LLM Inference Acceleration

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

108 papers in the last four weeks, up 120% on the four weeks before. 1.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 677

Aug 1, 2026cs.AI

CURE: Local Uncertainty Repair for Block-Parallel Speculative Decoding

Speculative decoding mitigates the latency of sequential generation in autoregressive Large Language Models (LLMs) by interleaving draft generation with target verification. However, existing parallel drafting backends often suffer from rapid accuracy degradation over long horizons, leading to high rejection rates during verification and suboptimal wall-clock speedups. We observe that drafting errors are not uniformly distributed but typically stem from localized high-uncertainty tokens that destabilize downstream generation trajectories. Motivated by this token error pattern, we propose CURE, a budget-aware dynamic repair tree designed to repair errors at uncertainty focal points without incurring prohibitive tree-verification overheads. Specifically, our method uses predictive confidence margins to dynamically locate candidate error tokens within a block-parallel draft, expands bounded repair paths only at these fragile nodes, and employs a novel repair resynchronization mechanism to realign draft states post-verification. Evaluations on code-generation benchmarks (HumanEval, MBPP, and LiveCodeBench-lite) and mathematical reasoning benchmark (GSM8K) demonstrate that CURE increases the average accepted length by 4.2-7.5% over parallel baselines without repair, translating to an end-to-end speedup of

2.66-3.49\times

over target-only decoding. Furthermore, we provide a plug-and-play repair module compatible with standard parallel drafting frameworks. We also characterize the trade-off between draft compute and verification efficiency.

Aofan Liu, Jingxiang Meng, Fangxin Liu +1

Speculative Decoding LLM Inference Acceleration Parallel Decoding

Aug 1, 2026cs.CL

AdaMTP: An Adaptive Training Paradigm for Multi-Token Prediction

Multi-Token Prediction (MTP) has emerged as an effective paradigm that augments a shared Large Language Model backbone with auxiliary heads, training the model to predict several future tokens in parallel to enrich its supervision signal and accelerate inference. However, existing training frameworks adopt a rigid, fixed-length prediction horizon, disregarding the highly non-uniform information density of natural language and code. Forcing the auxiliary heads to predict across high-entropy semantic boundaries injects noisy, conflicting training signals; because these heads share the backbone's latent representations, the resulting gradients backpropagate and interfere with the model's core capabilities. We propose AdaMTP, an adaptive training paradigm that dynamically aligns the prediction horizon with the intrinsic predictability of the sequence. At its core, an entropy-based segmentation algorithm leverages the base model to detect sudden surges in uncertainty as semantic boundaries, partitioning sequences into variable-length groups. Each token is assigned an adaptive prediction depth, and a dynamically masked MTP objective suppresses the loss for predictions that cross these boundaries, attenuating the noisy gradients that degrade the backbone. Across mathematical reasoning, code generation, and general benchmarks on three backbones (Llama-3.1-8B, Qwen-2.5-7B, Gemma-3-12B), AdaMTP consistently outperforms standard MTP in both task performance and inference speedup.

Ziqiang Cui, Han Shi, Bowei He +8

City University of Hong Kong · Huawei Technologies · Mohamed bin Zayed University of Artificial Intelligence +2

Multi-Token Prediction LLM Inference Acceleration LLM Training

Jul 31, 2026cs.CL

TokTier: Exact Stateful CPU+GPU Tokenization for Agentic LLM Serving

LLM serving caches prompt KV state, yet most front ends still re-tokenize the full request on every call. Coding agents pay most: sessions repeatedly submit a long transcript after a small append, which can shift token boundaries near the end of the prior sequence. Across 153,951 calls the median append is ~1.4K characters; only 1.0-3.6% of calls start or rebuild a session, yet those carrymulti-million-character contexts. Fleet prompt-cache hit rate is 94.1%, and as it approaches 0.99, tokenization grows from 10% to 64% of time to first token (TTFT) in component measurements. TokTier is a stateful CPU+GPU tokenization service for this two-mode workload, under one contract: emitted token IDs are always identical to full reference tokenization. For session continuations it re-tokenizes a small window around the append and splices only when a per-request check finds a stable pre-tokenization boundary; failed checks widen the window or fall back to full reference tokenization. For calls without a reusable prefix it runs exact GPT-family regex pre-tokenization and BPE on a GPU. A sampled shadow verifier re-checks live traffic. Across 17 production tokenizer families, differential campaigns cover 1.5x10^10 split checks, a 12.4 TB real-text corpus, and 93,000+ replayed agent steps, with zero divergence. Incremental repair takes 0.5-1.1 ms from 100K to 3M characters, up to 437x faster than HF tokenization and 2.1x faster at 1M characters than the strongest cache-based baseline (Gigatoken) fully prewarmed. GPU tokenization encodes a 1M-character request in 0.87 ms, up to 491x below HF and 23.4x below the fastest published CPU method on the same protocol. With vLLM, median TTFT drops 16-34% and P99 TTFT 23% under recorded bursts. Under a 50 ms P99 objective, a four-core repair pool plus one GPU sustains 1,821 requests/s, where a 16-core stateless front end saturates at 40 requests/s.

