Searcharxiv⌕ Search

arXiv subjects

Junchen Jiang

Publications and source records attributed to Junchen Jiang.

At least 19 recordsLinked to original sources
arXiv2026

Divergent-Convergent Reasoning: Scaling Test-Time Compute through Structured Solution Synthesis

Test-time compute can substantially improve Large Language Model (LLM) reasoning performance, yet how and when additional compute helps remains poorly understood. We study Divergent-Convergent Reasoning (DCR), a simple two-phase primitive consisting of an exploration phase that generates multiple candidate solutions followed by a convergent reconciliation phase. We present three core results. First, we show that even a single reconciliation step can reliably amplify correct minority reports: across datasets, DCR often recovers the correct answer when correct exploration outputs are in the minority, a regime where majority voting fails. Second, we introduce recursive DCR, an autoregressive reconciliation system that iteratively analyzes disagreements and allocates additional test-time compute. Recursive DCR achieves higher accuracy than fixed-compute baselines-reaching 93.3% on AIME 2024 and 92.0% on AIME 2025-while using roughly 27% less compute on average, demonstrating that attentive resource allocation is superior to uniform scaling. Third, we analyze disagreement among exploration outputs via a simple, training-free dispersion metric. Dispersion reveals a structured relationship between disagreement and test-time gains: in regimes where DCR is effective, higher disagreement among exploration outputs is associated with larger accuracy improvements from reconciliation. Together, these results show that disagreement, often viewed as noise, can be systematically exploited to improve test-time reasoning and reveal emerging scaling laws for agentic LLM systems.

cs.AI↗
arXiv2026

An Internet for the KV Cache: Rethinking Classical Infrastructure Boundaries in the LLM Inference Age

LLM inference has become a global-scale, heterogeneous workload spanning agents, retrieval, tool-use, code execution and multi-modal reasoning. These workloads naturally enable context reuse from overlapping inputs, creating a major opportunity to store and reuse the contexts' KV Caches instead of recomputing them. However, model-side advances that shrink the KV Cache and system-side advances that reduce compute, storage, and transfer costs are evolve independently within legacy cloud boundaries. We argue that future inference infrastructure should allow decoupling of compute and KV Cache storage across cloud and datacenters. The network becomes an active distribution channel; bandwidth, latency and pricing directly determines how the KV Cache should be managed. We propose a vision for an Internet for the KV Cache, with KV Cache management working as a content-distribution system. In this view, KV Cache storage and recompute decisions are driven by model, infrastructure, and application metrics, to enable adaptive, content-driven decisions for minimizing latency and cost.

cs.NI↗
arXiv2026

Learning Agent Execution for KV-Cache Management in Agentic Serving

Multi-agent LLM systems have emerged as an important deployment paradigm for AI services, where each user request is decomposed into a sequence of specialized agents. Across these workflows, every agent repeatedly executes a fixed context consisting of system prompts, tool definitions, and few-shot examples, creating substantial opportunities for KV-cache reuse. Existing LLM serving systems, however, manage KV-cache reactively using prefix caching and recency-based replacement, causing reusable agent contexts to be evicted before their next invocation and forcing repeated recomputation. We present CacheScout, an agent-aware KV-cache runtime layer for multi-agent LLM serving. The key insight is that future KV-cache reuse is governed by agent execution semantics rather than cache recency alone. CacheScout captures these semantics by learning agent execution transitions online, without requiring predefined workflow graphs or offline training, and uses the learned execution model to guide both cache eviction and proactive prefetching while leaving the serving critical path unchanged. We implement CacheScout on top of vLLM. Across representative real-world multi-agent workloads, CacheScout improves KV-cache hit rate by 10-18 percentage points, reduces mean TTFT by 18-45%, lowers mean per-turn latency by 29-38%, and increases peak throughput by up to 57%. These benefits also generalize to larger models, reducing TTFT by up to 54% while sustaining 37% higher throughput.

