arXiv subjects

Myungjin Lee

Publications and source records attributed to Myungjin Lee.

At least 19 recordsLinked to original sources

arXiv2026

Robust Federated Learning Under Real-World Client Churn

Federated Learning (FL) enables training shared models on private, on-device data, but production deployments remain constrained to slow, multi-day refresh cycles due to the complexity of coordinating massive client populations. For applications such as feed ranking, ad targeting, and personalized recommendation, model freshness: the ability to rapidly adapt to new user-local data is critical for maximizing objectives like click-through rate. This lag leaves models stale and unresponsive to volatile data distributions driven by viral trends and shifting user intent. Bridging this gap requires addressing three challenges overlooked by existing FL systems: transient client availability, dynamic data heterogeneity, and delays between model predictions and observable outcomes. We present FeLiX, an FL orchestration framework that minimizes wall-clock time-to-target accuracy on live interaction streams. FeLiX introduces three primitives: (i) streaming-aware availability tiers that leverage lightweight telemetry to identify ready clients at scale; (ii) fresh-utility selection, a dual-tier mechanism that prioritizes statistically valuable updates from devices able to meet tight refresh deadlines; and (iii) informativeness-aware, delay-robust aggregation that incorporates late, high-value updates containing ground-truth outcomes without biasing the global model toward stale distributions. Unlike prior systems that rely on unrealistic oracular knowledge of client availability, FeLiX achieves near-oracular performance in real-world settings. Across CIFAR-10, Google Speech, and realistic low-availability traces, FeLiX reduces wall-clock time-to-target accuracy by up to 2.37X while reducing communication bandwidth by 1.30X compared to state-of-the-art synchronous and asynchronous FL baselines.

cs.LG↗

arXiv2026

Sangam: Efficiently Serving Diffusion LLMs with the AR Stack

Diffusion language models (dLLMs) generate text by iteratively denoising a masked response and can commit multiple output positions per model invocation. Their bidirectional attention prevents exact autoregressive-style KV caching, since committing one position shifts the KV activations of all others. Approximate caching techniques such as Fast-dLLM and dKV-Cache refresh KV activations repeatedly and reuse them across intervening decodes, inducing a repeated prefill/decode structure. This makes AR serving mechanisms relevant to dLLMs, but not directly applicable. dLLM decodes are block-sized rather than token-sized, prefills recur, and bidirectional attention precludes the chunked prefill mechanism used for stall-free colocated serving. We present Sangam, a serving system for cached dLLM inference. Sangam introduces a deficit token-budget scheduler that admits in-flight decodes first, admits whole indivisible prefills only when the accumulated token budget allows, and carries unused budget forward. This achieves amortized stall-free scheduling. Disaggregated serving avoids prefill-decode interference but suffers from prefill/decode resource partitioning problem. Sangam adopts a hybrid serving strategy, overflowing prefills onto decode workers to relieve prefill under-provisioning, and uses the same deficit-budget scheduler to protect those workers' decodes from the overflow. We show that like AR serving, dLLM serving design space is governed by prefill-decode interference and prefill/decode partitioning. Colocated serving is most effective on decode-heavy workloads, cutting mean latency by 9-20% over hybrid execution on LLaDA-8B ShareGPT; while hybrid execution is most effective on prefill-heavy workloads, cutting mean latency by 8-20% over colocated execution on Dream-7B arXiv. Sangam is available at https://github.com/UT-InfraAI/sangam.

cs.DC↗

arXiv2026

Harmonia: End-to-End RAG Serving Optimization

Retrieval-Augmented Generation (RAG) improves the reliability of large language models by integrating external knowledge, but serving RAG pipelines efficiently is challenging because requests traverse heterogeneous components spanning LLM inference, databases, and CPU-side processing. We present Harmonia, an end-to-end RAG serving framework that addresses these bottlenecks through (i) a flexible pipeline specification interface for composing custom workflows, (ii) heterogeneity-aware deployment that provisions and configures components as a distributed inference system, and (iii) a closed-loop runtime controller that monitors load and execution progress and reduces SLO violations through request prioritization and auto-scaling. Across four RAG applications, Harmonia outperforms commercial alternatives, improving throughput by more than 2.04x while reducing SLO violations by up to 78.4 percent.