Zhenyu Zhang, Zhichao Cao

Arizona State University · Tempe, Arizona, USA

LLM Serving LLM Inference Acceleration Subword Tokenization

Jul 31, 2026cs.AI

Decode-Branch Transformers: Decoupling the Primary Prefill Path from Additional Decode Computation

As large language models serve ever more requests, cumulative inference cost is growing relative to the one-time cost of training. In typical serving, prompt prefill runs in parallel and is compute-bound, whereas autoregressive decode is sequential and memory-traffic-bound. Conventional width or depth scaling raises both costs together, since every added layer is evaluated in both phases and enlarges the weights read at each decode step. We instead ask whether additional learned computation can be allocated to continuation prediction while preserving prompt-wide primary computation and a single KV cache. We realize this with the Decode-Branch Transformer. Its primary path alone processes the prompt and writes the KV cache; the decode branch is omitted during prefill and activated only from the final prompt position onward, adding continuation computation without writing state or affecting the primary path. The paths share attention, MLP, and output matrices, using separate token embeddings with lightweight coupling. Grouped decode reuses loaded weight tiles and the primary KV cache across both paths, so the added arithmetic does not proportionally increase dominant memory traffic or decode latency. Across matched-token comparisons, Decode-Branch achieves lower validation loss across architectures and data settings. In MoE models, the primary and branch expert fan-outs become independent knobs for trading prompt cost, decode cost, and predictive quality. We study two expert-allocation regimes, holding prefill or decode computation fixed, and expose a prefill-decode-quality trade-off enabled by phase-specific expert allocation.

Liming Liu, Mingze Wang, Tuo Zhao

Georgia Institute of Technology · Peking University

Efficient Transformer Inference Inference-Time Scaling LLM Inference Acceleration

Jul 31, 2026cs.CL

Studying quantization trade-offs for efficient inference deployment in machine translation

Deploying large language models in realistic server environments poses challenges, as the system needs to provide high-quality responses with low latency. Quantization is a common approach to reduce the memory footprint and improve inference efficiency, yet its impact on latency and throughput is rarely evaluated under controlled, orchestration-level workloads. In this work we study the quantization trade-offs of two translation model families, EuroLLM \citep{martins2025eurollm} and Hy-MT2 \citep{zheng2026hy} across five models ranging from 1.7B to 22B for efficient deployment on a single A100 or H100 GPU. We demonstrate that combining a document-chunking strategy with W4A8 or W8A8 quantization improves the latency-throughput Pareto-curve under a wide range of workloads. Furthermore, since standard machine translation (MT) benchmarks rely on isolated sentences and fail to capture long-context dynamics, we introduce a document-level evaluation from WMT24++ to assess how text chunking strategies affect translation quality under quantization. Our results reveal that standard segment-level evaluation can fail to predict the interaction between quantization and long-context document translation. While Hy-MT2 remains robust under quantization, EuroLLM shows strong sensitivity and translation quality collapses rapidly for all considered quantization formats. Overall, our experiments show that the trade-off between inference efficiency and translation quality depends not only on the quantization format, but also on the choice of text chunking strategy.

Jim Zhao, Sohir Maskey, Koen Oostermeijer +2

University of Basel · Aleph Alpha Research · Graphcore

LLM Quantization LLM Inference Acceleration Document-Level MT Machine Translation MT Evaluation

Jul 31, 2026cs.SD

ParaASR: Multi-Token Prediction for Fast and Long-Context LLM-Based Speech Recognition

Audio-encoder-LLM-decoder architectures have become the dominant paradigm for modern automatic speech recognition (ASR), improving transcription quality through large-scale language modeling. However, the cost of autoregressive decoding scales with decoder size, creating a fundamental trade-off between recognition quality and serving latency. We argue this trade-off is not inherent: unlike open-ended text generation, ASR outputs are strongly anchored to the input speech signal, providing a natural inductive bias toward high-parallelism decoding. Building on this, we introduce ParaASR, an ASR system that leverages Multi-Token Prediction (MTP) to let a 4B LLM decoder emit multiple tokens per forward step. Starting from a publicly available audio-language foundation, the model first establishes a robust autoregressive recognizer and then aligns five future-token branches through a staged optimization recipe. At inference, it proposes a six-token continuation per step and admits only the verified prefix into the transcript, preserving the safety of standard autoregressive decoding. The average accepted length reaches 5.0 out of 6 proposed tokens, confirming that the deterministic structure of speech makes ASR an especially natural setting for multi-token decoding. ParaASR further retains a native 32K-context window and transcribes up to 30 minutes of audio in a single pass. Across diverse benchmarks, it attains average error rates of 2.97%, 3.68%, and 3.70% on Chinese, English, and long-form evaluations, respectively, while reaching a real-time factor (RTF) as low as 0.0053. These results show that decoder scaling, low-latency inference, and long-context transcription need not be competing goals when future-token proposals are anchored by the acoustic signal and guarded by autoregressive verification.

