arXiv subjects

Baris Kasikci

Publications and source records attributed to Baris Kasikci.

At least 19 recordsLinked to original sources

arXiv2026

Themis: Software-Defined Hardware Prefetching

Data cache misses represent a significant portion of stall cycles in datacenter workloads. Hardware prefetchers that reduce such stalls by fetching data ahead of time have become increasingly sophisticated. However, to achieve high coverage, they have to prefetch aggressively, generating many inaccurate accesses that waste memory bandwidth. This is problematic in datacenter environments where memory bandwidth is a limited resource due to high multi-tenancy. We observe that for datacenter workloads, inaccurate prefetches can be effectively filtered on a data page granularity, without sacrificing prefetch coverage. However, storing per-page metadata about prefetch usefulness in hardware is costly, so we propose a novel hardware-software interface for data prefetching: The software directs the hardware on where to prefetch, and the hardware identifies and issues prefetches in the regions of interest. We propose Themis, a profile-guided hardware prefetching solution that implements this new interface. Themis utilizes page-level hints stored in page-table entries to disable the prefetcher for certain data pages at runtime. Themis requires no binary or ISA changes and can be used to optimize processes without disrupting their execution. Themis is also orthogonal to existing works on prefetching and can be applied to optimize any hardware prefetcher. Our results show that Themis is able to achieve around 40% reduction in useless prefetch requests, resulting in speedup for all the evaluated prefetchers for datacenter workloads, including 4.1% for BOP, 3.1% for SPP+PPF, and 1.4% for Pythia.

cs.AR↗

arXiv2026

TraceLab: Characterizing Coding Agent Workloads for LLM Serving

Coding agents are rapidly becoming a major application of agentic LLMs, but serving them efficiently remains challenging. Progress on this challenge requires understanding real workload patterns, yet the data needed for such analysis is largely absent. Existing public traces and benchmarks do not capture real, day-to-day coding-agent usage across multiple agents and model families for serving-system analysis. To help fill this gap, we collect and release a trace of roughly 4,300 coding-agent sessions, containing about 350,000 LLM steps and 430,000 tool calls from our own day-to-day use of Claude Code and Codex. Our analysis shows that coding-agent workloads feature long autonomous loops, long contexts with short outputs, diverse and heavily-tailed tool calls, and high but imperfect prefix cache hit rates. These findings point to concrete opportunities for optimizing serving, including lower-overhead tool calling, append-length-aware prefill, semantic-aware tool-latency prediction, and improved KV-cache management around human-paced gaps. We release the dataset, trace collection pipeline, and analysis code at https://github.com/uw-syfi/TraceLab.git the project website is https://tracelab.cs.washington.edu.

cs.LG↗

arXiv2026

M*: A Modular, Extensible, Serving System for Multimodal Models

We are entering a new era of composite model architectures that integrate diverse components such as vision encoders, language backbones, diffusion and flow heads, audio codecs, action generators, and world-model predictors. Such architectures underpin a broad class of multimodal models, including unified multimodal models, omni models, speech-language models, vision-language-action policies, and world models. However, existing model serving frameworks were built on narrow assumptions about model structure, making them ill-suited to accommodate this new architectural diversity. Here we present M*, a universal serving system for efficient serving of composite AI models. M* represents models as dataflow graphs, processing requests spanning diverse modalities and tasks as traversals over these graphs. The core insight is a modular abstraction that supports arbitrary composition of model components, flexible placement onto a physical cluster, and model-agnostic optimizations within a distributed runtime. We call this abstraction the Walk Graph and show how it can concisely capture composite models from a broad range of families. We instantiate M* on representative models and find that it achieves, on average, 20% lower end-to-end latency than vLLM-Omni for text-to-image workloads on BAGEL, while delivering up to 2.9x lower real-time factor and 2.7x higher throughput for text-to-speech workloads on Qwen3-Omni. M* also outperforms the V-JEPA 2-AC rollout baseline for robotic planning by up to 12.5x. Thus, our work paves the road towards more efficient serving of complex models with minimal developer effort.

cs.LG↗

arXiv2026

BlendServe: Optimizing Offline Inference for Auto-regressive Large Models with Resource-aware Batching