cs.DC↗

arXiv2026

KLAS: Using Similarity to Stitch Neural Networks for Improved Accuracy-Efficiency Tradeoffs

Given the wide range of deployment targets, flexible model selection is essential for optimizing performance within a given compute budget. Recent work demonstrates that stitching pretrained models within a model family enables cost-effective interpolation of the accuracy-efficiency tradeoff space. Stitching transforms intermediate activations from one pretrained model into another, producing a new interpolated stitched network. Such networks provide a pool of deployment options along the accuracy-efficiency spectrum. However, existing stitching approaches often yield suboptimal tradeoffs and lack generalizability, as they primarily rely on heuristics to select stitch configurations. We argue that constructing improved accuracy-efficiency tradeoffs requires explicitly capturing and leveraging the similarity between pretrained models being stitched. To this end, we introduce KLAS, a novel stitch selection framework that automates and generalizes stitch selection across model families by leveraging KL divergence between intermediate representations. KLAS identifies the most promising binary stitches from the $O(k^2n^2)$ possibilities for $k$ pretrained models of depth $n$. Through comprehensive experiments, we demonstrate that KLAS improves the accuracy-efficiency curve of stitched models at the same finetuning cost as baselines. KLAS achieves up to $1.21\%$ higher ImageNet-1K top-1 accuracy at the same computational cost, or maintains accuracy with a $1.33\times$ reduction in FLOPs.

cs.LG↗

arXiv2026

Cornserve: A Distributed Serving System for Any-to-Any Multimodal Models

Any-to-Any models are an emerging class of multimodal models that accept combinations of multimodal data (e.g., text, image, video, audio) as input and generate them as output. Serving these models are challenging; different requests with different input and output modalities traverse different paths through the model computation graph, and each component of the model have different scaling characteristics. We present Cornserve, a distributed serving system for generic Any-to-Any models. Cornserve provides a flexible task abstraction for expressing Any-to-Any model computation graphs, enabling component disaggregation and independent scaling. The distributed runtime dispatches compute to the data plane via an efficient record-and-replay execution model that keeps track of data dependencies, and forwards tensor data between components directly from the producer to the consumer. Built on Kubernetes with approximately 23K new lines of Python, Cornserve supports diverse Any-to-Any models and delivers up to 3.81$\times$ higher throughput and 5.79$\times$ lower tail latency. Cornserve is open-source, and the demo video is available on YouTube.

cs.LG↗

arXiv2026

Cornfigurator: Automated Planning for Any-to-Any Multimodal Model Serving

Any-to-Any models are an emerging class of multimodal models that accept combinations of text and multimodal data as input and generate them as output, introducing heterogeneous computation paths and component scaling characteristics. There are existing mechanisms for deploying Any-to-Any models--or special cases of them--for inference serving, but they either require manual effort and expertise to tune, or do not generalize to generic Any-to-Any models. We present Cornfigurator, the first deployment planner for generic Any-to-Any model inference serving. The goal of Cornfigurator is to maximize the overall goodput of serving the model, defined as the throughput of requests meeting their latency targets. To do so, based on model and workload characteristics, Cornfigurator explores the full spectrum of deployment strategies, from colocation to disaggregation and mixing different strategies. Cornfigurator performs coarse-to-fine statistical evaluation to efficiently navigate the large space of candidate plans. Plans generated by Cornfigurator either match or deliver 1.12$\times$-6.32$\times$ higher goodput compared to existing systems and expert-tuned deployment plans.

cs.LG↗

arXiv2026

On Harnessing Idle Compute at the Edge for Foundation Model Training

The foundation-model ecosystem remains highly centralized because training requires immense compute resources and is therefore largely limited to large cloud operators. Edge-assisted foundation model training that harnesses spare compute on edge devices offers a more democratized alternative. However, existing edge-training approaches fall short: they struggle to match cloud-training performance, scale to larger models, fit within device memory limits, or keep communication overhead manageable. They also do not handle device heterogeneity and churn satisfactorily. We introduce Cleave, built on a structural insight: each GEMM has an asymmetric I/O pattern -- its input matrices, sent over downlink, are much larger than the partial output blocks returned over uplink -- matching edge networks where downlink bandwidth exceeds uplink by 2--10x. Exploiting this alignment with a parameter-server-centric architecture, Cleave makes per-device communication \emph{decrease} as more devices join, rather than stay constant as in conventional TP. Decomposing training into independent sub-GEMM tasks yields one scheduling abstraction that unifies memory constraints, communication overhead, and fault tolerance under device churn. Our evaluation shows that Cleave achieves cloud-comparable GPU training performance and outperforms state-of-the-art edge-training methods by 4--10x in per-batch runtime at the same device counts. Beyond this shared operating range, Cleave scales to thousands of heterogeneous devices -- a regime where prior edge-training systems cannot operate -- and achieves at least 100x faster recovery from device failures.