cs.AI↗
arXiv2026

Argo: Efficient Importance Labeling for Enterprise Email Systems

Email importance labeling has long been a critical yet challenging problem for businesses and individuals. Traditional approaches; such as keyword matching, user-defined rules, and sender-based heuristics; demand extensive manual feature engineering and fail to scale effectively or generalize. Recent advances in large language models (LLMs) demonstrate strong potential and a natural fit for this task, offering deep contextual understanding and superior labeling quality. However, using LLM models like GPT-4.1 at enterprise email volumes incurs prohibitive computational costs and hinders real-world deployment. We explore the trade-off space of using alternative labeling schemes as opposed to GPT4.1 scale LLMs, with the goal of achieving near GPT level labeling quality with significantly lower cost. We develop Argo, an enterprise email labeling framework, where we construct a profiler to efficiently search the cost quality trade-off space of labeling and identify cost-efficient alternatives to labeling emails. Additionally, we design an on-demand provisioning scheme to intelligently scale Argo with real time load, to minimize cost increases during peak load inference. Over 3 open-source email datasets, Argo achieves 148-167X inference cost reduction with negligible quality degradation and 20-640000X lower profiling costs, making large-scale, context-aware email labeling practical for enterprises.

cs.MA↗
arXiv2026

VeriCache: Turning Lossy KV Cache into Lossless LLM Inference

The large size of the KV cache has become a major bottleneck for serving LLMs with increasing context lengths. In response, many KV cache compression methods, such as token dropping and quantization, have been proposed. However, almost all of these methods are inherently lossy-despite minimal accuracy degradation for short outputs, their outputs increasingly diverge from full-KV-cache outputs as more tokens are decoded, which leads to catastrophic failures in code generation and tool calling. We present VeriCache, the first inference framework that ensures the same output as full-KV-cache decoding but largely preserves the high decoding throughput of a range of KV cache compression algorithms. VeriCache uses the compressed KV cache to draft tokens, then verifies them against the full KV cache. While it may seem like just speculative decoding, VeriCache requires addressing a key system challenge to work-keeping the full KV cache out of GPU memory and minimizing the overhead of swapping it in for verification. The insight is two-fold: (1) compressed-KV decoding can be parallelized with full-KV swap, because one is HBM-bandwidth-bound and the other is PCIe/network-bound, and (2) the compressed KV cache often produces output similar to the full KV cache, allowing a long drafting horizon to amortize each full-KV swap. VeriCache applies to both long-context decoding and remote prefix caching, supports a broad family of token-dropping and quantization methods through a uniform compressor interface, and composes with traditional speculative decoding. Experimental results show that VeriCache achieves up to 4X higher throughput than full-KV inference while producing identical outputs.

cs.AR↗
arXiv2026

The Workload-Router-Pool Architecture for LLM Inference Optimization: A Vision Paper from the vLLM Semantic Router Project

Over the past year, the vLLM Semantic Router project has released a series of work spanning: (1) core routing mechanisms -- signal-driven routing, context-length pool routing, router performance engineering, policy conflict detection, low-latency embedding models, category-aware semantic caching, user-feedback-driven routing adaptation, hallucination detection, and hierarchical content-safety classification for privacy and jailbreak protection; (2) fleet optimization -- fleet provisioning and energy-efficiency analysis; (3) agentic and multimodal routing -- multimodal agent routing, tool selection, CUA security, and multi-turn context memory and safety; (4) governance and standards -- inference routing protocols and multi-provider API extensions. Each paper tackled a specific problem in LLM inference, but the problems are not independent; for example, fleet provisioning depends on the routing policy, which depends on the workload mix, shifting as organizations adopt agentic and multimodal workloads. This paper distills those results into the Workload-Router-Pool (WRP) architecture, a three-dimensional framework for LLM inference optimization. Workload characterizes what the fleet serves (chat vs. agent, single-turn vs. multi-turn, warm vs. cold, prefill-heavy vs. decode-heavy). Router determines how each request is dispatched (static semantic rules, online bandit adaptation, RL-based model selection, quality-aware cascading). Pool defines where inference runs (homogeneous vs. heterogeneous GPU, disaggregated prefill/decode, KV-cache topology). We map our prior work onto a 3x3 WRP interaction matrix, identify which cells we have covered and which remain open, and propose twenty-one concrete research directions at the intersections, each grounded in our prior measurements, tiered by maturity from engineering-ready to open research.

