rimitive \textbf{R}easoning and \textbf{T}asking \textbf{S}ystem), a VLA foundation model that reformulates pretraining through Goal-Conditioned Reinforcement Learning. By treating language instructions as goals and employing contrastive reinforcement learning, PRTS learns a unified embedding space where the inner product of state-action and goal embeddings approximates the log-discounted goal occupancy, the probability of reaching the language-specified goal from the current state-action, quantitatively assessing physical feasibility beyond static semantic matching. PRTS draws this dense goal-reachability supervision directly from offline trajectories without reward annotations, and folds it into the VLM backbone via a role-aware causal mask, incurring negligible overhead over vanilla behavior cloning. This paradigm endows the high-level reasoning system with intrinsic goal reachability awareness, bridging semantic reasoning and temporal task progress, and further benefits goal-conditioned action prediction. Pretrained on 167B tokens of diverse manipulation and embodied-reasoning data, PRTS reaches state-of-the-art performance on LIBERO, LIBERO-Pro, LIBERO-Plus, SimplerEnv, and a real-world suite of 14 complex tasks, with particularly substantial gains on long-horizon, contact-rich, and zero-shot novel-instruction settings, confirming that injecting goal-reachability awareness significantly improves both execution success and long-horizon planning of general-purpose robotic foundation policies.

cs.AI↗

arXiv2026

Swift-SVD: Theoretical Optimality Meets Practical Efficiency in Low-Rank LLM Compression

The deployment of Large Language Models is constrained by the memory and bandwidth demands of static weights and dynamic Key-Value cache. SVD-based compression provides a hardware-friendly solution to reduce these costs. However, existing methods suffer from two key limitations: some are suboptimal in reconstruction error, while others are theoretically optimal but practically inefficient. In this paper, we propose Swift-SVD, an activation-aware, closed-form compression framework that simultaneously guarantees theoretical optimum, practical efficiency and numerical stability. Swift-SVD incrementally aggregates covariance of output activations given a batch of inputs and performs a single eigenvalue decomposition after aggregation, enabling training-free, fast, and optimal layer-wise low-rank approximation. We employ effective rank to analyze local layer-wise compressibility and design a dynamic rank allocation strategy that jointly accounts for local reconstruction loss and end-to-end layer importance. Extensive experiments across six LLMs and eight datasets demonstrate that Swift-SVD outperforms state-of-the-art baselines, achieving optimal compression accuracy while delivering 3-70X speedups in end-to-end compression time. Our code is available at https://github.com/hiahei/Swift-SVD.

cs.CL↗

arXiv2026

CALVO: Improve Serving Efficiency for LLM Inferences with Intense Network Demands

Distributed prefix caching has become a core technique for efficient LLM serving. However, for long-context requests with high cache hit ratios, retrieving reusable KVCache blocks from remote servers has emerged as a new performance bottleneck. Such network-intensive LLM inference is expected to become increasingly common as agentic AI workloads continue to grow. However, existing LLM inference engines remain largely compute-centric: they treat KVCache loading as a subordinate phase to GPU execution and often fail to account for its delay explicitly during scheduling. We present CALVO, an LLM serving engine that treats KVCache loading as a first-class concern. CALVO decouples KVCache loading and GPU computation into independently managed, asynchronously progressing stages, enabling better utilization of network, PCIe, and computation resources. In addition, CALVO incorporates KVCache loading delay as an explicit component of per-request service cost, leading to more accurate scheduling decisions. Experiments on a real testbed with diverse long-context workloads show that CALVO substantially improves the efficiency of network-intensive LLM inference, achieving up to 61.67% higher SLO attainment than the baseline.

cs.DC↗

arXiv2026

Tele-Omni: a Unified Multimodal Framework for Video Generation and Editing

Recent advances in diffusion-based video generation have substantially improved visual fidelity and temporal coherence. However, most existing approaches remain task-specific and rely primarily on textual instructions, limiting their ability to handle multimodal inputs, contextual references, and diverse video generation and editing scenarios within a unified framework. Moreover, many video editing methods depend on carefully engineered pipelines tailored to individual operations, which hinders scalability and composability. In this paper, we propose Tele-Omni, a unified multimodal framework for video generation and editing that follows multimodal instructions, including text, images, and reference videos, within a single model. Tele-Omni leverages pretrained multimodal large language models to parse heterogeneous instructions and infer structured generation or editing intents, while diffusion-based generators perform high-quality video synthesis conditioned on these structured signals. To enable joint training across heterogeneous video tasks, we introduce a task-aware data processing pipeline that unifies multimodal inputs into a structured instruction format while preserving task-specific constraints. Tele-Omni supports a wide range of video-centric tasks, including text-to-video generation, image-to-video generation, first-last-frame video generation, in-context video generation, and in-context video editing. By decoupling instruction parsing from video synthesis and combining it with task-aware data design, Tele-Omni achieves flexible multimodal control while maintaining strong temporal coherence and visual consistency. Experimental results demonstrate that Tele-Omni achieves competitive performance across multiple tasks.

