Tingfeng Lan

Tingfeng Lan, Zirui Wang, Yunjia Zheng et al. · cmu

Modern AI models are growing rapidly in size and redundancy, leading to significant storage and distribution challenges in model hubs. We present TensorHub, a tensor-centric system for reducing storage overhead through fine-grained deduplication and compression. TensorHub leverages tensor-level fingerprinting and clustering to identify redundancy across models without requiring annotations. Our design enables efficient storage reduction while preserving model usability and performance. Experiments on real-world model repositories demonstrate substantial storage savings with minimal overhead.

Qinlong Wang, Tingfeng Lan, Yinghao Tang et al.

Deep learning recommendation models (DLRM) rely on large embedding tables to manage categorical sparse features. Expanding such embedding tables can significantly enhance model performance, but at the cost of increased GPU/CPU/memory usage. Meanwhile, tech companies have built extensive cloud-based services to accelerate training DLRM models at scale. In this paper, we conduct a deep investigation of the DLRM training platforms at AntGroup and reveal two critical challenges: low resource utilization due to suboptimal configurations by users and the tendency to encounter abnormalities due to an unstable cloud environment. To overcome them, we introduce DLRover-RM, an elastic training framework for DLRMs designed to increase resource utilization and handle the instability of a cloud environment. DLRover-RM develops a resource-performance model by considering the unique characteristics of DLRMs and a three-stage heuristic strategy to automatically allocate and dynamically adjust resources for DLRM training jobs for higher resource utilization. Further, DLRover-RM develops multiple mechanisms to ensure efficient and reliable execution of DLRM training jobs. Our extensive evaluation shows that DLRover-RM reduces job completion times by 31%, increases the job completion rate by 6%, enhances CPU usage by 15%, and improves memory utilization by 20%, compared to state-of-the-art resource scheduling frameworks. DLRover-RM has been widely deployed at AntGroup and processes thousands of DLRM training jobs on a daily basis. DLRover-RM is open-sourced and has been adopted by 10+ companies.

Yinghao Tang, Xueding Liu, Boyuan Zhang et al.

Infographics are composite visual artifacts that combine data visualizations with textual and illustrative elements to communicate information. While recent text-to-image (T2I) models can generate aesthetically appealing images, their reliability in generating infographics remains unclear. Generated infographics may appear correct at first glance but contain easily overlooked issues, such as distorted data encoding or incorrect textual content. We present IGENBENCH, the first benchmark for evaluating the reliability of text-to-infographic generation, comprising 600 curated test cases spanning 30 infographic types. We design an automated evaluation framework that decomposes reliability verification into atomic yes/no questions based on a taxonomy of 10 question types. We employ multimodal large language models (MLLMs) to verify each question, yielding question-level accuracy (Q-ACC) and infographic-level accuracy (I-ACC). We comprehensively evaluate 10 state-of-the-art T2I models on IGENBENCH. Our systematic analysis reveals key insights for future model development: (i) a three-tier performance hierarchy with the top model achieving Q-ACC of 0.90 but I-ACC of only 0.49; (ii) data-related dimensions emerging as universal bottlenecks (e.g., Data Completeness: 0.21); and (iii) the challenge of achieving end-to-end correctness across all models. We release IGENBENCH at https://igen-bench.vercel.app/.

Zirui Wang, Yunjia Zheng, Tingfeng Lan et al.

The explosive growth of AI-generated images has created a sustainability challenge for storage infrastructure. Platforms like Midjourney and Adobe Firefly already host billions of generative images, yet conventional object stores persist them as blobs with full-resolution pixels, consuming huge amounts of storage capacity and bandwidth. Unlike natural photos, however, AI-generated images can be deterministically reconstructed from compact, model-native latent tensors, making persistent image storage fundamentally redundant. This paper presents LatentBox, a latent-first storage system for AI-generated images. LatentBox treats compressed latents as durable storage objects and uses on-demand GPU reconstruction on the read path to trade inexpensive compute for large persistent storage savings. Our design is guided by the first large-scale analysis of AI-generated image access we are aware of, based on a 35-month, 2-billion-request production trace from a major generative-content platform. Motivated by the trace analysis, LatentBox keeps frequently accessed images in decoded pixel format for fast hits, stores less-active objects as compressed latents to expand effective cache capacity, and continuously adjusts the splits between the image and latent cache to optimize user-perceived access latency.We build a LatentBox prototype and evaluate it with the production trace. LatentBox reduces persistent storage by 78.7% with competitive or even lower mean and tail latency over a pure image-based storage.

Yinghao Tang, Yupeng Xie, Yingchaojie Feng et al.

Interactive documents help readers engage with complex ideas through dynamic visualization, interactive animations, and exploratory interfaces. However, creating such documents remains costly, as it requires both domain expertise and web development skills. Recent Large Language Model (LLM)-based agents can automate content creation, but directly applying them to interactive document generation often produces outputs that are difficult to control. To address this, we present ViviDoc, to the best of our knowledge the first work to systematically address interactive document generation. ViviDoc introduces a multi-agent pipeline (Planner, Styler, Executor, Evaluator). To make the generation process controllable, we provide three levels of human control: (1) the Document Specification (DocSpec) with SRTC Interaction Specifications (State, Render, Transition, Constraint) for structured planning, (2) a content-aware Style Palette for customizing writing and interaction styles, and (3) chat-based editing for iterative refinement. We also construct ViviBench, a benchmark of 101 topics derived from real-world interactive documents across 11 domains, along with a taxonomy of 8 interaction types and a 4-dimensional automated evaluation framework validated against human ratings (Pearson r > 0.84). Experiments show that ViviDoc achieves the highest content richness and interaction quality in both automated and human evaluation. A 12-person user study confirms that the system is easy to use, provides effective control over the generation process, and produces documents that satisfy users.

