Searcharxiv⌕ Search

arXiv subjects

Danyang Zhuo

Publications and source records attributed to Danyang Zhuo.

At least 19 recordsLinked to original sources

MARS: Towards High Goodput and Low KV Thrashing in Agent Serving under GPU Memory Constraints

Self-hosted LLM agents offer greater control over sensitive data and model customization while avoiding per-request API fees. However, serving concurrent, multi-turn workflows within limited GPU memory makes it difficult to achieve high workflow-level goodput. In particular, as workflows accumulate context, their reusable key-value (KV) caches can easily outgrow GPU capacity, triggering repeated eviction and recovery, known as KV cache thrashing. This process consumes resources and delays workflow completion. Although existing approaches propose various methods, our motivation study shows that the evaluated baselines still incur substantial thrashing overhead or fail to address the workflow-level goodput issue. We present MARS, an agent serving system that coordinates workflow admission and in-engine scheduling to achieve high workflow-level goodput on memory-constrained devices while avoiding KV cache thrashing. Specifically, outside the engine, an admission controller selects root workflows using initial prefix footprint estimates and live KV availability, adapting admissions to memory pressure. Inside, an agent-centric scheduler prioritizes turn completion and coordinates prefill allocation with KV reclamation to reduce disruptive evictions. Across two models and three GPU configurations, MARS achieves 1.59x-7.56x mean end-to-end latency speedups over the fastest baseline. Integrated as the serving backend for OpenCode, Codex, and OpenHands, MARS further improves mean workflow completion time by 1.17x-1.87x. Our source code is publicly available at https://github.com/Afterglow231/MARS_preview .

cs.OS↗

Tideway: Phase-Aware Tiered Memory Management for MoE Inference on Consumer Devices

Large language model (LLM) agents are rapidly becoming a dominant paradigm for AI applications. Recently, there is a growing demand to deploy large-scale Mixture-of-Experts (MoE) models locally on consumer PCs to balance data privacy, API costs, and inference efficiency. However, unlike datacenter servers with abundant DRAM, consumer devices are tightly constrained in both GPU VRAM and host RAM, forcing massive MoE weights to spill across VRAM, host RAM, and SSD. Under such spilling, we observe that existing systems fail to fully utilize this memory hierarchy, treating the tiers as independent caches and thrashing in inference. We present TIDEWAY, a system for efficient inference of large MoE models on memory-constrained consumer devices. Our core insight is to manage host RAM and GPU VRAM as a unified memory pool. Within this pool, TIDEWAY couples phase-aware memory layouts, which adapt to the distinct access patterns of prefill and decode, with an execution engine that optimizes CUDA Graph execution under expert paging. Our evaluations on representative consumer hardware with frontier MoE models demonstrate that TIDEWAY significantly outperforms state-of-the-art systems. On average, TIDEWAY achieves speedups of 1.97x in end-to-end latency and 2.40x in Time-to-First-Token (TTFT), along with 1.61x higher decode throughput against the fastest baseline. Our source code is publicly available at https://github.com/Afterglow231/Tideway .

cs.OS↗

Trellis: Bringing Compiler-Guided Repair into Agentic Code Generation

Coding agents regularly generate programs that do not compile and require substantial tokens to repair. Yet most repairs are small and local to the edit that caused them. We present Trellis, a system that integrates incremental compiler feedback into code generation to improve code correctness while controlling repair costs. Trellis uses rollback-based search for targeted repair and asynchronous checking and checkpointing for efficient execution. We adapt Clang and tsc to Trellis's checker interface and integrate Trellis transparently into OpenCode. On C/C++ and TypeScript software engineering tasks, Trellis improves compilation rate by up to 13% and test pass rate by up to 11% with comparable token consumption to baseline OpenCode.

cs.SE↗

Toward Automated Detection of Documentation Inconsistencies in Electronic Health Records