cs.LG↗
arXiv2026

The 1/W Law: An Analytical Study of Context-Length Routing Topology and GPU Generation Gains for LLM Inference Energy Efficiency

How many tokens can a GPU inference cluster deliver per watt? Across deployments of identical hardware, the answer varies by 40x -- not because of software inefficiency, but because of the serving context window. We derive the 1/W law: tokens per watt halves every time the context window doubles. A larger context window shrinks the KV-cache concurrency limit while leaving GPU power draw roughly unchanged. At 64K context, an H100 holds 16 sequences in flight (tok/W = 1.5); at 4K context, the same H100 holds 256 sequences (tok/W = 17.6). Routing topology -- which determines the effective context window each GPU services -- is a more powerful energy lever than buying newer hardware. Working from published H100 power measurements, a calibrated logistic power model, and a roofline throughput model, we derive these results analytically using the inference-fleet-sim framework; no new hardware experiments were conducted. Two-pool context-length routing (FleetOpt) delivers roughly 2.5x better tok/W over a homogeneous fleet, while upgrading from H100 to B200 delivers roughly 1.7x. The gains are independent: combining FleetOpt with B200 yields 4.25x over the H100 homogeneous baseline. B200/H200 numbers are analytical projections (+-20% uncertainty); H100 results are calibrated to published measurements. For MoE models, active-parameter weight streaming adds a third lever. Qwen3-235B-A22B (22B active) reaches roughly 37.8 tok/W at 8K context on H100 -- 5.1x better than Llama-3.1-70B -- because decode time scales with activated weights, not total parameters. MoE dispatch overhead is excluded, so this is an upper bound.

cs.DC↗
arXiv2026

inference-fleet-sim: A Queueing-Theory-Grounded Fleet Capacity Planner for LLM Inference

Sizing a GPU fleet for LLM inference is harder than it looks. The obvious questions -- how many GPUs, which type, where to split a two-pool fleet -- have no closed-form answers. They depend on the full token-length distribution, the routing policy, and queueing dynamics that turn ugly under heavy-tailed workloads. Existing tools optimize per-engine configuration for a fixed GPU count; none of them address the upstream question of how many GPUs to buy and how to arrange them. inference-fleet-sim fills that gap. It combines analytical M/G/c queueing with discrete-event simulation (DES) to find the minimum-cost fleet configuration that empirically meets a P99 TTFT SLO. It includes a physics-informed GPU performance model covering A10G, A100, and H100 across monolithic, two-pool-routed, and disaggregated topologies, all without requiring access to real hardware. We run the tool on seven fleet-planning scenarios drawn from two public workload traces (LMSYS, Azure) and one synthetic agent-heavy trace. Each one surfaces a result that simple analysis gets wrong -- the right split threshold, the cheapest GPU type, whether an apparently idle fleet is actually broken -- and shows why joint simulation of queueing, routing, and hardware is necessary to find it.

cs.DC↗
arXiv2026

FleetOpt: Analytical Fleet Provisioning for LLM Inference with Compress-and-Route as Implementation Mechanism