Qingjian Lin, Yuxin Li, Haoyang Zhang +14

StepFun · NTU · PKU +3

Multi-Token Prediction Automatic Speech Recognition LLM Inference Acceleration Speech Language Models

Jul 31, 2026cs.CL

Mixture-of-Translators: Translating KV Caches Across Heterogeneous Large Language Models

Heterogeneous Large Language Model (LLM) systems increasingly rely on shared contexts, retrieved evidence, and multi-agent dialogue histories, yet their internal key-value (KV) caches remain model-specific and cannot be reused across architectures. Consequently, each model must repeatedly prefill or store caches for the same context, limiting the scalability of multi-model reasoning and long-context generation. We propose Mixture-of-Translators(MoT), a cache translation framework that maps context KV caches from a source LLM into the cache space of a target LLM. Unlike prior approaches that depend on a single projection path or global shared latent space, MoT uses multiple translator modules to capture diverse source--target mappings. To further reduce residual translation error, we introduce a Context Correction Loss that aligns the replayed target trajectory with the native target trajectory. We reveal two competing failure modes in cache translation: propagated translation shift from early injection and last-state shift from late injection. MoT addresses them through translator mixtures and target-side correction. Across homogeneous and heterogeneous translations among Qwen2.5, GPT-2, and OPT models, MoT preserves downstream QA performance, including Qwen2.5-7B-scale translation with 51.0% average closed-set QA accuracy and 0.43 average extractive QA F1. In practical case studies, MoT enables quality-preserving memory reuse for multi-agent reasoning and retains 96.3% of direct-context quality in long-context cache-augmented generation, demonstrating scalable KV cache reuse across heterogeneous LLMs.

Jin-woo Lee, Minkyung Song, Junghyun Oh +4

Chungnam National University · KISTI

Memory-Augmented Language Models KV Caching LLM Inference Acceleration

Jul 31, 2026cs.AI

LinearKV: One Cached State Suffices for Position-Independent Caching in Hybrid LLMs

LLM serving is increasingly accelerated by position-independent caching (PIC). Existing PIC methods, however, are built for full-attention models, where a token-indexed KV cache underlies its core operations: matching reusable token chunks, concatenating their KV entries, and selectively recomputing a few tokens to restore cross-chunk context. Hybrid LLMs break these primitives---they replace most attention layers with linear recurrences that expose only a fixed-size state, leaving no token-indexed KV to concatenate or to locally repair. This raises a natural question: can PIC benefit hybrid models, and what would it take? We present LinearKV, a training-free hybrid-PIC framework. Its key insight is a \emph{decoupled initialization}: each linear layer maps its

K

matched local states to a single initial state, while full-attention layers concatenate their KV as before. LinearKV is therefore compatible with existing PIC methods, reusing their token selection and recomputation as-is. Under this framework, we find that a \emph{single cached state} suffices as the linear layer's initializer. The algebraically principled alternative---composing all

K

cached states into the exact full-prefix state, as concurrent work HYPIC does---is unnecessary and, on some architectures, even harmful. We compare the two across three hybrid models and three PIC selectors. On the two GDN models the two tie, both recovering most of full quality (up to

92\%

); on the Mamba-2 model, exact composition instead collapses under every selector---under EPIC, for instance, it recovers only

46.6\%

of full quality, versus

86.8\%

for a single cached block initializer. A single state initializer is also cheaper, cutting time-to-first-token to

0.46\times

full prefill versus a further

5

--

17\%

overhead for exact composition; results hold across LongBench QA and RULER at 8K--32K.

Yirui Liu, Ruoling Qi, Longwen Wang +5

Institute of Artificial Intelligence, China Telecom (TeleAI) · Shanghai Jiao Tong University · Xi’an Jiaotong University +1