Yinghao Tang, Yupeng Xie, Yingchaojie Feng et al.

Interactive articles help readers engage with complex ideas through exploration, yet creating them remains costly, requiring both domain expertise and web development skills. Recent LLM-based agents can automate content creation, but naively applying them yields uncontrollable and unverifiable outputs. We present ViviDoc, a human-agent collaborative system that generates interactive educational documents from a single topic input. ViviDoc introduces a multi-agent pipeline (Planner, Executor, Evaluator) and the Document Specification (DocSpec), a human-readable intermediate representation that decomposes each interactive visualization into State, Render, Transition, and Constraint components. The DocSpec enables educators to review and refine generation plans before code is produced, bridging the gap between pedagogical intent and executable output. Expert evaluation and a user study show that ViviDoc substantially outperforms naive agentic generation and provides an intuitive editing experience. Our project homepage is available at https://vividoc-homepage.vercel.app/.

Zhengmao Ye, Dengchun Li, Zetao Hu et al.

Transformer-based, pre-trained large language models (LLMs) have demonstrated outstanding performance across diverse domains, particularly in the emerging {\em pretrain-then-finetune} paradigm. Low-Rank Adaptation (LoRA), a parameter-efficient fine-tuning method, is commonly used to adapt a base LLM to multiple downstream tasks. Further, LLM platforms enable developers to fine-tune multiple models and develop various domain-specific applications simultaneously. However, existing model parallelism schemes suffer from high communication overhead and inefficient GPU utilization when training multiple LoRA tasks across GPUs and machines. In this paper, we present mLoRA, a parallelism-efficient fine-tuning system designed for training multiple LoRA across GPUs and machines. mLoRA introduces a novel LoRA-aware pipeline parallelism scheme that efficiently pipelines independent LoRA adapters and their distinct fine-tuning stages across GPUs and machines, along with a new LoRA-efficient operator to enhance GPU utilization during pipelined LoRA training. Our extensive evaluation shows that mLoRA can significantly reduce average fine-tuning task completion time, e.g., by 30\%, compared to state-of-the-art methods like FSDP. More importantly, mLoRA enables simultaneous fine-tuning of larger models, e.g., two Llama-2-13B models on four NVIDIA RTX A6000 48GB GPUs, which is not feasible for FSDP due to high memory requirements. Hence, mLoRA not only increases fine-tuning efficiency but also makes it more accessible on cost-effective GPUs. mLoRA has been deployed in AntGroup's production environment.

Tingfeng Lan, Yusen Wu, Bin Ma et al.

Fine-tuning large language models (LLMs) often exceeds GPU memory limits, prompting systems to offload model states to CPU memory. However, existing offloaded training frameworks like ZeRO-Offload treat all parameters equally and update the full model on the CPU, causing severe GPU stalls, where fast, expensive GPUs sit idle waiting for slow CPU updates and limited-bandwidth PCIe transfers. We present ZenFlow, a new offloading framework that prioritizes important parameters and decouples updates between GPU and CPU. ZenFlow performs in-place updates of important gradients on GPU, while asynchronously offloading and accumulating less important ones on CPU, fully overlapping CPU work with GPU computation. To scale across GPUs, ZenFlow introduces a lightweight gradient selection method that exploits a novel spatial and temporal locality property of important gradients, avoiding costly global synchronization. ZenFlow achieves up to 5x end-to-end speedup, 2x lower PCIe traffic, and reduces GPU stalls by over 85 percent, all while preserving accuracy.

Yinghao Tang, Tingfeng Lan, Xiuqi Huang et al.

Existing Large Language Model (LLM) serving systems prioritize maximum throughput. They often neglect Service Level Objectives (SLOs) such as Time to First Token (TTFT) and Time Per Output Token (TPOT), which leads to suboptimal SLO attainment. This paper introduces SCORPIO, an SLO-oriented LLM serving system designed to maximize system goodput and SLO attainment for workloads with heterogeneous SLOs. Our core insight is to exploit SLO heterogeneity for adaptive scheduling across admission control, queue management, and batch selection. SCORPIO features a TTFT Guard, which employs least-deadline-first reordering and rejects unattainable requests, and a TPOT Guard, which utilizes a VBS-based admission control and a novel credit-based batching mechanism. Both guards are supported by a predictive module. Evaluations demonstrate that SCORPIO improves system goodput by up to 14.4X and SLO adherence by up to 46.5% compared to state-of-the-art baselines.

Zhaoyuan Su, Tingfeng Lan, Zirui Wang et al.

Efficiently serving large language models (LLMs) under dynamic and bursty workloads remains a key challenge for real-world deployment. Existing serving frameworks and static model compression techniques fail to adapt to workload fluctuations, leading to either service-level objective (SLO) violations under full-precision serving or persistent accuracy degradation with static quantization. We present MorphServe, a dynamic, workload-aware LLM serving framework based on morphological adaptation. MorphServe introduces two asynchronous, token-level runtime mechanisms: quantized layer swapping, which selectively replaces less impactful layers with quantized alternatives during high-load periods, and pressure-aware KV cache resizing, which dynamically adjusts KV cache capacity in response to memory pressure. These mechanisms enable state-preserving transitions with minimum runtime overhead and are fully compatible with modern scheduling and attention techniques. Extensive experiments on Vicuna and Llama family models with real-world workloads demonstrate that MorphServe reduces average SLO violations by 92.45 percent and improves the P95 TTFT latency by 2.2x-3.9x compared to full-precision serving, without compromising generation quality. These results establish MorphServe as a practical and elastic solution for LLM deployment in dynamic environments.