Objective:To characterize the kinds of internal documentation inconsistencies a general-domain large language model (LLM) can surface from real-world discharge summaries, and to identify recurring failure modes that limit reliability at scale. Materials and Methods: We applied a two-stage LLM pipeline---open-ended candidate identification (Gemini 2.5 Pro) followed by context-grounded verification (Gemini 2.5 Flash)---to 3000 MIMIC-IV-Note discharge summaries drawn as three contiguous 1000-note batches from distinct positions in the corpus. Two clinicians independently reviewed a domain-stratified sample of 268 retained candidates; a third adjudicated disagreements. Results: Stage~1 proposed 11333 candidates; 3461 were retained after Stage~2 verification---machine-generated candidates---with at least one in 69.7% of admissions. Medication (43.2%) and diagnosis or clinical-finding (25.3%) candidates predominated. In the reviewed sample, reviewers agreed on 91.0\% of candidates, and 253 of 268 (94.4\%) were judged valid after adjudication. Representative examples from all seven domains carried direct implications for clinical reasoning or patient safety. Review also revealed recurring failure modes involving temporal reasoning, evolving diagnoses, and outpatient-prescribing conventions. Discussion: Detection is highly context-dependent: judging a flagged pair requires locating each statement and assessing whether the conflict reflects a true contradiction or missing context. We propose a graded ontology spanning strict contradiction, ambiguity, and gap, with a schema characterizing each flagged case by type, section, scope, domain, attribute, and inconsistency axis. Conclusion: This formative study establishes a methodological foundation and conceptual framework to guide subsequent validated, large-scale EHR-inconsistency analysis.

cs.CL↗

Beaver: Elastic GPU Sharing between ML and Latency-Critical vRAN Workloads

Within the shared industry vision of AI-RAN, AI-and-RAN seeks to co-locate virtualized radio access network (vRAN) workloads and AI services on shared GPUs. This sharing is inherently asymmetric: vRAN workload is latency-critical, whereas the machine learning (ML) workload is a throughput-oriented, best-effort co-tenant. We present Beaver, a GPU sharing system that jointly manages compute and memory resources while protecting the vRAN's strict processing deadline. Beaver sizes the vRAN's streaming multiprocessor (SM) allocation from each slot's scheduled workload, repartitions SM allocations at slot granularity, and rewrites compiled ML kernels to yield HBM bandwidth during the vRAN's memory-critical phases. We implement Beaver and evaluate it using NVIDIA Aerial with heterogeneous multi-cell workloads, real-world cellular traces, production inference kernels, and full-stack LLM serving. On an H200 GPU, Beaver keeps the vRAN's p99.9 latency within its 1.5ms uplink deadline while retaining 74% of Llama-3.3-70B serving throughput. It also incurs no observed deadline misses under replayed cellular traces, protects a 375us downlink deadline, and generalizes to other GPUs including A100, GB10 and GH200.

cs.DC↗

Vosti: Specifying, Implementing, and Verifying Deterministic LLM Inference

LLM inference systems may vary batch composition, prompt chunking, prefill/decode execution, and KV-cache reuse, eviction, or recomputation. These optimizations should not affect system outputs. Production systems, including vLLM's batch-invariant mode and SGLang's deterministic mode, target this goal but lack a formal system-level specification. We formalize deterministic LLM inference: under a fixed model and deployment configuration, requests with the same prompt and initial sampler state produce bitwise-identical logits at corresponding output positions across executions. Our tests find that these production modes produce different logits under some execution variations. To address this limitation, we present Vosti, an inference engine designed and verified against this specification. Vosti chooses kernels independently of runtime engine state and ties cached KV values to their logical token prefixes. Its proof decomposes at the engine/GPU kernel boundary: a Verus inductive proof establishes that scheduling and the paged, prefix-sharing KV-cache preserve output logits, while a Triton analyzer proves bitwise-equal selected kernel outputs across batches, query lengths, and paged KV-cache layouts. Vosti produces bitwise-identical logits across every tested execution variation and achieves performance comparable to vLLM's batch-invariant mode on decode-heavy workloads, while providing a stronger, formally verified determinism guarantee.

cs.DC↗

Agentic AI for Scalable and Robust Optical Systems Control

We present AgentOptics, an agentic AI framework for high-fidelity, autonomous optical system control built on the Model Context Protocol (MCP). AgentOptics interprets natural language tasks and executes protocol-compliant actions on heterogeneous optical devices through a structured tool abstraction layer. We implement 64 standardized MCP tools across 8 representative optical devices and construct a 410-task benchmark to evaluate request understanding, role-aware responses, multi-step coordination, robustness to linguistic variation, and error handling. We assess two deployment configurations--commercial online LLMs and locally hosted open-source LLMs--and compare them with LLM-based code generation baselines. AgentOptics achieves 87.7%--99.0% average task success rates, significantly outperforming code-generation approaches, which reach up to 50% success. We further demonstrate broader applicability through five case studies extending beyond device-level control to system orchestration, monitoring, and closed-loop optimization. These include DWDM link provisioning and coordinated monitoring of coherent 400 GbE and analog radio-over-fiber (ARoF) channels; autonomous characterization and bias optimization of a wideband ARoF link carrying 5G fronthaul traffic; multi-span channel provisioning with launch power optimization; closed-loop fiber polarization stabilization; and distributed acoustic sensing (DAS)-based fiber monitoring with LLM-assisted event detection. These results establish AgentOptics as a scalable, robust paradigm for autonomous control and orchestration of heterogeneous optical systems.