KV Caching LLM Inference Acceleration Hybrid Attention

Jul 30, 2026cs.AI

WIDE: Boosting Adaptive LLM Inference via Token-level Dynamic Width Pruning

Pruning is a promising approach for improving the efficiency of LLMs. Existing static structured pruning methods are hardware-friendly and can deliver practical throughput gains, but their input-agnostic computation allocation often causes substantial accuracy degradation under aggressive sparsity. Recent dynamic sparsity methods improve quality retention by adapting computation to individual inputs, yet they remain largely limited to coarse-grained structural decisions and their practical acceleration under real-world inference scenarios remains challenging. To address these challenges, we present WIDE, the first end-to-end differentiable token-level dynamic width pruning framework designed for both prefill and decode scenarios. WIDE enables fine-grained computation allocation by allowing each token to dynamically select attention-head groups and FFN-channel groups, extending dynamic pruning beyond layer-level decisions to neuron-block-level granularity. Through a two-stage training pipeline, WIDE learns effective token-wise sparse execution patterns and achieves substantially better quality retention than existing approaches. To make such fine-grained dynamic pruning practical, we further propose a pruning--kernel co-design framework that decomposes dynamic sparsity acceleration into mask reordering, hardware-agnostic block-level skipping, and hardware-dependent intra-block skipping, enabling efficient execution across different granularities. At 50% sparsity, WIDE provides 55.1% performance boost when compared to the state-of-the-art dynamic depth pruning under calibration-only settings. Under prefill and decoding inference workloads, WIDE achieves close-to-theoretical kernel-level speedups of up to 1.98x for prefill and 4.95x for decoding, as well as 1.68x and 1.55x end-to-end acceleration. Our code is available at https://github.com/EIT-NLP/LLM-Pruning/tree/main/WIDE.

Haozhe Hu, Hao Wu, Peiran Yin +3

Ningbo Institute of Digital Twin, Eastern Institute of Technology, Ningbo · Munich Center for Machine Learning, LMU Munich

LLM Pruning Token Pruning LLM Inference Acceleration Attention Head Pruning

Jul 30, 2026cs.CL

CoMem: Reusing Transformer Depth across Queries with Persistent Intermediate Residuals

Repeated queries over shared documents repeatedly execute the same lower transformer layers. We introduce CoMem, which makes split depth j an explicit reusable-context axis: write one depth-j residual per token, select a bounded chunk set, and resume only layers [j:L). Among document-reuse systems we are aware of, CoMem jointly makes split depth a tunable serving axis and isolates it with a matched j=0 endpoint. On Qwen3-8B, j=12 reduces selected-pack Read from 931.9 to 664.4 ms (1.403x), with a 3.12-point RULER cost (95% CI [2.36, 3.93]); a continuous-prefix oracle recovers the full gap. The resulting depth axis quantifies a quality-latency-storage trade-off; a separate same-adapter, Write-inclusive pipeline is 2.74x faster. Equal-latency raw replay leads by 11.56 points with BM25, directly measuring an applicability boundary of prepaid depth rather than hiding it. CoMem stores 8 KiB/token versus 144 KiB/token for a protocol-aligned same-Qwen3 CacheBlend-style diagnostic; the cohorts and adaptation budgets are not matched. A context-position factorization identifies missing lower-layer document context as the dominant tested multikey error, and a 32-token overlap raises 92.5 to 98.5 without increasing persistent bytes or per-query Read. CoMem opens transformer depth as a measurable, tunable dimension for repeated-query long-context serving.

Hanzuo Liu, Xuan Qi, Chunyu Liu +6

Tsinghua University · Tencent

Memory-Augmented Language Models Long-Context Language Model Inference LLM Inference Acceleration

Jul 30, 2026cs.CL

A Sparse Glimpse of the Whole: Train-Free Self-Speculative Decoding

Speculative decoding alleviates the memory-bandwidth bottleneck in large language model inference, but its acceleration is jointly constrained by drafting overhead, token acceptance, and speculation length. We present a unified efficiency analysis showing that extending the speculation horizon can reduce rather than improve speedup when the marginal acceptance probability falls below the relative drafting cost. Guided by this analysis, we introduce SparseSpec-L, a training-free self-speculative decoding framework for long-context inference. SparseSpec-L generates lightweight drafts directly from the target model using a dynamically sparsified and recallable KV cache. It recycles per-head attention statistics produced during full-context verification as a no-extra-forward importance signal, allowing critical historical tokens to be recalled without permanently discarding the dense KV cache. An online entropy-based controller further selects the speculation length according to expected step-wise efficiency. Experiments across multiple long-context tasks and model scales show consistent end-to-end acceleration, with up to speedup over autoregressive decoding while preserving the target model's output distribution.

Yuesong Liu, Yuan Zeng, Min Lyu +3

University of Science and Technology of China

Long-Context Language Model Inference Speculative Decoding LLM Inference Acceleration

Jul 30, 2026cs.AR

LightRot: A Light-Weighted Rotation Scheme and Architecture for Accurate Low-Bit Large Language Model Inference

As large language models (LLMs) continue to demonstrate exceptional capabilities across various domains, the challenge of achieving energy-efficient and accurate inference becomes increasingly critical. This work presents LightRot, a lightweight rotation scheme and dedicated hardware accelerator designed for low-bit LLM inference. The proposed architecture integrates Grouped Local Rotation (GLR) and Outlier Direction Aligning (ODA) algorithms with a hierarchical Fast Hadamard Transform (FHT)-based rotation unit to address key challenges in low-bit quantization, including the energy overhead of rotation operations. The proposed accelerator, implemented in a 28nm CMOS process, achieves a peak energy efficiency of 27.4 TOPS/W for 4-bit inference, surpassing prior state-of-the-art designs. Unlike conventional approaches that rely on higher-precision inference or evaluate on basic language modeling tasks like GPT-2, LightRot is optimized for advanced models such as LLaMA2-13B and LLaMA3-8B. Its performance is further validated on MT-Bench, demonstrating robust applicability to real-world conversational scenarios and redefining benchmarks for chat-based AI systems. By synergizing algorithmic innovations and hardware efficiency, this work sets a new paradigm for scalable, low-bit LLM inference, paving the way for sustainable AI advancements.