Offline batch inference, which leverages the flexibility of request batching to achieve higher throughput and lower costs, is becoming more popular for latency-insensitive applications. Meanwhile, recent progress in model capability and modality makes requests more diverse in compute and memory demands, creating unique opportunities for throughput improvement by resource overlapping. However, a request schedule that maximizes resource overlapping can conflict with the schedule that maximizes prefix sharing, a widely-used performance optimization, causing sub-optimal inference throughput. We present BlendServe, a system that maximizes resource utilization of offline batch inference by combining the benefits of resource overlapping and prefix sharing using a resource-aware prefix tree. BlendServe exploits the relaxed latency requirements in offline batch inference to reorder and overlap requests with varied resource demands while ensuring high prefix sharing. We evaluate BlendServe on a variety of synthetic multi-modal workloads and show that it provides up to $1.44\times$ throughput boost compared to widely-used industry standards, vLLM and SGLang.

cs.LG↗

arXiv2026

Ekka: Automated Diagnosis of Silent Errors in LLM Inference

LLM serving frameworks are quickly evolving with a complex software stack and a vast number of optimizations. The rapid development process can introduce silent errors where output quality silently degrades without any explicit error signals. Diagnosing silent errors is notoriously difficult due to the substantial semantic gap between the high-level symptoms and the low-level root causes. We observe that diagnosis of silent errors can be effectively framed as a differential debugging problem by leveraging the existence of semantically correct reference implementations. We propose Ekka, an automated diagnosis system that identifies root causes by systematically aligning and comparing intermediate execution states between a target and a reference framework. We constructed a benchmark of real-world silent errors from popular serving frameworks, where Ekka shows 80% pass@1 diagnosis accuracy and 88% pass@5 diagnosis accuracy, outperforming state-of-the-art systems. Ekka also diagnoses 4 new silent errors from serving frameworks, all of which have been confirmed by the developers.

cs.DC↗

arXiv2026

MURMUR: An Efficient Inference System for Long-Form ASR

Long-form automatic speech recognition (ASR) requires both high accuracy and low latency, but existing systems force a trade-off between the two. Chunk-based pipelines process audio in parallel windows for low latency, but lose cross-chunk context and need brittle heuristics to align speakers and timestamps at boundaries. Long-context ASR models resolve everything in a single pass for better accuracy, but are an order of magnitude slower. We propose Murmur, an inference system that overcomes this trade-off by operating at two levels. At the inter-chunk level, we revisit the chunk-based pipeline for modern long-context ASR, treating chunk size as a tunable hyperparameter, and show that intermediate chunk sizes strike a good balance of accuracy and latency. At the intra-chunk level, we exploit attention sparsity through a sliding window KV cache eviction policy applied to both output and speech tokens. On AMI-IHM, Murmur matches single-pass accuracy while reducing latency by 4.2x, with further gains from token eviction at less than 1% relative tcpWER degradation. The code of Murmur is available at https://github.com/uw-syfi/Murmur.

cs.LG↗

arXiv2026

DynaFlow: Transparent and Flexible Intra-Device Parallelism via Programmable Operator Scheduling

Intra-device parallelism addresses resource under-utilization in ML inference and training by overlapping the execution of operators with different resource usage. However, its wide adoption is hindered by a fundamental conflict with the static, sequential programming model of existing frameworks. Integrating these strategies requires invasive, model-specific code overhauls, representing an intractable engineering cost. This is further amplified by the high sensitivity of strategies to execution contexts (e.g., workload, model architecture, hardware), forcing developers to implement and maintain multiple specialized solutions. To address this, we propose DynaFlow, a framework that enables the transparent and flexible integration of intra-device parallelism by decoupling the logical model definition from the physical execution schedule. DynaFlow introduces a flexible frontend with annotations for graph partitioning and a programmable interface for defining custom intra-device parallelism strategies. Its efficient backend manages complex control/data-flow asynchronously, uses custom memory management to eliminate copy overheads, and preserves compatibility with optimizations like CUDA Graphs and TorchInductor. We demonstrate that DynaFlow can integrate representative parallelism strategies into 6 state-of-the-art ML systems with minimal code changes, achieving up to a 1.29x throughput improvement. DynaFlow is publicly available at https://github.com/uw-syfi/DynaFlow.