eess.SY↗

LowRankArena: A Standardized Evaluation Platform for SVD-Based LLM Compression

SVD-based low-rank compression has become a fast-growing direction for reducing the memory and computational cost of large language models (LLMs). However, meaningful comparison across existing studies remains difficult as prior evaluations use varied benchmarks, inconsistent ratios, and diverse setups, often failing to isolate low-rank effects from auxiliary techniques. As a result, it remains unclear whether reported gains reflect method-level improvements or differences in evaluation protocol. This lack of comparability highlights the need for a unified, reproducible evaluation platform. To address this problem, we present LowRankArena, a standardized evaluation platform for SVD-based LLM compression. LowRankArena unifies task versions, uniform-precision compression budgets, comparison regimes, and inference measurements, and provides a reproducible pipeline with over 3 TiB released compressed checkpoints. Using LowRankArena, our aligned audit of five representative SVD methods reveals that prior findings are highly conditional under standardized protocols: clear leaders and performance tiers shift across backbones and keep ratios, multiple-choice accuracy can hide large perplexity degradation, and nominal low-rank savings yield workload-dependent and often limited end-to-end speedups. Our code is available at: https://github.com/Zishan-Shao/lowrankarena.git.

cs.CL↗

Iteration Without Elaboration: A Simple ReAct Architecture Suffices for Text-to-SQL Generation

Modern text-to-SQL systems have become increasingly elaborate, relying on schema-linking modules, retrieval-augmented prompting, candidate generation, and multi-stage refinement pipelines. While effective, these additions introduce substantial latency and engineering overhead. To this end, we present \textbf{ReAct-SQL}, a simple yet effective zero-shot ReAct-style framework built solely on iterative reasoning and a constrained action space defined by a typed Domain-Specific Language (DSL) of 15 relational operations, rather than free-form SQL generation. The model incrementally issues DSL calls, observes compiled-SQL execution feedback, and revises its reasoning through interaction. On corrected BIRD mini-dev and EHR-SQL, ReAct-SQL achieves \textbf{84.5\%} and \textbf{73.9\%} accuracy, respectively, matching substantially more elaborate baselines while running up to $8\times$ faster. Incremental ablations further show that iteration primarily improves grounding, while the DSL improves compositional reliability.

cs.CL↗

InfoFlow KV: Information-Flow-Aware KV Recomputation for Long Context

Retrieval-augmented generation (RAG) for long-context question answering is bottlenecked by inference-time prefilling over large retrieved contexts. A common strategy is to precompute key-value (KV) caches for individual documents and selectively recompute a small subset of tokens to restore global causal dependencies, but existing methods rely on heuristics or representation discrepancies without modeling whether selected tokens can effectively influence generation. We cast selective KV recomputation as an information flow problem and show that a simple attention-norm signal from the query reliably identifies tokens that are both semantically relevant and structurally positioned to propagate information, when computed under an inference-consistent RoPE geometry. We therefore reconstruct global positional assignments for retrieved chunks and introduce an information-flow-guided chunk reordering strategy. Experiments on Large Language Model and Vision-Language Model benchmarks demonstrate consistent gains over prior methods under comparable latency.

cs.LG↗

Optimus: Elastic Decoding for Efficient Diffusion LLM Serving

Large language model (LLM) serving is fundamentally limited by inefficient hardware utilization. Autoregressive (AR) decoding underutilizes GPUs due to its strictly sequential execution, while diffusion LLMs (DLLMs) improve throughput by decoding multiple tokens per iteration. However, fixed block-size diffusion decoding exhibits strong load sensitivity: large blocks exploit idle GPU resources under low load, but saturate early and incur substantial redundant computation under high load. As a result, throughput gains vanish beyond saturation, and no single decoding granularity performs well across dynamic serving workloads. We present Optimus, a serving system that enables elastic decoding for diffusion LLMs by dynamically adapting decoding granularity to runtime load. The key idea is to treat decoding granularity as a runtime control variable, balancing GPU utilization and token efficiency. Optimus combines chunked decoding, which enables fine-grained execution without retraining, with saturation-aware scheduling, a closed-loop mechanism that selects chunk sizes based on runtime conditions. Together with system-level optimizations and customized attention kernels, Optimus achieves significant performance improvements while preserving model accuracy. Experiments show that Optimus delivers up to 6.1x throughput improvement over AR decoding and 4.3x improvement over fixed-block diffusion LLM, while maintaining stable performance across diverse load regimes and improving end-to-end serving capacity under latency constraints. The source code is available at https://github.com/dubcyfor3/Optimus.