Sangjin Kim, Yuseon Choi, Jungjun Oh +2

PIM Semiconductor Design Research Center (AI-PIM), Daejeon 34141, South Korea

LLM Quantization AI Accelerator Inference Rotation-Based Quantization LLM Inference Acceleration Low-Bit Quantization

Jul 30, 2026cs.AR

GyRot: Leveraging Hidden Synergy between Rotation and Fine-grained Group Quantization for Low-bit LLM Inference

Low-bit quantization is essential for efficient LLM inference, and both rotation and fine-grained group quantization have shown individual promise. However, their combination often leads to accuracy degradation or hardware overhead due to a mismatch between the global nature of rotation and the localized behavior of group scaling. We propose GyRot, a quantization framework and hardware accelerator that bridges this gap through algorithm-hardware co-design. GyRot introduces Coarse Rotation, Fine Grouping (CoRFiG) and Harmonic-Aligned Permutation (HAP) to enable cooperative integration of rotation and group quantization, enhancing quantizability while relaxing scaling factor precision. To further reduce hardware cost, we reformulate asymmetric quantization and introduce a zero-point rounding strategy that enables fully integer dequantization. Implemented on an INT4-based tensor PE architecture, GyRot achieves state-of-the-art 4-bit accuracy across LLaMA-family models, while delivering up to 3.4x speedup and 3.6x energy efficiency over baseline LLM accelerators. These results validate GyRot's practical effectiveness for scalable and energy-efficient LLM deployment.

Sangjin Kim, Yuseon Choi, Byeongcheol Kim +2

School of Electrical Engineering, Korea Advanced Institue of Science and Technology (KAIST)

AI Accelerator Inference Rotation-Based Quantization LLM Inference Acceleration Low-Bit Quantization Neural Network Quantization

Jul 30, 2026cs.LG

Back from the Future: Key-Value Cache Management by Counter-Causal Surprise

Key-value (KV) cache management through compression and eviction strategies has emerged as an important research direction in recent years. Computational demands of large language models (LLMs) and their multi-modal variants during output generation can be partially alleviated by caching previous key and value calculations needed by subsequent scaled dot-product attention operations. However, this leads to another problem: the size of the resulting KV cache grows linearly with context length and quickly consumes all available GPU memory when either the prompt or the generated output are long. KV cache management periodically prunes entries from the cache thereby reducing its memory footprint while attempting to retain sufficient information for accurate generation. A by-product is faster inference speed. We propose a simple yet effective KV eviction scheme motivated by the insight that past tokens which can be well-predicted from more recent tokens are redundant and their associated keys and values can be removed from the cache. To score entries for eviction we run the model on the tokens in their original order, reusing the key and value representations already stored in the KV cache, and applying a counter-causal attention mask so that each position attends only to its future context. This is in-distribution, tied directly to the actual cache contents, and requires no additional training. To further reduce cost, we additionally propose a fast single-layer approximation that restricts the counter-causal pass to the last transformer layer, achieving a significant speedup per refresh cycle at marginal accuracy cost. We evaluate our strategy on various open-source LLMs and benchmark datasets showing competitive or improved performance over other state-of-the-art methods. Reference code is available at https://github.com/metacognitionai/counter_causal.

Stephen Gould, Anton van den Hengel

Metacognition AI · Australian National University · Adelaide University

KV Caching KV-Cache Eviction KV-Cache Management LLM Inference Acceleration KV-Cache Pruning

Jul 30, 2026cs.LG

Prox: Training-Free FFN Activation Sparsity via Approximate Intermediate-Channel Salience in LLMs

Feed-forward networks (FFNs) dominate memory traffic and computation in large language model (LLM) inference, making them a primary target for activation sparsification. However, existing training-free methods suffer substantial model-quality degradation at high sparsity due to limitations in their channel-selection strategies. We observe that the SwiGLU intermediate state provides a highly effective channel-selection signal, but obtaining it requires costly dense computation. To address this, we present \emph{Prox}, a two-stage training-free framework for sparse SwiGLU FFNs. Prox hinges on the key insight: sparse execution requires only the channel mask induced by the intermediate state, which can be constructed from the magnitude ranking of its entries rather than their exact values. Specifically, Stage 1 uses input sparsity and quantized proxy weights to construct a shared mask; Stage 2 computes the selected channels exactly, enabling sparse execution of all three projections. Across ten LLMs from six model families, Prox outperforms training-free baselines at all sparsity levels, achieves up to a

1.99\times

end-to-end decoding speedup at 70% FFN sparsity, and is compatible with quantization and sparse attention.