cs.DC↗

arXiv2026

TeleRAG: Efficient Retrieval-Augmented Generation Inference with Lookahead Retrieval

Retrieval-augmented generation (RAG) extends large language models (LLMs) with external data sources to enhance factual correctness and domain coverage. Modern RAG pipelines rely on large datastores, creating a significant system challenge: achieving high throughput and low latency is difficult, especially when GPU memory is limited. To address these challenges, we propose TeleRAG, an efficient inference system that reduces latency and improves throughput with minimal GPU memory requirements. The core innovation of TeleRAG is lookahead retrieval, a prefetching mechanism that predicts required data and transfers them from CPU to GPU in parallel with LLM generation. In addition, TeleRAG adopts a prefetching scheduler and a cache-aware scheduler to support efficient multi-GPU inference with minimal overhead. Evaluations show TeleRAG achieves up to a 1.53x average end-to-end latency reduction (single-query) and 1.83x higher average throughput (batched), as well as good scalability in throughput. This confirms the practical utility of TeleRAG for faster and more memory-efficient deployments of RAG applications.

cs.DC↗

arXiv2026

Unleashing Scalable Context Parallelism for Foundation Models Pre-Training via FCP

Context parallelism (CP) has been widely adopted to support the growing context length in foundation model pretraining. However, existing designs fail to handle the large variation in sequence length from training datasets, resulting in suboptimal performance. These methods often over-shard short sequences, leading to compute inefficiency and excessive communication, or process long and short sequences separately without proper bin-packing, causing workload imbalance. In this paper, we propose FCP, a flexible context parallelism paradigm that shards and schedules sequences at block-level granularity. Instead of relying on rigid communication topologies such as ring, FCP enables arbitrary peer-to-peer communication, allowing flexible placement of sequence blocks across workers. By bin-packing blocks from both short and long sequences, FCP achieves both high compute efficiency and balanced workload distribution. Extensive evaluations show that FCP attains near-linear scalability on up to 256 NVIDIA GPUs, with 1.13x-2.21x improvement in the attention MFU.

cs.DC↗

arXiv2026

VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?

For years, we have built LLM serving systems like any other critical infrastructure: a single general-purpose stack, hand-tuned over many engineer-years, meant to support every model and workload. In this paper, we take the opposite bet: a multi-agent loop that automatically synthesizes bespoke serving systems for different usage scenarios. We propose VibeServe, the first agentic loop that generates entire LLM serving stacks end-to-end. VibeServe uses an outer loop to plan and track the search over system designs, and an inner loop to implement candidates, check correctness, and measure performance on the target benchmark. In the standard deployment setting, where existing stacks are highly optimized, VibeServe remains competitive with vLLM, showing that generation-time specialization need not come at the cost of performance. More interestingly, in non-standard scenarios, VibeServe outperforms existing systems by exploiting opportunities that generic systems miss in six scenarios involving non-standard model architectures, workload knowledge, and hardware-specific optimizations. Together, these results suggest a different point in the design space for infrastructure software: generation-time specialization rather than runtime generality. Code is available at https://github.com/uw-syfi/vibe-serve.

cs.AI↗

arXiv2026

VoxServe: Streaming-Centric Serving System for Speech Language Models

Deploying modern Speech Language Models (SpeechLMs) in streaming settings requires systems that provide low latency, high throughput, and strong guarantees of streamability. Existing systems fall short of supporting diverse models flexibly and efficiently. We present VoxServe, a unified serving system for SpeechLMs that optimizes streaming performance. VoxServe introduces a model-execution abstraction that decouples model architecture from system-level optimizations, thereby enabling support for diverse SpeechLM architectures within a single framework. Building on this abstraction, VoxServe implements streaming-aware scheduling and an asynchronous inference pipeline to improve end-to-end efficiency. Evaluations across multiple modern SpeechLMs show that VoxServe achieves 10-20x higher throughput than existing implementations at comparable latency while maintaining high streaming viability. The code of VoxServe is available at https://github.com/vox-serve/vox-serve.

