2026-07-08
DepthWeave-KV: Token-Adaptive Cross-Layer Residual Factorization for Long-Context KV Cache Compression
针对长上下文KV缓存内存开销问题，提出跨层残差分解的自适应压缩方法，提升检索性能并降低内存占用。
Authors: Anna Cordoba, Adam Puente Tercero, Nerea Angulo Hijo, Mar Linares Tercero, Julia Barrientos, Ainhoa Miranda, Jesus Olivera | Date: 2026-07-07
Link: http://arxiv.org/abs/2607.06523v1
Abstract
Long-context language model inference is increasingly limited by the memory bandwidth and capacity required to store key-value caches, yet existing compression methods often apply uniform budgets across layers or tokens and degrade retrieval when lexical cues and semantic states require different preservation. We introduce DepthWeave-KV, a token-adaptive cache compression method that factorizes key and value states across neighboring transformer layers using shared low-rank channel bases while retaining lightweight token-specific residuals where attention behavior is sensitive. DepthWeave-KV combines cross-depth residual factorization with a token-conditional depth router that allocates higher reconstruction rank to instruction-bearing and retrieval-critical tokens, and uses calibration-free online error tracking from attention-output probes to adapt compression during generation without retraining the base model. A fused CUDA implementation jointly performs basis lookup, residual dequantization, and attention projection to reduce decode-time memory traffic. Across LongBench, Needle-in-a-Haystack, L-Eval, and long-form QA and summarization benchmarks, DepthWeave-KV achieves near-full-cache task quality with substantially lower memory use, improving average score and retrieval accuracy over prior compressed caches while reaching 8.3x KV memory reduction and 72.8 tokens per second at 64K context.FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference
针对长上下文KV缓存压缩的检索退化问题，提出频率引导的深度共享方法，平衡压缩率与多步推理准确性。
Authors: Anna Córdoba, Adam Puente Tercero, Nerea Angulo Hijo, Mar Linares Tercero, Julia Barrientos, Ainhoa Miranda, Jesús Olivera | Date: 2026-07-07
Link: http://arxiv.org/abs/2607.06519v1
Abstract
Long-context LLM inference is increasingly limited by the memory and bandwidth cost of KV caches, yet aggressive compression can remove the layer-specific evidence needed for retrieval and multi-step reasoning. We introduce FreqDepthKV, an inference-time cache compression method that factorizes adjacent-layer KV states into shared low-frequency depth components and sparse high-frequency residuals. A lightweight online probe assigns attention heads to shared-depth, residual-depth, or exact cache modes according to their contribution to reconstruction-sensitive attention logits, allowing the compression policy to adapt to prompt structure without retraining. Across long-context question answering, needle retrieval, summarization, and code generation benchmarks, FreqDepthKV preserves task accuracy under substantially smaller cache budgets. With a 32k-token prefill window, FreqDepthKV reaches 58.3 Exact Match, 63.0 F1, 32.5 ROUGE-L, and 48.1 pass@1, closely matching full KV while outperforming prior compressed-cache methods. It also improves decoding throughput to 70.4 tokens/s, reduces TTFT to 2.06 seconds, and lowers peak KV memory to 6.2 GB, achieving a 3.9x effective compression ratio.UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods
针对MoE模型在超级计算集群上的通信瓶颈，提出重构专家并行通信库，优化执行并行性与通信效率。
Authors: Yipeng Liu, Chang Liu, Si Shen, Jiaqi Zheng, Mingfan Li, Yuyang Yang, Guanhua Li, Yuquan Zhang, Yimeng Xu, Zhongzhe Hu, Zhiyuan Huang, Qihang Duan, Junsong Wang, Wenkai Ling, Baochuan Yang, Xianzhi Yu, Han Bao, Yijie Chen, Guihai Chen | Date: 2026-07-07
Link: http://arxiv.org/abs/2607.06202v1
Abstract