cs.DC↗

arXiv2026

GENSERVE: Efficient Co-Serving of Heterogeneous Diffusion Model Workloads

Diffusion models have emerged as the prevailing approach for text-to-image (T2I) and text-to-video (T2V) generation, yet production platforms must increasingly serve both modalities on shared GPU clusters while meeting stringent latency SLOs. Co-serving such heterogeneous workloads is challenging: T2I and T2V requests exhibit vastly different compute demands, parallelism characteristics, and latency requirements, leading to significant SLO violations in existing serving systems. We present GENSERVE, a co-serving system that leverages the inherent predictability of the diffusion process to optimize serving efficiency. A central insight is that diffusion inference proceeds in discrete, predictable steps and is naturally preemptible at step boundaries, opening a new design space for heterogeneity-aware resource management. GENSERVE introduces step-level resource adaptation through three coordinated mechanisms: intelligent video preemption, elastic sequence parallelism with dynamic batching, and an SLO-aware scheduler that jointly optimizes resource allocation across all concurrent requests. Experimental results show that GENSERVE improves the SLO attainment rate by up to 44% over the strongest baseline across diverse configurations.

cs.DC↗

arXiv2026

FedMOA: Federated GRPO for Personalized Reasoning LLMs under Heterogeneous Rewards

Group Relative Policy Optimization (GRPO) has recently emerged as an effective approach for improving the reasoning capabilities of large language models through online multi-objective reinforcement learning. While personalization on private data is increasingly vital, traditional Reinforcement Learning (RL) alignment is often memory-prohibitive for on-device federated learning due to the overhead of maintaining a separate critic network. GRPO's critic-free architecture enables feasible on-device training, yet transitioning to a federated setting introduces systemic challenges: heterogeneous reward definitions, imbalanced multi-objective optimization, and high training costs. We propose FedMOA, a federated GRPO framework for multi-objective alignment under heterogeneous rewards. FedMOA stabilizes local training through an online adaptive weighting mechanism via hypergradient descent, which prioritizes primary reasoning as auxiliary objectives saturate. On the server side, it utilizes a task- and accuracy-aware aggregation strategy to prioritize high-quality updates. Experiments on mathematical reasoning and code generation benchmarks demonstrate that FedMOA consistently outperforms federated averaging, achieving accuracy gains of up to 2.2% while improving global performance, personalization, and multi-objective balance.

cs.LG↗

arXiv2026

Erasing Your Voice Before It's Heard: Training-free Speaker Unlearning for Zero-shot Text-to-Speech

Modern zero-shot text-to-speech (TTS) models offer unprecedented expressivity but also pose serious crime risks, as they can synthesize voices of individuals who never consented. In this context, speaker unlearning aims to prevent the generation of specific speaker identities upon request. Existing approaches, reliant on retraining, are costly and limited to speakers seen in the training set. We present TruS, a training-free speaker unlearning framework that shifts the paradigm from data deletion to inference-time control. TruS steers identity-specific hidden activations to suppress target speakers while preserving other attributes (e.g., prosody and emotion). Experimental results show that TruS effectively prevents voice generation on both seen and unseen opt-out speakers, establishing a scalable safeguard for speech synthesis. The demo and code are available on http://mmai.ewha.ac.kr/trus.