cs.CV↗

arXiv2026

KV-CoRE: Benchmarking Data-Dependent Low-Rank Compressibility of KV-Caches in LLMs

Large language models rely on kv-caches to avoid redundant computation during autoregressive decoding, but as context length grows, reading and writing the cache can quickly saturate GPU memory bandwidth. Recent work has explored KV-cache compression, yet most approaches neglect the data-dependent nature of kv-caches and their variation across layers. We introduce KV-CoRE KV-cache Compressibility by Rank Evaluation), an SVD-based method for quantifying the data-dependent low-rank compressibility of kv-caches. KV-CoRE computes the optimal low-rank approximation under the Frobenius norm and, being gradient-free and incremental, enables efficient dataset-level, layer-wise evaluation. Using this method, we analyze multiple models and datasets spanning five English domains and sixteen languages, uncovering systematic patterns that link compressibility to model architecture, training data, and language coverage. As part of this analysis, we employ the Normalized Effective Rank as a metric of compressibility and show that it correlates strongly with performance degradation under compression. Our study establishes a principled evaluation framework and the first large-scale benchmark of kv-cache compressibility in LLMs, offering insights for dynamic, data-aware compression and data-centric model development.

cs.CL↗

arXiv2026

TeleBoost: A Systematic Alignment Framework for High-Fidelity, Controllable, and Robust Video Generation

Post-training is the decisive step for converting a pretrained video generator into a production-oriented model that is instruction-following, controllable, and robust over long temporal horizons. This report presents a systematical post-training framework that organizes supervised policy shaping, reward-driven reinforcement learning, and preference-based refinement into a single stability-constrained optimization stack. The framework is designed around practical video-generation constraints, including high rollout cost, temporally compounding failure modes, and feedback that is heterogeneous, uncertain, and often weakly discriminative. By treating optimization as a staged, diagnostic-driven process rather than a collection of isolated tricks, the report summarizes a cohesive recipe for improving perceptual fidelity, temporal coherence, and prompt adherence while preserving the controllability established at initialization. The resulting framework provides a clear blueprint for building scalable post-training pipelines that remain stable, extensible, and effective in real-world deployment settings.

cs.CV↗

arXiv2026

Beyond Binary: Turning Partial Success into Dense Verifiable Rewards for Reinforcement Learning in Code Generation

Effective reward design is a central challenge in Reinforcement Learning (RL) for code generation. Mainstream test-suite-level outcome rewards enforce functional correctness but induce sparsity, while external Reward Models (RMs) provide dense supervision at the cost of misalignment and additional overhead. Since code evaluation naturally yields multiple test-case-level outcomes, partial success, i.e., passing a subset of test cases, offers an intrinsic, verifiable source of dense supervision. In this paper, we propose VeRPO (Verifiable Dense Reward Policy Optimization), an RL framework that systematically turns verifiable partial success into reliable dense rewards. We analyze partial-success rewards using a weighted sum formulation, theoretically identifying a critical cardinality bias that causes policy updates to disproportionately favor gains from easy-test successes over progress on frontier tests. Based on this, VeRPO introduces a dynamic, density-calibrated local reward that explicitly corrects this bias and provides robust dense supervision from partial success. To enhance alignment with end-to-end functional correctness, VeRPO further integrates the local dense reward with global execution outcomes. Extensive experiments across diverse benchmarks and settings demonstrate that VeRPO outperforms outcome-driven and RM-based baselines, achieving up to +8.83 pass@1 gain with negligible time cost (< 0.02%) and zero GPU memory overhead.

cs.LG↗

arXiv2025

TeleWorld: Towards Dynamic Multimodal Synthesis with a 4D World Model

World models aim to endow AI systems with the ability to represent, generate, and interact with dynamic environments in a coherent and temporally consistent manner. While recent video generation models have demonstrated impressive visual quality, they remain limited in real-time interaction, long-horizon consistency, and persistent memory of dynamic scenes, hindering their evolution into practical world models. In this report, we present TeleWorld, a real-time multimodal 4D world modeling framework that unifies video generation, dynamic scene reconstruction, and long-term world memory within a closed-loop system. TeleWorld introduces a novel generation-reconstruction-guidance paradigm, where generated video streams are continuously reconstructed into a dynamic 4D spatio-temporal representation, which in turn guides subsequent generation to maintain spatial, temporal, and physical consistency. To support long-horizon generation with low latency, we employ an autoregressive diffusion-based video model enhanced with Macro-from-Micro Planning (MMPL)--a hierarchical planning method that reduces error accumulation from frame-level to segment-level-alongside efficient Distribution Matching Distillation (DMD), enabling real-time synthesis under practical computational budgets. Our approach achieves seamless integration of dynamic object modeling and static scene representation within a unified 4D framework, advancing world models toward practical, interactive, and computationally accessible systems. Extensive experiments demonstrate that TeleWorld achieves strong performance in both static and dynamic world understanding, long-term consistency, and real-time generation efficiency, positioning it as a practical step toward interactive, memory-enabled world models for multimodal generation and embodied intelligence.