The deployment of Mixture-of-Experts (MoE) models on production high-bandwidth superpods, such as NVIDIA's NVL72/576 and Huawei's CloudMatrix384, introduces critical challenges beyond raw interconnect bandwidth. While these systems provide unified global address spaces and high-bandwidth fabrics, their full potential for sparse MoE communication is hindered by three fundamental bottlenecks: (1) Strict execution serialization imposed by coarse-grained Bulk Synchronous Parallel (BSP) orchestration of interdependent communication phases; (2) Prohibitive synchronization overhead that fails to scale alongside high interconnect bandwidth; and (3) Severe load imbalance resulting from distance-agnostic scheduling of irregular token traffic. To eliminate these bottlenecks, we introduce UBEP (Unified-Bus Expert Parallelism), a production-ready communication library that rethinks MoE's All-to-All primitives for modern superpod architectures. Through large scale experiments, UBEP reduces All-to-All latency by up to 52.4% and MoE inference Time Per Output Token (TPOT) by up to 11.1%.MemDefrag: Latent Memory Defragmentation for Large Language Models
针对大语言模型潜在记忆更新时的性能退化问题，提出潜在内存碎片整理机制，提升记忆检索准确性。
Authors: Ruiyi Yan, Zhuoyuan Mao, Yiwen Guo | Date: 2026-07-07
Link: http://arxiv.org/abs/2607.05969v1
Abstract
Latent memory, which stores past knowledge fragments as per-layer hidden states, has emerged as a promising paradigm (e.g., MemoryLLM and M+) for long-term memory in large language models (LLMs). However, the paradigm suffers from significant performance degradation during memory updates, due to positional encoding misalignment and the absence of any tracing mechanism to distinguish target memory fragments from irrelevant ones. To discover such a tracing mechanism, we probe the layer-wise attention density over stored memory fragments, and find that a small set of middle transformer layers consistently concentrates the highest density on the target fragment - exposing an inherent tracing signal. In light of this, we propose MemDefrag, a training-free and model-agnostic framework that (1) uses a middle-layer tracing signal to conduct memory defragmentation (rank, reorder, and filter memories), and (2) applies an informativeness-guided proportional forgetting mechanism once capacity is exceeded. Experiments show that MemDefrag substantially outperforms MemoryLLM and M+ on knowledge retention (e.g., 43.0% vs. 17.4%/17.6% after 50 memory updates) and long-context benchmarks, and generalizes well across various LLMs and latent-memory variants.Think Before You Grid-Search: Floor-First Triage for LLM Serving
针对LLM服务优化中的配置搜索低效问题，提出基于残差驱动的分层诊断工作流，提升资源分配效率。
Authors: Yihua Liu | Date: 2026-07-07
Link: http://arxiv.org/abs/2607.05876v1
Abstract
LLM serving optimization typically benchmarks many configurations and reaches for heavy profilers when latency targets are missed. We argue for the reverse discipline: estimation is the analytical layer of profiling -- without it, optimization degenerates to grid search. Floor First is a residual-driven triage workflow. Each decode step is modeled as a five-dimensional resource vector (HBM bytes, FLOPs, network bytes, network messages, KV capacity); summing within a resource and maximizing across resources gives an optimistic floor, the plain sum a pessimistic one. Where a measurement lands inside this [max, sum] interval reads out overlap quality before any profiler is opened, and profilers escalate only on residuals above a stated threshold. Deployment alternatives are compared by wall ordering -- which resource wall binds first as load grows -- rather than by point benchmarks. The account is compositional: new attention or state-space variants enter by declaring one module, and the workflow ships as a zero-dependency calculator plus an agent skill that enforces the discipline in agentic optimization loops. As a case study we analyze a DeepSeek-V3.2-style 671B MoE/MLA model on 16 NVIDIA H20 GPUs, whose ridge point of ~74 FLOP/byte (vs ~590 for H100) makes it an extreme decode-oriented part. The floors show TP16 decoding is KV-capacity-limited to ~70 concurrent 8K requests; sparse attention removes the KV-bandwidth term but not the capacity wall; an EP16+DP-attention layout accepts slightly worse same-batch weight traffic for an order-of-magnitude higher capacity wall (~644) -- while single-stream latency favors TP by 2.4x. The layout judgment is thus a computable function of the operating point, explaining why production deployments on identical hardware have shipped opposite attention layouts.Akashic: A Low-Overhead LLM Inference Service with MemAttention