Modern LLM GPU fleets are provisioned for worst-case context lengths that the vast majority of requests never approach, wasting GPU capacity on idle KV-cache slots. We present FleetOpt, a framework that starts from first principles: given a workload's prompt-length CDF and a P99 TTFT target, derive the minimum-cost fleet analytically, then deploy it in practice. The analytical core models each pool as an M/G/c queue and derives that the minimum-cost fleet is a two-pool architecture -- a short-context pool and a long-context pool -- with an optimal boundary B* satisfying an equal marginal GPU cost condition across both pools. The fundamental barrier to achieving B* is the cost cliff: a hard routing step where requests just above B* consume 8x--42x more GPU capacity than requests just below it (depending on the context window ratio), creating a structural disincentive to lower the boundary. Compress-and-Route (C&R) is the implementation mechanism that resolves this barrier. Gateway-layer extractive compression trims borderline requests below B* before the engine ever sees them, converting the hard hardware boundary into a software parameter read from the workload CDF. The two components are unified in the FleetOpt offline planner: given a CDF and SLO, it returns the optimal (n_s*, n_l*, B*, gamma*) in under 1 ms. On three production traces, the combined framework reduces total GPU cost by 6--82% versus a homogeneous fleet, with C&R contributing 1--44 percentage points beyond plain pool routing depending on workload archetype. The analytical model is validated against a discrete-event simulator (inference-fleet-sim) with <= 3% error on predicted GPU utilization across all pools and workloads.

cs.DC↗
arXiv2026

Outcome-Aware Tool Selection for Semantic Routers: Latency-Constrained Learning Without LLM Inference

Semantic routers in LLM inference gateways select tools in the critical request path, where every millisecond of added latency compounds across millions of requests. We propose Outcome-Aware Tool Selection (OATS), which interpolates tool embeddings toward the centroid of queries where they historically succeed -- an offline process that adds no parameters, latency, or GPU cost at serving time. On MetaTool (199~tools, 4,287~queries), this improves NDCG@5 from 0.869 to 0.940; on ToolBench (2,413~APIs), from 0.834 to 0.848. We also evaluate two learned extensions: a 2,625-parameter MLP re-ranker and a 197K-parameter contrastive adapter. The MLP re-ranker hurts or matches baseline when outcome data is sparse relative to the tool set; the contrastive adapter provides comparable gains on MetaTool (NDCG@5: 0.931). All methods are evaluated on the same held-out 30\% test split. The practical takeaway is to start with the zero-cost refinement and add learned components only when data density warrants it. All mechanisms run within single-digit millisecond CPU budgets.

cs.LG↗
arXiv2026

Token-Budget-Aware Pool Routing for Cost-Efficient LLM Inference

Production vLLM fleets provision every instance for worst-case context length, wasting 4-8x concurrency on the 80-95% of requests that are short and simultaneously triggering KV-cache failures -- OOM crashes, preemption storms, and request rejections. Both problems share a single root cause: configuration-traffic mismatch. We propose token-budget-aware pool routing: estimate each request's total token budget using a self-calibrating per-category bytes-per-token ratio, then dispatch it to one of two vLLM pools -- a high-throughput short pool or a high-capacity long pool -- each right-sized for its workload class. The ratio is learned online via exponential moving average from usage.prompt_tokens feedback, requiring no tokenizer. A closed-form cost model, savings = alpha * (1 - 1/rho), predicts fleet-level GPU savings from two observable quantities: the short-traffic fraction alpha and the throughput gain ratio rho. On traces from the Azure LLM Inference Dataset and LMSYS-Chat-1M serving Llama-3-70B on A100 GPUs, token-budget routing reduces GPU instances by 17-39% (\$1.2-2.0M/yr at 1,000 req/s), with savings verified by a self-contained discrete-event simulator. A case study projecting Qwen3-235B-A22B on AMD MI300X at 10,000 req/s shows \$15.4M/yr in savings. The algorithm adds O(1) dispatch overhead, self-calibrates across content types without a tokenizer, and composes with PagedAttention, continuous batching, and prefill-decode disaggregation.