cs.LG↗

arXiv2025

Magneton: Optimizing Energy Efficiency of ML Systems via Differential Energy Debugging

The training and deployment of machine learning (ML) models have become extremely energy-intensive. While existing optimization efforts focus primarily on hardware energy efficiency, a significant but overlooked source of inefficiency is software energy waste caused by poor software design. This often includes redundant or poorly designed operations that consume more energy without improving performance. These inefficiencies arise in widely used ML frameworks and applications, yet developers often lack the visibility and tools to detect and diagnose them. We propose differential energy debugging, a novel approach that leverages the observation that competing ML systems often implement similar functionality with vastly different energy consumption. Building on this insight, we design and implement Magneton, an energy profiler that compares energy consumption between similar ML systems at the operator level and automatically pinpoints code regions and configuration choices responsible for excessive energy use. Applied to 9 popular ML systems spanning LLM inference, general ML frameworks, and image generation, Magneton detects and diagnoses 16 known cases of software energy inefficiency and further discovers 8 previously unknown cases, 7 of which have been confirmed by developers.

cs.DC↗

arXiv2025

Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding

Reasoning language models have demonstrated remarkable capabilities on challenging tasks by generating elaborate chain-of-thought (CoT) solutions. However, such lengthy generation shifts the inference bottleneck from compute-bound to memory-bound. To generate each token, the model applies full attention to all previously generated tokens, requiring memory access to an increasingly large KV-Cache. Consequently, longer generations demand more memory access for every step, leading to substantial pressure on memory bandwidth. To address this, we introduce SparseSpec, a speculative decoding framework that reuses the same model as the draft and target models (i.e., self-speculation). SparseSpec features a novel sparse attention mechanism, PillarAttn, as the draft model, which accurately selects critical tokens via elegantly reusing information from the verification stage. Furthermore, SparseSpec co-designs self-speculation with three system innovations: (1) a unified scheduler to batch token drafting and verification, (2) delayed verification for CPU/GPU overlap, and (3) dynamic KV-Cache management to maximize memory utilization. Across various models and datasets, SparseSpec outperforms state-of-the-art solutions, with an up to 2.13x throughput speedup.

cs.LG↗

arXiv2025

AgentFlux: Decoupled Fine-Tuning & Inference for On-Device Agentic Systems

The deployment of Large Language Models (LLMs) as agentic orchestrators has revolutionized task automation, but the need for privacy-preserving, cost-effective solutions demands on-device inference capabilities. However, local LLMs consistently underperform compared to frontier models in tool calling scenarios, struggling with both tool selection from large tool sets and accurate argument generation for complex parameter structures. We introduce a methodology that disaggregates a tool-calling task into two distinct subtasks: tool selection and argument generation. We propose "decoupled fine-tuning", a novel post-training approach that employs LoRA fine-tuning to create dedicated LoRA adapters for tool selection and tool-specific argument generation using separate loss masking for each of the subtasks. Furthermore, we present AgentFlux, an inference framework that leverages the LoRA adapters created using decoupled fine-tuning to perform efficient agent orchestration with the help of local models on end-user devices. AgentFlux decomposes the tool-call generation step into tool selection and argument generation, and dynamically loads the corresponding LoRA adapters to generate tool calls. Additionally, AgentFlux implements hierarchical orchestration to restrict the number of tools required for tool selection. Our experiments on the MCP-Bench benchmark demonstrate that the Qwen-2.5-7B model trained using decoupled fine-tuning improves the tool calling accuracy of the base model by 46%, and outperforms other local reasoning, non-reasoning and fine-tuned models of similar size in all cases, and models that are 2x larger, in most cases.