cs.DC↗

DecodeShare: Tracing the Shared Subspace of LLM Decode-Time Decisions

Large language models (LLMs) handle many tasks with one set of parameters, but under KV-cached inference it is unclear what task-general structure, if any, is used at decode time rather than during prefill. We propose DecodeShare, a protocol that identifies a low-dimensional subspace consistently shared across tasks in decode-time hidden states, and then tests its causal role by removing that subspace only during decoding. In our experiments, disturbing the discovered shared subspace degrades decision performance far more than disturbing either a prefill-derived or random subspace under the same intervention budget. We further show this decode-shared subspace has practical consequences for activation steering: common steering directions can overlap the task-general decode channel. Projecting out this shared subspace directly separates the functional roles of the two components, while evaluating steering vectors at decode-time yields more reliable signal for downstream deployment than prefill-based proxies. Despite its compactness, the shared subspace can serve as a high-leverage causal channel at decode time. Code is available at: https://github.com/Zishan-Shao/decodeshare.git.

cs.AI↗

FlashSVD v1.5: Making Low-Rank Transformers Inference Actually Fast

SVD-based Low-rank compression reduces transformer parameters and nominal FLOPs, but these savings often translate poorly into real LLM serving speedups. We show that this gap is largely a runtime problem: factorized checkpoints fragment execution paths, and the resulting overhead differs substantially between prefill and autoregressive decode. We present FlashSVD v1.5, a unified inference runtime for serving SVD-compressed transformers. FlashSVD v1.5 maps diverse public SVD compression families to a common factorized representation and combines phase-specific kernels with dense-KV decode, packed MLP execution, and per-layer CUDA-graph replay to reorganize the low-rank serving path into a thin runtime. Across representative decoder-serving settings, FlashSVD v1.5 achieves up to 2.55x decode and 2.39x end-to-end speedup, and it attains 1.48x average decode and 1.44x average end-to-end speedup across multiple popular SVD compression families. These results suggest that practical low-rank acceleration requires runtime co-design, not compression algorithms alone. Our code is available at: https://github.com/Zishan-Shao/FlashSVD.

cs.LG↗

Parallel Prefix Verification for Speculative Generation

We introduce PARSE (PArallel pRefix Speculative Engine), a speculative generation framework that accelerates large language model (LLM) inference by parallelizing prefix verification on a semantic level. Existing speculative decoding methods are fundamentally limited by token-level equivalence: the target model must verify each token, leading to short acceptance lengths and modest speedups. Moving to semantic or segment-level verification can substantially increase acceptance granularity, but prior approaches rely on sequential verification, introducing significant overhead and limiting practical gains. PARSE introduces parallel prefix verification, enabling semantic-level verification without sequential checks. Given a full draft from a draft model, the target model evaluates correctness across multiple prefixes in a single forward pass using a custom attention mask, directly identifying the maximal valid prefix. This eliminates sequential segment verification, and makes verification compute-efficient. PARSE is orthogonal to token-level speculative decoding and can be composed with it for additional gains. Across models and benchmarks, PARSE delivers $1.25\times$ to $4.3\times$ throughput gain over the target model, and $1.6\times$ to $4.5\times$ when composed with EAGLE-3, all with negligible accuracy degradation. This demonstrates parallel prefix verification as an effective, general approach to accelerating LLM inference.