Jinyi Liu, Wei Chen, Pengyu Chen +4

Institute of Software, Chinese Academy of Sciences · University of Chinese Academy of Sciences · School of Computer Science, Nanjing University of Posts and Telecommunications

Activation Sparsity Structured Sparsity LLM Inference Acceleration Neural Network Pruning

Jul 30, 2026cs.AI

DeepResearch Agent System

The DeepResearch Agent System is a large language model system engineered for deep information retrieval, multi-step reasoning, and autonomous research tasks. Built upon a sparse activation architecture with 30 billion total parameters of which only 3 billion are activated per token, the system achieves state-of-the-art performance on multiple agent search benchmarks while delivering 3.2 times faster inference compared to dense counterparts of equivalent scale. The system supports a 128K-token context window with hierarchical attention mechanisms that yield 18.7% accuracy and 23.4% recall improvements over standard long-context approaches. A dual-mode reasoning engine provides both a ReAct paradigm for basic multi-step problem solving and an IterResearch mode for high-performance iterative research with up to 20 reasoning steps, collectively delivering a 31.2% accuracy improvement over single-pass baselines. Multi-tool coordination integrates retrieval, computation, web search, and file parsing modules to achieve 92.1% tool-use accuracy. A reinforcement learning optimization framework based on the GRPO algorithm provides token-level policy gradients that improve training stability by 35% and accelerate convergence by 42%. An automated data synthesis pipeline with seed-based expansion achieves a 92.5% usability rate. Benchmark results include 87.3% on Humanity's Last Exam, 85.3% on BrowserComp Chinese, and 91.2% on WebWalkerQA. The system is fully open-sourced, including data synthesis, training, and inference code, and supports applications in academic research, business analysis, R&D support, and education.

Yong Huang, Yulu Huang, for the team Collaboration

Deep Research Agents LLM Inference Acceleration Autonomous Research Agents

Jul 29, 2026cs.DC

InferScale: GPU-Native KV Injection for Personalized LLM Serving

Large language models are increasingly deployed with persistent personalized context, such as accumulated memory profiles or long conversation histories, that is shared across a user's many requests. Production memory systems (e.g., Mem0, MemGPT, and Zep) retrieve a relevant subset of this memory and inject it into the prompt, forcing the serving engine to repeatedly prefill the same content. As the retrieval budget grows, time-to-first-token (TTFT) increases even though the underlying memory is reused across requests. We present InferScale, a GPU-native LLM memory system that replaces repeated prompt prefilling with reusable KV state. InferScale precomputes each memory fact's KV representation, stores it alongside a semantic embedding on the GPU, retrieves relevant facts at serving time, and injects their KV directly into vLLM's paged cache. To support dynamically assembled memories under rotary position embeddings, we introduce Chunked RoPE, which stores keys before rotation and applies their serving-time positions during injection. However, encoding memory facts independently omits the cross-fact context available during joint prefilling. We mitigate this with Context-Window Encoding, which encodes each memory fact together with a small window of preceding conversation context while caching only the target fact's KV. InferScale is implemented through vLLM's KV-connector interface, requiring neither engine modifications nor model fine-tuning. Across three open-weight models on LoCoMo, InferScale keeps TTFT nearly constant as the retrieval budget increases: at k=50 it reduces TTFT by 72-79% (3.6-4.8x), achieves 60.3% accuracy versus 63.3% for Mem0 without serving-time recomputation, and delivers 3.7-4.5x the throughput under concurrent load. Reusable KV state thus decouples memory-conditioned serving latency from retrieved-context size while preserving application quality.

Peter Li, Prashant Pandey

Northeastern University Boston, Massachusetts, USA

LLM Serving Memory-Augmented Language Models KV Caching Persistent Memory for Language Models LLM Inference Acceleration KV-Cache Reuse

Jul 29, 2026cs.LG

Beyond KV Reconstruction: Functional Reconstruction for MLA Draft Models in Speculative Decoding