eess.AS↗

arXiv2026

Nalar: An agent serving framework

LLM-driven agentic applications increasingly automate complex, multi-step tasks, but serving them efficiently remains challenging due to heterogeneous components, dynamic and model-driven control flow, long-running state, and unpredictable latencies. Nalar is a ground-up agent-serving framework that cleanly separates workflow specification from execution while providing the runtime visibility and control needed for robust performance. Nalar preserves full Python expressiveness, using lightweight auto-generated stubs that turn agent and tool invocations into futures carrying dependency and context metadata. A managed state layer decouples logical state from physical placement, enabling safe reuse, migration, and consistent retry behavior. A two-level control architecture combines global policy computation with local event-driven enforcement to support adaptive routing, scheduling, and resource management across evolving workflows. Together, these mechanisms allow Nalar to deliver scalable, efficient, and policy-driven serving of heterogeneous agentic applications without burdening developers with orchestration logic. Across three agentic workloads, Nalar cuts tail latency by 34--74\%, achieves up to $2.9\times$ speedups, sustains 80 RPS where baselines fail, and scales to 130K futures with sub-500 ms control overhead.

cs.DC↗

arXiv2026

Making MoE-based LLM Inference Resilient with Tarragon

Mixture-of-Experts (MoE) models are increasingly used to serve LLMs at scale, but failures become common as deployment scale grows. Existing systems exhibit poor failure resilience: even a single worker failure triggers a coarse-grained, service-wide restart, discarding accumulated progress and halting the entire inference pipeline during recovery--an approach clearly ill-suited for latency-sensitive, LLM services. We present Tarragon, a resilient MoE inference framework that confines the failures impact to individual workers while allowing the rest of the pipeline to continue making forward progress. Tarragon exploits the natural separation between the attention and expert computation in MoE-based transformers, treating attention workers (AWs) and expert workers (EWs) as distinct failure domains. Tarragon introduces a reconfigurable datapath to mask failures by rerouting requests to healthy workers. On top of this datapath, Tarragon implements a self-healing mechanism that relaxes the tightly synchronized execution of existing MoE frameworks. For stateful AWs, Tarragon performs asynchronous, incremental KV cache checkpointing with per-request restoration, and for stateless EWs, it leverages residual GPU memory to deploy shadow experts. These together keep recovery cost and recomputation overhead extremely low. Our evaluation shows that, compared to state-of-the-art MegaScale-Infer, Tarragon reduces failure-induced stalls by 160-213x (from ~64 s down to 0.3-0.4 s) while preserving performance when no failures occur.

cs.DC↗

arXiv2026

Software-Defined Agentic Serving

As multi-agent LLM pipelines grow in complexity, existing serving paradigms fail to adapt to the dynamic serving conditions. We argue that agentic serving systems should be programmable and system-aware, unlike existing serving which statically encode the parameters. In this work, we propose a new SDN-inspired agentic serving framework that helps control the key attributes of communication based on runtime state. This architecture enables serving-efficient, responsive agent systems and paves the way for high-level intent-driven agentic serving.

cs.DC↗

arXiv2025

JITServe: SLO-aware LLM Serving with Imprecise Request Information

The integration of Large Language Models (LLMs) into applications ranging from interactive chatbots to multi-agent systems has introduced a wide spectrum of service-level objectives (SLOs) for responsiveness. These include latency-sensitive requests emphasizing per-token latency in streaming chat, deadline-sensitive requests requiring rapid full responses to trigger external tools, and compound requests with evolving dependencies across multiple LLM calls. Despite-or perhaps, because of-this workload diversity and unpredictable request information (e.g., response lengths and dependencies), existing request schedulers have focused on aggregate performance, unable to ensure application-level SLO needs. This paper presents JITServe, the first SLO-aware LLM serving system designed to maximize service goodput (e.g., the number of tokens meeting request SLOs) across diverse workloads. JITServe novelly schedules requests using imprecise request information and gradually relaxes this conservatism by refining request information estimates as generation progresses. It applies a grouped margin goodput maximization algorithm to allocate just enough serving bandwidth to satisfy each request's SLO just-in-time (JIT), maximizing residual capacity for others, while deciding the composition of requests in a batch to maximize efficiency and goodput with provable guarantees. Our evaluation across diverse realistic workloads, including chat, deep research, and agentic pipelines, shows that JITServe improves service goodput by 1.4x-6.3x, alternatively achieving 28.5%-83.2% resource savings, compared to state-of-the-art designs.

cs.DC↗

arXiv2025

SuperGen: An Efficient Ultra-high-resolution Video Generation System with Sketching and Tiling