cs.AI↗

arXiv2025

Jenga: Responsive Tiered Memory Management without Thrashing

A heterogeneous memory has a single address space with fast access to some addresses (a fast tier of DRAM) and slow access to other addresses (a capacity tier of CXL-attached memory or NVM). A tiered memory system aims to maximize the number of accesses to the fast tier via page migrations between the fast and capacity tiers. Unfortunately, previous tiered memory systems can perform poorly due to (1) allocating hot and cold objects in the same page and (2) abrupt changes in hotness measurements that lead to thrashing. This paper presents Jenga, a tiered memory system that addresses both problems. Jenga's memory allocator uses a novel context-based page allocation strategy. Jenga's accurate measurements of page hotness enable it to react to memory access behavior changes in a timely manner while avoiding thrashing. Compared to the best previous tiered memory system, Jenga runs memory-intensive applications 28% faster across 10 applications, when the fast tier capacity matches the working set size, at a CPU overhead of <3% of a single core and a memory overhead of <0.3%

cs.ET↗

arXiv2025

LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation

Modern automatic speech recognition (ASR) models, such as OpenAI's Whisper, rely on deep encoder-decoder architectures, and their encoders are a critical bottleneck for efficient deployment due to high computational intensity. We introduce LiteASR, a low-rank compression scheme for ASR encoders that significantly reduces inference costs while maintaining transcription accuracy. Our approach leverages the strong low-rank properties observed in intermediate activations: by applying principal component analysis (PCA) with a small calibration dataset, we approximate linear transformations with a chain of low-rank matrix multiplications, and further optimize self-attention to work in reduced dimensionality. Evaluation results show that our method can compress Whisper large-v3's encoder size by over 50%, matching Whisper medium's size with better transcription accuracy, thereby establishing a new Pareto frontier of accuracy and efficiency. The code of LiteASR is available at https://github.com/efeslab/LiteASR.

cs.LG↗

arXiv2025

PolyServe: Efficient Multi-SLO Serving at Scale

Advances in Large Language Models (LLMs) have led to a surge of LLM-powered applications. These applications have diverse token-generation latency requirements. As a result, simply classifying workloads as latency-sensitive (LS) or best-effort (BE) overlooks the nuances within the latency-sensitive category and results in suboptimal user experiences and scheduling opportunities. However, efficiently serving requests with multiple SLO requirements poses significant challenges. First, all requests within a batch generate new tokens simultaneously, which can misalign them with their distinct SLO requirements. Moreover, while existing systems focus on auto-scaling for handling various overall request rates, the diversity of SLOs necessitates fine-grained auto-scaling among these SLO tiers. Finally, unlike LS/BE scenarios, where BE requests can be aborted at any time to ensure the SLO attainment of LS requests, those with different latency-sensitive SLOs cannot tolerate prolonged delays, and tail latency must be controlled. To tackle these challenges, we propose PolyServe, a novel multi-SLO scheduling policy at scale that maintains high SLO attainment while maximizing throughput. PolyServe first groups requests into multiple bins based on their per-token latency requirement, then schedules each bin to a subset of the server fleet. PolyServe routes requests to the highest-load but still SLO-attainable server to create a load gradient that facilitates auto-scaling. To increase utilization, PolyServe permits looser-SLO requests to share tighter-SLO instances when their own servers are saturated. PolyServe uses profiling data to guide scheduling decisions and manage tail latency through request-wait-time-aware scheduling, dynamic chunking, and continuous chunked prefill prediction. PolyServe achieves 1.23x goodput gain compared to existing policies, achieving up to 92.5% of optimal goodput.

cs.DC↗

arXiv2025

gigiProfiler: Diagnosing Performance Issues by Uncovering Application Resource Bottlenecks

Diagnosing performance bottlenecks in modern software is essential yet challenging, particularly as applications become more complex and rely on custom resource management policies. While traditional profilers effectively identify execution bottlenecks by tracing system-level metrics, they fall short when it comes to application-level resource contention caused by waiting for application-level events. In this work, we introduce OmniResource Profiling, a performance analysis approach that integrates system-level and application-level resource tracing to diagnose resource bottlenecks comprehensively. gigiProfiler, our realization of OmniResource Profiling, uses a hybrid LLM-static analysis approach to identify application-defined resources offline and analyze their impact on performance during buggy executions to uncover the performance bottleneck. gigiProfiler then samples and records critical variables related to these bottleneck resources during buggy execution and compares their value with those from normal executions to identify the root causes. We evaluated gigiProfiler on 12 real-world performance issues across five applications. gigiProfiler accurately identified performance bottlenecks in all cases. gigiProfiler also successfully diagnosed the root causes of two newly emerged, previously undiagnosed problems, with the findings confirmed by developers.

cs.PF↗