Multi-head latent attention (MLA) is increasingly important for long-context LLM inference because compact latent states replace the growing key-value (KV) cache and reduce decoding memory traffic. Yet most capable open checkpoints use multi-head or grouped-query attention (MHA/GQA), so conversion is needed to obtain MLA's cache efficiency without retraining from scratch. Speculative decoding offers complementary acceleration, but its speedup depends on agreement between draft proposals and target verification. We find that direct MHA/GQA-to-MLA conversion can sharply reduce this agreement: low-rank factorization and RoPE handling introduce attention-function errors that may be tolerable for standalone generation but substantially lower draft-token acceptance. We therefore formulate MLA draft construction as functional reconstruction rather than cache compression. Our end-to-end (E2E) method optimizes each converted MLA attention module to reproduce the post-output-projection response of its original MHA/GQA counterpart on calibration hidden states. This converter-agnostic post-conversion procedure preserves the converted cache and inference graph and requires neither verifier logits nor verifier supervision. We evaluate 192 model-converter-backend-method-task configurations spanning four Llama/Qwen draft-target pairs, TransMLA and MHA2MLA, HF and vLLM, and four 200-prompt tasks. With a 0.5-percentage-point reporting tolerance, Functional Reconstruction materially improves acceptance in 37 of 64 matched task cells, leaves 26 practically unchanged, and materially decreases one. Code and evaluation artifacts are available at https://github.com/swyhahaha/FunctionalMLA.

Weiye Shi, Fanxu Meng, Muhan Zhang

Institute for Artificial Intelligence, Peking University

Speculative Decoding LLM Inference Acceleration Attention Mechanisms Multi-Head Latent Attention

Jul 29, 2026cs.CL

Revisiting Lossy Verification in Speculative Decoding: Mechanisms, Trade-offs, and Failure Modes

Speculative Decoding (SD) accelerates large language model inference by allowing a lightweight draft model to propose tokens that are subsequently verified in parallel by a larger target model. Recent approaches introduce lossy verification schemes to further improve efficiency by relaxing strict distributional matching. Yet such relaxation silently rewrites the decoding distribution, and the resulting acceleration can come at the cost of unstable, sometimes severely degraded generation quality. In this work, we present a principled analysis of the distributions induced by lossy verification methods. We show that many seemingly distinct approaches differ only superficially and can be classified into two categories: truncation-based verification and collaborative verification. We further construct a diagnostic evaluation framework across curated benchmarks. For truncation-based methods, we identify a fundamental pitfall: performance can degrade significantly compared to the true truncation sampling baseline due to distributional distortion. For collaborative verification, we uncover a key principles: controlling the overshoot of draft probabilities relative to target probabilities is essential to prevent low-quality outputs. Our code is available at https://github.com/ZhouYuxuanYX/Fast-HSD.

Tianyu Wang, Yuxuan Zhou, Wenbin Wang +3

1Independent Researcher · 2Baidu Inc. · 3Zhejiang University

Speculative Decoding LLM Inference Acceleration

Jul 29, 2026cs.IR

OneLatent: Latent Reasoning for Efficient Foundation Recommendation Models

Large language models (LLMs) have demonstrated strong reasoning capabilities, motivating their use as the backbone of foundation recommendation models (FRMs). Existing methods enhance recommendations through explicit Chain-of-Thought (CoT) reasoning under a Think-then-Answer paradigm. However, explicit CoT incurs substantial inference overhead by generating lengthy reasoning traces and relies on manually designed templates that struggle to capture diverse, dynamic user interests. We propose OneLatent, an efficient latent reasoning framework that compresses explicit reasoning traces into several learnable latent tokens, enabling Latent-Reason-then-Answer inference without generating verbose traces. OneLatent first introduces Multi-View Adaptive CoT (MV-ACoT), which creates diverse, high-quality teacher-generated supervision by exploring user interests from multiple perspectives and automatically adapting reasoning complexity to each instance. Building on pretrained FRMs, it then uses a three-stage latent-token alignment paradigm to progressively internalize CoT traces into learnable latent tokens. Finally, a multistage curriculum-based post-training strategy activates latent-token reasoning for downstream recommendation tasks. Experiments on an industrial-scale Kuaishou dataset and the public Kuaishou LLM-Rec benchmark show that OneLatent consistently outperforms explicit CoT-based methods and traditional baselines. Compared with the Think and No-Think variants of FRMs, OneLatent improves SID@64 by 17.44% and 9.33%, respectively, while achieving over 17x higher online inference throughput. We further develop a production serving system for scalable, real-time FRM inference. An online A/B test in Kuaishou's local-services advertising scenario shows that deploying OneLatent with this system yields an estimated 9.6% revenue lift over strong online baselines, including OneRec and OneReason.