cs.CV↗

arXiv2025

Computation-Bandwidth-Memory Trade-offs: A Unified Paradigm for AI Infrastructure

Large-scale artificial intelligence models are transforming industries and redefining human machine collaboration. However, continued scaling exposes critical limitations in hardware, including constraints on computation, bandwidth, and memory. These dimensions are tightly interconnected, so improvements in one often create bottlenecks in others, making isolated optimizations less effective. Balancing them to maximize system efficiency remains a central challenge in scalable AI design. To address this challenge, we introduce {Computation-Bandwidth-Memory Trade-offs}, termed the {AI Trinity}, a unified paradigm that positions {computation}, {bandwidth}, and {memory} as coequal pillars for next-generation AI infrastructure. AI Trinity enables dynamic allocation of resources across these pillars, alleviating single-resource bottlenecks and adapting to diverse scenarios to optimize system performance. Within this framework, AI Trinity identifies three fundamental trade-offs: (1) {More Computation$\rightarrow$Less Bandwidth}, wherein computational resources are exploited to reduce data transmission under limited bandwidth conditions, (2) {More Bandwidth$\rightarrow$Less Memory}, which exploits abundant communication capacity to populate or refresh memory when local storage resources are constrained, and (3) {More Memory$\rightarrow$Less Computation}, whereby storage capacity are utilized to mitigate redundant computation when computational costs are prohibitive. We illustrate the effectiveness of AI Trinity through representative system designs spanning edge-cloud communication, large-scale distributed training, and model inference. The innovations embodied in AI Trinity advance a new paradigm for scalable AI infrastructure, providing both a conceptual foundation and practical guidance for a broad range of application scenarios.

cs.DC↗

arXiv2025

Janus: Disaggregating Attention and Experts for Scalable MoE Inference

Serving large Mixture-of-Experts (MoE) models is challenging because of their large memory footprints, heterogeneous resource demands, and highly dynamic inference workloads. Most existing MoE inference systems deploy the entire model as a monolithic unit, forcing attention and MoE layers to share the same resource configuration despite their different scaling behaviors and resource bottlenecks. Such coarse-grained provisioning leads to resource inefficiency and suboptimal performance. We present JANUS, a scalable and resource-efficient MoE inference system built around three key principles. First, JANUS disaggregates attention and MoE layers onto separate GPU worker pools, enabling independent resource provisioning for the two layer types, and uses an adaptive two-phase communication mechanism for low-latency data exchange. Second, because MoE-layer execution is often memory-bound and highly sensitive to activated-expert imbalance, JANUS introduces a lightweight, microsecond-scale activation scheduler that balances per-layer activated experts across MoE instances to reduce inference latency. Third, JANUS employs a fine-grained, SLO-aware resource scaling scheme that jointly selects attention resources, MoE resources, and expert placement to minimize GPU cost under token-level SLOs. Evaluation shows that JANUS improves per-GPU throughput by up to 4.7x over state-of-the-art MoE inference baselines while satisfying token-level latency SLOs.

cs.DC↗

arXiv2025

Efficient Unified Caching for Accelerating Heterogeneous AI Workloads

Modern AI clusters, which host diverse workloads like data pre-processing, training and inference, often store the large-volume data in cloud storage and employ caching frameworks to facilitate remote data access. To avoid code-intrusion complexity and minimize cache space wastage, it is desirable to maintain a unified cache shared by all the workloads. However, existing cache management strategies, designed for specific workloads, struggle to handle the heterogeneous AI workloads in a cluster -- which usually exhibit heterogeneous access patterns and item storage granularities. In this paper, we propose IGTCache, a unified, high-efficacy cache for modern AI clusters. IGTCache leverages a hierarchical access abstraction, AccessStreamTree, to organize the recent data accesses in a tree structure, facilitating access pattern detection at various granularities. Using this abstraction, IGTCache applies hypothesis testing to categorize data access patterns as sequential, random, or skewed. Based on these detected access patterns and granularities, IGTCache tailors optimal cache management strategies including prefetching, eviction, and space allocation accordingly. Experimental results show that IGTCache increases the cache hit ratio by 55.6% over state-of-the-art caching frameworks, reducing the overall job completion time by 52.2%.