针对多轮交互中上下文重放成本高的问题，提出基于MemAttention的低开销记忆系统，提升服务效率与输出质量。
Authors: Yang Liu, Zhaokai Luo, Huayi Jin, Ruozhou He, Chenchen Hong, Zhiyong Wang, Yifei Liu, Yunfei Gu, Chentao Wu, Junhao Hu | Date: 2026-07-07
Link: http://arxiv.org/abs/2607.05708v1
Abstract
Recent LLM-based agent systems continuously accumulate context across multi-turn interactions, tool invocations, and cross-session workflows. Replaying the full history for every request quickly becomes impractical: long contexts increase prefill cost, may exceed context limits, and often bury task-relevant evidence in irrelevant content, degrading both serving efficiency and output quality. We propose Akashic, a low-overhead memory system built around MemAttention, which organizes context into bounded chunks and models semantic relationships across chunks, preserving cross-chunk evidence without repeatedly rewriting the full history. Akashic further applies hardware-software co-designed memory placement to co-locate likely co-retrieved chunks, reducing retrieval fragmentation and I/O overhead. Across four representative workloads and three model sizes, Akashic improves task accuracy by up to 10.2 points, throughput by up to 1.21x, and sustainable request rate by up to 1.88x over strong prior memory baselines.A Coin Flip Per Token: Bernoulli Sparse Steering of Large Language Models
针对激活引导中密集干预影响流畅性的问题，提出随机令牌/块引导方法，实现高效行为控制。
Authors: Nima Eshraghi, Lovedeep Gondara, Yuqing Huang, Sagarika Suresh, Leizer Teran, Jithin Pradeep, Xiaotong Xu, Fanny Chevalier | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.05615v1
Abstract
Activation steering via sparse autoencoders (SAEs) enables behavioral control of large language models without task-specific fine-tuning, but standard methods apply the steering signal at every generated token, incurring constant per-token perturbation that risks degrading fluency. We ask: is dense intervention necessary? We introduce Stochastic Token Steering (STS), which gates each token independently with probability $p$, and Stochastic Block Steering (SBS), which gates a leading window once per sequence; neither requires a reward model or learned gating policy. Across two model families and two behavioral tasks, steering only 50% of the tokens recovers most of the dense-steering effect while preserving fluency, and steering as few as 30% surpasses prompt-based control. The optimal steering magnitude scales inversely with the intervention ratio, revealing that SAE-mediated control is rate-limited: the behavioral outcome depends on cumulative signal dosage across a sequence.ResonatorLM: Causal Resonant Field Mixing for Efficient Long-Context Language Modelin
针对Transformer在长上下文建模中的效率瓶颈，提出因果共振场混合架构，提升长序列建模效率。
Authors: Archie Chaudhury | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.05583v1
Abstract
Contemporary language models are dominated by the transformer architecture, which leverages self-attention mechanisms to enable more efficient, parallelized training across a wide set of documents and corpora. This has allowed transformers to effectively model data across a wide range of modalities and contexts. However, transformers, along with their conventional counterparts such as recurrent neural networks (RNNs) and convolutional neural networks (CNNs), often struggle to maintain efficiency when processing long contexts. We introduce ResonatorLM, a new mechanism that replaces attention with a physics-derived alternative. ResonatorLM treats token sequences as a single, driven one-dimensional latent field and replaces attention dot products with causal functions of damped resonators. We implement ResonatorLM on a traditional network architecture and test it on standard long-context modeling tasks. We find that in a small, 6M matched setting, training and prefill speedups increase with sequence length, decode speed reaches 6.47x compared to that of a standard, optimized transformer at 32K tokens, and accuracy reaches 61.31 percent (compared to 55.32 percent) on WikiText.Adaptive Inference Batching using Policy Gradients