cs.DC↗
arXiv2025

EVICPRESS: Joint KV-Cache Compression and Eviction for Efficient LLM Serving

Reusing KV cache is essential for high efficiency of Large Language Model (LLM) inference systems. With more LLM users, the KV cache footprint can easily exceed GPU memory capacity, so prior work has proposed to either evict KV cache to lower-tier storage devices, or compress KV cache so that more KV cache can be fit in the fast memory. However, prior work misses an important opportunity: jointly optimizing the eviction and compression decisions across all KV caches to minimize average generation latency without hurting quality. We propose EVICPRESS, a KV-cache management system that applies lossy compression and adaptive eviction to KV cache across multiple storage tiers. Specifically, for each KV cache of a context, EVICPRESS considers the effect of compression and eviction of the KV cache on the average generation quality and delay across all contexts as a whole. To achieve this, EVICPRESS proposes a unified utility function that quantifies the effect of quality and delay of the lossy compression or eviction. To this end, EVICPRESS's profiling module periodically updates the utility function scores on all possible eviction-compression configurations for all contexts and places KV caches using a fast heuristic to rearrange KV caches on all storage tiers, with the goal of maximizing the utility function scores on each storage tier. Compared to the baselines that evict KV cache or compress KV cache, EVICPRESS achieves higher KV-cache hit rates on fast devices, i.e., lower delay, while preserving high generation quality by applying conservative compression to contexts that are sensitive to compression errors. Evaluation on 12 datasets and 5 models demonstrates that EVICPRESS achieves up to 2.19x faster time-to-first-token (TTFT) at equivalent generation quality.

cs.OS↗
arXiv2025

When to Reason: Semantic Router for vLLM

Large Language Models (LLMs) demonstrate substantial accuracy gains when augmented with reasoning modes such as chain-of-thought and inference-time scaling. However, reasoning also incurs significant costs in inference latency and token usage, with environmental and financial impacts, which are unnecessary for many simple prompts. We present a semantic router that classifies queries based on their reasoning requirements and selectively applies reasoning only when beneficial. Our approach achieves a 10.2 percentage point improvement in accuracy on the MMLU-Pro benchmark while reducing response latency by 47.1% and token consumption by 48.5% compared to direct inference with vLLM. These results demonstrate that semantic routing offers an effective mechanism for striking a balance between accuracy and efficiency in open-source LLM serving systems

cs.ET↗
arXiv2025

LMCache: An Efficient KV Cache Layer for Enterprise-Scale LLM Inference

KV cache has traditionally been stored in GPU memory to accelerate the decoding phase of large language model (LLM) inference. However, it is increasingly necessary to move KV caches outside GPU devices, to enable cache reuse across different queries and inference engines. Our real-world usage statistics confirm this trend: over time, the total KV cache stored by users has grown rapidly, far exceeding the capacity of GPU memory. Despite this need, there lacks an efficient solution for offloading and transferring KV caches. We present LMCACHE, the first and so far the most efficient open-source KV caching solution, which extracts and stores KV caches generated by modern LLM engines (vLLM and SGLang) out of the GPU memory and shares them across engines and queries. LMCACHE supports both cache offloading (prefix reuse across queries) and prefill-decode (PD) disaggregation (cross-engine/GPU cache transfer). LMCACHE's high performance and wide adoption stem from the following contributions: (1) highly optimized KV cache data movement powered by batched data movement operations, compute and I/O pipelining; (2) a modular KV cache connector component, decoupling LMCACHE from the rapid evolution of inference engines; (3) a first-class control API for flexible cache orchestration across GPU, CPU, storage, and network layers. Our evaluation shows that combining LMCACHE with vLLM achieves up to 15x improvement in throughput across workloads such as multi-round question answering and document analysis. Large-scale adoption of LMCACHE in enterprise settings provides us valuable insights, for example, fetching KV cache from remote storage has unsurprisingly benefits to prefill delay, and that context truncation, which is a widely applied technique in industry, can greatly reduce prefix cache hit ratio by half. The source code of LMCACHE is at: https://github.com/LMCache/LMCache.

cs.LG↗
arXiv2025

ShadowServe: Interference-Free KV Cache Fetching for Distributed Prefix Caching

Distributed prefix caching accelerates long-context LLM serving by reusing KV cache entries for common context prefixes. However, KV cache fetches can become a bottleneck when network bandwidth is limited. Compression mitigates the bandwidth issue, but can degrade overall performance when decompression interferes with model computation. We present ShadowServe, the first SmartNIC-accelerated, interference-free prefix caching system for LLM serving. ShadowServe separates a control plane on the host and a data plane fully offloaded to the SmartNIC, which eliminates interference to both host GPU and CPU. To overcome the SmartNIC's limited compute and memory resources, we design a chunked pipeline that parallelizes data plane operations across the SmartNIC's compute resources, and a minimal-copy memory management scheme that reduces memory pressure on the SmartNIC. Compared to state-of-the-art solutions, ShadowServe achieves up to 2.2x lower loaded time-per-output-token (TPOT), and reduces time-to-first-token (TTFT) by up to 1.38x in low-bandwidth scenarios (<= 20 Gbps), translating to up to 1.35x higher throughput.