cs.DC↗

arXiv2024

Efficient Training of Large Language Models on Distributed Infrastructures: A Survey

Large Language Models (LLMs) like GPT and LLaMA are revolutionizing the AI industry with their sophisticated capabilities. Training these models requires vast GPU clusters and significant computing time, posing major challenges in terms of scalability, efficiency, and reliability. This survey explores recent advancements in training systems for LLMs, including innovations in training infrastructure with AI accelerators, networking, storage, and scheduling. Additionally, the survey covers parallelism strategies, as well as optimizations for computation, communication, and memory in distributed LLM training. It also includes approaches of maintaining system reliability over extended training periods. By examining current innovations and future directions, this survey aims to provide valuable insights towards improving LLM training systems and tackling ongoing challenges. Furthermore, traditional digital circuit-based computing systems face significant constraints in meeting the computational demands of LLMs, highlighting the need for innovative solutions such as optical computing and optical networks.

cs.DC↗

arXiv2024

InternLM2 Technical Report

The evolution of Large Language Models (LLMs) like ChatGPT and GPT-4 has sparked discussions on the advent of Artificial General Intelligence (AGI). However, replicating such advancements in open-source models has been challenging. This paper introduces InternLM2, an open-source LLM that outperforms its predecessors in comprehensive evaluations across 6 dimensions and 30 benchmarks, long-context modeling, and open-ended subjective evaluations through innovative pre-training and optimization techniques. The pre-training process of InternLM2 is meticulously detailed, highlighting the preparation of diverse data types including text, code, and long-context data. InternLM2 efficiently captures long-term dependencies, initially trained on 4k tokens before advancing to 32k tokens in pre-training and fine-tuning stages, exhibiting remarkable performance on the 200k ``Needle-in-a-Haystack" test. InternLM2 is further aligned using Supervised Fine-Tuning (SFT) and a novel Conditional Online Reinforcement Learning from Human Feedback (COOL RLHF) strategy that addresses conflicting human preferences and reward hacking. By releasing InternLM2 models in different training stages and model sizes, we provide the community with insights into the model's evolution.

cs.CL↗

arXiv2024

CaraServe: CPU-Assisted and Rank-Aware LoRA Serving for Generative LLM Inference

Pre-trained large language models (LLMs) often need specialization for domain-specific tasks. Low-Rank Adaptation (LoRA) is a popular approach that adapts a base model to multiple tasks by adding lightweight trainable adapters. In this paper, we present CaraServe, a system that efficiently serves many LoRA adapters derived from a common base model. CaraServe maintains the base model on GPUs and dynamically loads activated LoRA adapters from main memory. As GPU loading results in a cold-start that substantially delays token generation, CaraServe employs a CPU-assisted approach. It early starts the activated adapters on CPUs for prefilling as they are being loaded onto GPUs; after loading completes, it then switches to the GPUs for generative LoRA inference. CaraServe develops a highly optimized synchronization mechanism to efficiently coordinate LoRA computation on the CPU and GPU. Moreover, CaraServe employs a rank-aware scheduling algorithm to optimally schedule heterogeneous LoRA requests for maximum service-level objective (SLO) attainment. We have implemented CaraServe and evaluated it against state-of-the-art LoRA serving systems. Our results demonstrate that CaraServe can speed up the average request serving latency by up to 1.4$\times$ and achieve an SLO attainment of up to 99%.

cs.DC↗

arXiv2020

Semi-Dynamic Load Balancing: Efficient Distributed Learning in Non-Dedicated Environments

Machine learning (ML) models are increasingly trained in clusters with non-dedicated workers possessing heterogeneous resources. In such scenarios, model training efficiency can be negatively affected by stragglers -- workers that run much slower than others. Efficient model training requires eliminating such stragglers, yet for modern ML workloads, existing load balancing strategies are inefficient and even infeasible. In this paper, we propose a novel strategy called semi-dynamic load balancing to eliminate stragglers of distributed ML workloads. The key insight is that ML workers shall be load-balanced at iteration boundaries, being non-intrusive to intra-iteration execution. We develop LB-BSP based on such an insight, which is an integrated worker coordination mechanism that adapts workers' load to their instantaneous processing capabilities by right-sizing the sample batches at the synchronization barriers. We have custom-designed the batch sizing algorithm respectively for CPU and GPU clusters based on their own characteristics. LB-BSP has been implemented as a Python module for ML frameworks like TensorFlow and PyTorch. Our EC2 deployment confirms that LB-BSP is practical, effective and light-weight, and is able to accelerating distributed training by up to $54\%$.

cs.DC↗