针对静态批处理策略无法适应动态负载的问题，提出基于强化学习的自适应批处理与路由策略，提升吞吐量与延迟。
Authors: Ruslan Sharifullin | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.05272v1
Abstract
Inference serving systems must balance throughput and latency under bursty, heterogeneous workloads, yet the industry standard remains static batching policies that require manual tuning and cannot adapt to shifting traffic. We investigate whether reinforcement learning (RL) can learn adaptive batching and routing policies that outperform these heuristics, training REINFORCE and PPO agents on a discrete-event simulator validated against queuing theory and production traces (Azure Functions, BurstGPT). We formulate the problem as an MDP over queue state, request type and GPU availability, evaluating across standard Poisson traffic, extreme bursts, real-world traces and heterogeneous multi-GPU routing. Our central finding is a clear boundary condition for RL's value in systems problems. In single-GPU settings, a well-tuned static batching policy is already near-optimal under Poisson-like arrivals and RL offers only marginal gains (+0.1% to +1.0%). In multi-GPU heterogeneous routing, however, where fast and slow requests compete for shared resources, the agent discovers a workload-segregation policy that eliminates Head-of-Line blocking, yielding a 3.5x (348%) improvement over Round-Robin and a 48% improvement over the strongest heuristic baseline (Shortest-Queue), with 60% higher throughput and 25% lower latency while respecting SLA constraints. The policy generalizes to unseen bursty and real-world traffic despite training only on synthetic Poisson arrivals and an attention-augmented policy network converges roughly 20% faster than an MLP baseline. These results suggest RL's advantage over engineered heuristics concentrates in combinatorial, multi-resource decisions rather than single-resource temporal scheduling, a practical distinction for deciding where learned policies justify their engineering cost in production inference infrastructure.Optimizing ML Workload Partitioning between CPUs and CIM Accelerators for Heterogeneous Computing
针对CIM加速器上ML工作负载分区未考虑硬件约束的问题，提出优化分区方法，提升计算效率与能效。
Authors: Joel Klein, Rebecca Pelke, Roberto Laudani, Jan Moritz Joseph, Rainer Leupers | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.05240v1
Abstract
Computing-in-Memory (CIM) accelerators execute Matrix-Vector Multiplications (MVMs) in memory, making them a compelling solution for Machine Learning (ML) workloads. However, existing ML workload partitioning approaches for CIM accelerators do not fully account for Resistive Random Access Memory (RRAM) constraints such as limited memory, high write latency, and limited endurance. They also neglect parallelism, low-level architectural effects, or the Central Processing Unit (CPU) as a complementary compute resource. To address these limitations, we propose an Integer Linear Programming (ILP)-based workload partitioning framework for heterogeneous CPU-CIM systems. It minimizes end-to-end inference latency under RRAM constraints, captures parallelism, and combines empirical profiling with analytical models. Using our framework, heterogeneous CPU-CIM execution achieves speedups of up to 30.9x over CPU-only execution on an edge CPU and 7.3x over a high-performance CPU. A Design Space Exploration (DSE) yields further design insights for future CIM accelerators.DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation
针对并行草稿生成中接受率衰减和批处理容量浪费问题，提出置信度调度的推测解码方法，提升高并发吞吐量。
Authors: Xin Cheng, Xingkai Yu, Chenze Shao, Jiashi Li, Yunfan Xiong, Yi Qian, Jiaqi Zhu, Shirong Ma, Xiaokang Zhang, Jiasheng Ye, Qinyu Chen, Chengqi Deng, Jiping Yu, Damai Dai, Zhengyan Zhang, Yixuan Wei, Yixuan Tan, Wenkai Yang, Runxin Xu, Yu Wu, Zhean Xu, Xuanyu Wang, Muyang Chen, Rui Tian, Xiao Bi, Zhewen Hao, Shaoyuan Chen, Huanqi Cao, Wentao Zhang, Anyi Xu, Huishuai Zhang, Dongyan Zhao, Wenfeng Liang | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.05147v1
Abstract