cs.DC↗
arXiv2025

AdaptCache: KV Cache Native Storage Hierarchy for Low-Delay and High-Quality Language Model Serving

Large language model (LLM) applications often reuse previously processed context, such as chat history and documents, which introduces significant redundant computation. Existing LLM serving systems address such redundant computation by storing the KV caches of processed context and loading the corresponding KV cache when a new request reuses the context. Further, as these LLM applications scale, the total size of KV caches becomes excessively large and requires both DRAM and SSD for full storage. However, prior work that stores KV caches in DRAM and SSD suffers from high loading delays, as most KV cache hits come from SSD, which is slow to load. To increase the KV cache hit rate on DRAM, we identify lossy KV cache compression as a promising approach. We design a lossy compression system that decides the compression algorithm, compression rate and device placement for each KV cache entry to maximise DRAM hits and minimise loading delay without significantly degrading generation quality. Compared to various static compression baselines across three tasks, our system AdaptCache achieves 1.43--2.4 x delay savings at the same quality and 6--55% quality improvements at the same delay.

cs.OS↗
arXiv2025

PrefillOnly: An Inference Engine for Prefill-only Workloads in Large Language Model Applications

Besides typical generative applications, like ChatGPT, GitHub Copilot, and Cursor, we observe an emerging trend that LLMs are increasingly used in traditional discriminative tasks, such as recommendation, credit verification, and data labeling. The key characteristic of these emerging use cases is that the LLM generates only a single output token, rather than an arbitrarily long sequence of tokens. We call this prefill-only workload. However, since existing LLM engines assume arbitrary output lengths, they fail to leverage the unique properties of prefill-only workloads. In this paper, we present PrefillOnly, the first LLM inference engine that improves the inference throughput and latency by fully embracing the properties of prefill-only workloads. First, since it generates only one token, PrefillOnly only needs to store the KV cache of only the last computed layer, rather than of all layers. This drastically reduces the GPU memory footprint of LLM inference and allows handling long inputs without using solutions that reduces throughput, such as cross-GPU KV cache parallelization. Second, because the output length is fixed, rather than arbitrary, PrefillOnly can precisely determine the job completion time (JCT) of each prefill-only request before it starts. This enables efficient JCT-aware scheduling policies such as shortest remaining job first. PrefillOnly can process upto 4x larger queries per second without inflating average and P99 latency.

cs.DC↗
arXiv2025

HyperRAG: Enhancing Quality-Efficiency Tradeoffs in Retrieval-Augmented Generation with Reranker KV-Cache Reuse

Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for enhancing the performance of large language models (LLMs) by integrating external knowledge into the generation process. A key component of RAG pipelines is the reranker, which selects the most relevant documents from a pool of retrieved candidates and significantly improves the quality of the generated responses. While rerankers refine the selection of retrieved documents in RAG pipelines, they introduce computational challenges that hinder high throughput and low latency. To address this problem, we propose HyperRAG, a system that optimizes the trade-off between quality and efficiency in RAG pipelines by leveraging KV-cache reuse for efficient reranker inference. By reusing document-side KV-cache, HyperRAG achieves both high-quality generation and system-level efficiency. To fully realize the benefits of KV-cache reuse, HyperRAG incorporates a range of system-level optimizations designed to enhance efficiency and scalability. Experiments show that HyperRAG achieves a 2 - 3 throughput improvement with decoder-only rerankers while also delivering higher downstream performance compared with traditional RAG service.

cs.CL↗