Diffusion models have recently achieved remarkable success in generative tasks (e.g., image and video generation), and the demand for high-quality content (e.g., 2K/4K videos) is rapidly increasing across various domains. However, generating ultra-high-resolution videos on existing standard-resolution (e.g., 720p) platforms remains challenging due to the excessive re-training requirements and prohibitively high computational and memory costs. To this end, we introduce SUPERGEN, an efficient tile-based framework for ultra-high-resolution video generation. SUPERGEN features a novel training-free algorithmic innovation with tiling to successfully support a wide range of resolutions without additional training efforts while significantly reducing both memory footprint and computational complexity. Moreover, SUPERGEN incorporates a tile-tailored, adaptive, region-aware caching strategy that accelerates video generation by exploiting redundancy across denoising steps and spatial regions. SUPERGEN also integrates cache-guided, communication-minimized tile parallelism for enhanced throughput and minimized latency. Evaluations show that SUPERGEN maximizes performance gains while achieving high output quality across various benchmarks.

cs.LG↗

arXiv2025

An Efficient and Adaptive Watermark Detection System with Tile-based Error Correction

Efficient and reliable detection of generated images is critical for the responsible deployment of generative models. Existing approaches primarily focus on improving detection accuracy and robustness under various image transformations and adversarial manipulations, yet they largely overlook the efficiency challenges of watermark detection across large-scale image collections. To address this gap, we propose QRMark, an efficient and adaptive end-to-end method for detecting embedded image watermarks. The core idea of QRMark is to combine QR Code-inspired error correction with tailored tiling techniques to improve detection efficiency while preserving accuracy and robustness. At the algorithmic level, QRMark employs a Reed-Solomon error correction mechanism to mitigate the accuracy degradation introduced by tiling. At the system level, QRMark implements a resource-aware multi-channel horizontal fusion policy that adaptively assigns more streams to GPU-intensive stages of the detection pipeline. It further employs a tile-based workload interleaving strategy to overlap data-loading overhead with computation and schedules kernels across stages to maximize efficiency. End-to-end evaluations show that QRMark achieves an average 2.43x inference speedup over the sequential baseline.

cs.DC↗

arXiv2025

Dora: QoE-Aware Hybrid Parallelism for Distributed Edge AI

With the proliferation of edge AI applications, satisfying user quality of experience (QoE) requirements, such as model inference latency, has become a first class objective, as these models operate in resource constrained settings and directly interact with users. Yet, modern AI models routinely exceed the resource capacity of individual devices, necessitating distributed execution across heterogeneous devices over variable and contention prone networks. Existing planners for hybrid (e.g., data and pipeline) parallelism largely optimize for throughput or device utilization, overlooking QoE, leading to severe resource inefficiency (e.g., unnecessary energy drain) or QoE violations under runtime dynamics. We present Dora, a framework for QoE aware hybrid parallelism in distributed edge AI training and inference. Dora jointly optimizes heterogeneous computation, contention prone networks, and multi dimensional QoE objectives via three key mechanisms: (i) a heterogeneity aware model partitioner that determines and assigns model partitions across devices, forming a compact set of QoE compliant plans; (ii) a contention aware network scheduler that further refines these candidate plans by maximizing compute communication overlap; and (iii) a runtime adapter that adaptively composes multiple plans to maximize global efficiency while respecting overall QoEs. Across representative edge deployments, including smart homes, traffic analytics, and small edge clusters, Dora achieves 1.1--6.3 times faster execution and, alternatively, reduces energy consumption by 21--82 percent, all while maintaining QoE under runtime dynamics.

cs.DC↗

arXiv2025

CountSteer: Steering Attention for Object Counting in Diffusion Models

Text-to-image diffusion models generate realistic and coherent images but often fail to follow numerical instructions in text, revealing a gap between language and visual representation. Interestingly, we found that these models are not entirely blind to numbers-they are implicitly aware of their own counting accuracy, as their internal signals shift in consistent ways depending on whether the output meets the specified count. This observation suggests that the model already encodes a latent notion of numerical correctness, which can be harnessed to guide generation more precisely. Building on this intuition, we introduce CountSteer, a training-free method that improves generation of specified object counts by steering the model's cross-attention hidden states during inference. In our experiments, CountSteer improved object-count accuracy by about 4% without compromising visual quality, demonstrating a simple yet effective step toward more controllable and semantically reliable text-to-image generation.

cs.CV↗