Speculative decoding accelerates Large Language Model (LLM) inference by decoupling draft generation from target verification. While recent parallel drafters efficiently propose long token sequences in a single forward pass, they suffer from rapid acceptance decay due to a lack of inter-token dependencies. Furthermore, indiscriminately verifying these extended blocks wastes critical batch capacity on tokens with high rejection risks, severely degrading throughput in high-concurrency serving systems. We introduce DSpark, a speculative decoding framework that unifies high-throughput parallel generation with adaptive, load-aware verification. To maintain draft quality, DSpark utilizes a semi-autoregressive architecture, coupling a parallel backbone with a lightweight sequential module, to introduce intra-block dependency modeling and mitigate suffix decay. To optimize system efficiency, DSpark employs confidence-scheduled verification, dynamically tailoring the verification length for each request based on estimated prefix survival probabilities and engine-specific throughput profiles. On offline benchmarks across diverse domains, DSpark substantially improves the accepted length over state-of-the-art autoregressive and parallel drafters. When deployed within the DeepSeek-V4 serving system under live user traffic, DSpark successfully mitigates verification waste. Compared to the established production baseline (MTP-1), DSpark accelerates per-user generation speeds by 60 to 85 percent at matched throughput levels. More importantly, by preventing severe throughput degradation under strict interactivity constraints, it enables performance tiers that were previously unattainable, shifting the Pareto frontier of our serving system.Communication-Aware Placement and Pruning for Efficient Mixture-of-Experts Inference
针对MoE模型分布式推理中通信量过大的问题，提出通信感知的专家放置与剪枝框架，降低通信开销。
Authors: Xiao Shi, Yingying Sun, Jiangsu Du, Zhiguang Chen, Yutong Lu | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.05116v1
Abstract
As MoE models scale to hundreds of experts, placement and pruning decisions increasingly dictate communication volume, affecting the performance of distributed inference across GPUs and nodes. We propose CAP (Communication-Aware Assignment and Pruning), a framework that considers computation, communication and accuracy together for efficient MoE inference through expert placement and pruning. It consists of three components: (1) Co-activation driven expert placement, which groups frequently co-activated experts to reduce inter-device and inter-node communication; (2) Communicationcomputation trade-off adjustment, which generates placements with different computational load and communication volume; and (3) Communication-aware expert pruning, which selectively removes routing destinations to reduce communication with limited accuracy degradation. By combining these components, CAP selects an efficient operating strategy for different hardware configurations. Across our single-node and multi-node experiments, it achieves 1.23x - 1.86 x throughput improvement over DeepSeek EPLB and sequential placement in vLLM, and preserves better model accuracy at the same target speedup under lossy acceleration.KVpop -- Key-Value Cache Compression with Predictive Online Pruning
针对KV缓存驱逐策略静态且不准确的问题，提出基于预测在线剪枝的压缩方法，提升缓存利用率与性能。
Authors: Lukas Hauzenberger, Niklas Schmidinger, Anamaria-Roberta Hartl, David Stap, Thomas Schmied, Sebastian Böck, Günter Klambauer, Sepp Hochreiter | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.05061v1
Abstract
Key-value (KV) cache growth is a major bottleneck in autoregressive decoding, as memory and bandwidth scale linearly with context length. Existing KV eviction methods often rely on static heuristics or proxy scores, which poorly track future token utility and cause brittle eviction as relevance shifts. To address this, we introduce KVpop, which learns a fixed-budget KV eviction policy by directly supervising the keep-or-drop decision. The scorer is trained against a novel future-attention target, computed efficiently without materializing dense attention maps. We further introduce a delayed memory-based scorer that, uniquely among learned eviction methods, defers scoring for a fixed number of steps to exploit near-future context. On AIME and HMMT mathematical reasoning, KVpop retains 98% of full-attention performance on Qwen3-4B at 75% KV cache compression and 97% at 88% compression, consistently outperforming established eviction baselines. Qwen3-8B shows even stronger results, reaching near-full teacher performance. These results show that supervising eviction with future-attention signals cuts memory costs while maintaining quality.Train Smarter, Not Longer: Memorization-Guided Data Reuse for Efficient LLM Training