cs.AI↗

Feature Anchors for Time-Series Sensor-Based Human Activity Recognition

Wearable Human Activity Recognition (HAR) still lacks a representation that is both explicit and adaptable. Handcrafted time-series features (TSFs) capture meaningful motion statistics and remain competitive on standard benchmarks, but they are usually used as fixed preprocessing outputs. Deep models learn adaptable representations directly from raw signals, but those representations are typically latent and difficult to inspect. We address this gap by treating handcrafted TSFs as feature anchors: explicit intermediate representations that remain inside the model and are adjusted by neural context instead of being discarded. We propose the Temporal Conditioning Network for Feature Anchors (TCNet), which extracts handcrafted anchors, encodes complementary time-domain and frequency-domain context from raw IMU windows, and predicts context-conditioned scale, bias, and gating parameters to modulate anchor groups directly in feature space. This design keeps anchor semantics visible while allowing the representation to adapt to the classification objective. Across five HAR benchmarks, TCNet achieves 70.2% mF1 on USC-HAD, 85.1% mF1 on Daphnet, 93.9% mF1 on MHealth, and 94.5% mF1 on PAMAP2. Relative to rTsfNet, it improves by 4.5 points on USC-HAD, 14.6 points on Daphnet, and 6.5 points on MHealth. Ablations show that the gains come primarily from anchor guidance rather than simple branch fusion, and feature-space analyses indicate that several discriminative TSF families are not reliably accessible in standard latent representations. These results suggest that, for HAR, handcrafted TSFs are most useful when they remain explicit and adaptable within the model. The code is available at: https://github.com/ni-x-lab/TCNet-har

cs.HC↗

Foundry: Template-Based CUDA Graph Context Materialization for Fast LLM Serving Cold Start

Modern LLM service providers increasingly rely on autoscaling and parallelism reconfiguration to respond to rapidly changing workloads, but cold-start latency remains a major bottleneck. While recent systems have reduced model weight loading to seconds, CUDA graph capture still takes tens of seconds to minutes and often dominates startup. Unfortunately, CUDA graphs cannot be naively serialized: beyond graph topology, they are tightly coupled to execution context, including device addresses embedded in kernel arguments and kernel code lazily loaded during warmup. Existing approaches either rely on brittle kernel-specific patching or heavyweight process-level checkpoint/restore that are inflexible to dynamic parallelism switching. We present Foundry, a template-based CUDA graph context materialization system that persists both graph topology and execution context during an offline processing stage, and reconstructs executable graphs online with negligible overhead. Foundry enforces deterministic memory layouts, automatically extracts and reloads kernel binaries required by captured graphs, and reduces online reconstruction costs through topology-based templating. For distributed serving, Foundry further enables a single-GPU offline capture to generate templates for multi-GPU deployments by patching only rank-dependent communication state. Across dense and MoE models up to 235B parameters, Foundry reduces cold-start latency by up to 99%, cutting the initialization time of Qwen3-235B-A22B from 10 minutes to 3.9 seconds while preserving the throughput gains of CUDA graphs.

cs.DC↗

Real-Time and Scalable Zak-OTFS Receiver Processing on GPUs

Orthogonal time frequency space (OTFS) modulation offers superior robustness to high-mobility channels compared to conventional orthogonal frequency-division multiplexing (OFDM) waveforms. However, its explicit delay-Doppler (DD) domain representation incurs substantial signal processing complexity, especially with increased DD domain grid sizes. To address this challenge, we present a scalable, real-time Zak-OTFS receiver architecture on GPUs through hardware--algorithm co-design that exploits DD-domain channel sparsity. Our design leverages compact matrix operations for key processing stages, a branchless iterative equalizer, and a structured sparse channel matrix of the DD domain channel matrix to significantly reduce computational and memory overhead. These optimizations enable low-latency processing that consistently meets the 99.9-th percentile real-time processing deadline. The proposed system achieves up to 906.52 Mbps throughput with a DD grid size of (16384,32) using 16QAM modulation over 245.76 MHz bandwidth. Extensive evaluations under a Vehicular-A channel model demonstrate strong scalability and robust performance across CPU (Intel Xeon) and multiple GPU platforms (NVIDIA Jetson Orin, RTX 6000 Ada, A100, and H200), highlighting the effectiveness of compute-aware Zak-OTFS receiver design for next-generation (NextG) high-mobility communication systems.

eess.SP↗

AXLearn: Modular, Hardware-Agnostic Large Model Training

AXLearn is a production system which facilitates scalable and high-performance training of large deep learning models. Compared to other state-of-art deep learning systems, AXLearn has a unique focus on modularity and support for hardware-agnostic training. AXLearn's internal interfaces between software components follow strict encapsulation, allowing different components to be assembled to facilitate rapid model development and experimentation on different hardware infrastructure. AXLearn maintains constant complexity as we scale the components in the system, compared to linear or quadratic complexity in state-of-the-art training systems. This allows integrating features such as Rotary Position Embeddings (RoPE) into AXLearn across hundred of modules with just 10 lines of code, compared to hundreds as required in other systems. At the same time, AXLearn maintains equivalent performance compared to state-of-the-art training systems. Finally, we share our experience in the development and operation of AXLearn at Apple.

cs.LG↗