针对多轮训练中数据重用效率低的问题，提出基于记忆窗口信号的数据重用指导方法，提升训练效率与性能。
Authors: Jingwei Zuo, Cong Zeng, Ilyas Chahed, Maksim Velikanov, Dhia Eddine Rhaiem, Pasquale Balsebre, Abhay Kumar, Younes Belkada, Hakim Hacid | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.04969v1
Abstract
The training paradigm of large language models has shifted from traditional one-pass training to multi-epoch training, as reasonable reuse of limited high-quality data can improve both model performance and sample efficiency. Meanwhile, excessive repetition introduces the risk of overfitting and diminishing returns. Determining when and how to reuse data effectively thus emerges as a natural but under-explored question. Through a novel observation of model's "Memorization Window" signals derived from loss retention dynamics and downstream evaluation scores, we propose "Memorization-guided Data Reuse", a training paradigm that adaptively determines when and how data should be reused, enabling principled decisions on the number of training epochs and the scheduling of data replays. Our preliminary experiments reveal a consistent memorization-driven regime: performance continues to improve with repetition far beyond current practice (e.g., the commonly cited four-epoch limit). While a full scheduler remains future work, these insights provide a foundation for memorization-aware training schedules, helping to determine reuse budgets and move toward training LLMs smarter rather than longer with limited high-quality data.Is Your NPU Ready for LLMs? Dissecting the Hidden Efficiency Bottlenecks in Mobile LLM Inference
针对移动设备LLM推理的硬件效率瓶颈问题，提出跨层测量研究与分析工具，揭示并优化能效瓶颈。
Authors: Guanyu Cai, Ruiming Tian, Lang Yang, Zhouhong Ren, Jinliang Yuan, Lingkun Li, Jiliang Wang | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.05475v1
Abstract
Deploying Large Language Models (LLMs) on mobile devices enhances privacy and reduces latency, but is severely bottlenecked by hardware inefficiency. We present the first comprehensive, cross-layer measurement study of mobile LLM inference, uniquely spanning five mainstream frameworks (e.g., llama.cpp, GENIE) and three hardware backends (CPU, GPU, NPU). To enable this analysis, we develop PowerBench, a fine-grained profiling tool that provides the first backend-specific energy attribution, moving beyond traditional device-level measurements. Our study yields three critical insights: (1) Framework-induced performance gaps are substantially amplified on NPUs, reaching up to 10x using custom operators due to divergent offloading and quantization strategies. (2) We identify a distinct phase split where NPUs excel at compute-bound prefilling, while CPUs outperform all other backends in memory-bound decoding. This is driven by the NPU's preference for large, fixed-shape workloads, which conflicts with the small-kernel, dynamic nature of decoding. (3) Backend-specific profiling uncovers substantial scheduling headroom missed by prior work. Suboptimal thread configurations, uncoordinated NPU sleep latencies, and CPU polling intervals result in up to 40% energy waste. Leveraging these findings, we present an energy-oriented best-practice configuration for mobile LLM inference. We estimate that this configuration could reduce energy consumption by up to 54.8% on the NPU backend across three datasets.Layer-Parallel Inference Reduces Encrypted Nonlinear Depth in Transformers
针对全同态加密Transformer推理中非线性深度瓶颈问题，提出结构化牛顿层并行方法，降低加密计算深度。
Authors: Ligong Han, Kai Xu, Hao Wang, Ruijiang Gao, Akash Srivastava | Date: 2026-07-06
Link: http://arxiv.org/abs/2607.04819v1