Hao Jiang, Peiru Du, Pengfei Yao +10

Kuaishou Technology, Beijing, China

LLM Inference Acceleration Large Language Model-Based Recommendation Latent Reasoning

Jul 28, 2026cs.LG

Spend Experts Where You Are Unsure: Confidence-Adaptive Routing for Mixture-of-Experts LoRA

Mixture-of-Experts (MoE) variants of Low-Rank Adaptation (LoRA) route every token to a fixed number of experts

k

. Tokens differ in how uncertain the model is about them, so a single k over-spends on easy tokens and under-serves hard ones. We observe that the router's output distribution is already a per-token uncertainty signal: peaked mass indicates confidence, while a flat distribution indicates ambiguity. We introduce CARE (Confidence-Adaptive Routing of Experts), which admits experts in a nucleus fashion. Experts are activated in decreasing router weight until their cumulative mass reaches a threshold, with a small extension when the admitted experts disagree. A budget thermostat calibrates the threshold so that the average number of active experts matches any target. CARE is a drop-in, single-forward-pass rule with no extra parameters. Across eight commonsense benchmarks on LLaMA-3.1-8B and Qwen2.5-7B, as well as math, code, and knowledge tasks, CARE improves over fixed top-k MoE-LoRA at matched compute and matches the fixed-k=4 baseline while activating fewer experts. The same confidence and disagreement signals also improve out-of-distribution detection over MSP, entropy, and multi-pass proxies. We support the design with nucleus fidelity, budget optimality, and an epistemic reading of disagreement, and we release code.

Tom Saliencro, Rohan Desai, Priya Nair +2

University of California, Irvine · University of Washington

Parameter-Free Mixture-of-Experts Routing LLM Inference Acceleration Expert Routing LLM Uncertainty Estimation Mixture-of-Experts Models

Jul 28, 2026cs.CL

AngelSpec: Towards Real-World High Performance Inference with Speculative Decoding

Speculative decoding accelerates large language model inference without changing the target distribution, but no single drafting structure performs best across real-world workloads. Autoregressive multi-token prediction (MTP) is a lightweight, stable proposal mechanism, whereas block-parallel diffusion amortizes drafting latency over much longer candidate sequences; the better choice depends strongly on the output distribution. We present AngelSpec, a unified training framework for MTP and block-parallel speculative decoding that addresses this heterogeneity at three levels. At the training level, rather than fitting one universal drafter to a uniform data mixture, we co-specialize structure and data: the MTP drafter is trained on diverse conversational data for high-entropy open-ended chat, and the block-diffusion drafter on code and mathematics data for longer predictable continuations. At the architecture level, we propose DFly, a block-diffusion framework combining a hybrid target-conditioning backbone with a predecessor-conditioned autoregressive head, improving target-feature utilization and intra-block dependency modeling while keeping generation parallel. At the inference level, both acceptance length and verification cost vary with domain, request, online load, and hardware, so DFly treats verification as a shared batch-level resource: it reallocates compute toward high-confidence prefixes across requests and combines expected utility with a profiled cost model to adapt verification depth online. Across the Hy3 series, DFly raises the average accepted length on Hy3-A21B by roughly 30% and attains the highest average throughput at every tested concurrency from 4 to 64, a 1.98-2.40x speedup over autoregressive decoding and 10.5-11.8% higher throughput than DFlash. We release AngelSpec to support training and extending these methods.

Hong Liu, Rui Cen, Junhan Shi +10

Tencent Inc.

LLM Inference Scheduling Speculative Decoding for Diffusion Language Models Speculative Decoding LLM Inference Acceleration Diffusion Language Model Decoding

Jul 28, 2026cs.AR

At-the-Roofline Sparse Tensor Contractions on Vector Processors for Transformer Inference

Fine-grained weight pruning and activation sparsification have emerged as effective approaches for reducing the compute and memory cost of inference for Transformer models. In the moderate-sparsity regime, Gustavson's dataflow provides a natural execution model for exploiting both activation and weight sparsity on vector processors through metadata-driven indexed accumulation. However, existing RVV architectures lack native support for this pattern, forcing kernels to rely on software index decoding and L1-backed indexed memory operations that keep sparse tensor contractions far below their roofline performance bound. We present Ventaglio, a runtime-configurable sparse execution unit coupled with RVV ISA extensions that drives sparse tensor contractions toward their roofline through indexed gather-accumulate-scatter support. Integrated into an open-source vector processing cluster and implemented in 12nm FinFET, Ventaglio accelerates sparse tensor contraction kernels by

6.9\text{--}7.4\times

over optimized RVV baselines, with only

3.1\%

area overhead for a cluster of tightly-L1 coupled vector processing elements. We build a performance-accurate instruction-level model of the Ventaglio extension, calibrate it against RTL implementation, and leverage it for scale-out performance analysis on a large

4\times4

multi-cluster system. Using a DuoGPT-pruned LLaMA-3-8B model with practical

40\text{--}60\%

dual sparsity, Ventaglio achieves

2.40\text{--}5.25\times

and

2.06\text{--}3.16\times

speedup over dense baselines during prefill and autoregressive decoding, respectively.

Bowen Wang, Chi Zhang, Diyou Shen +3

*IIS, ETH Zürich, Switzerland · ‡Computing Systems Lab, Huawei, Switzerland · †DEI, Universit`a di Bologna, Italy

Efficient Neural Network Inference Transformer Inference Activation Sparsity LLM Inference Acceleration

Jul 28, 2026cs.CL

CoSA: Accelerating Long-Context Inference via Proxy-Kernel Co-Designed Sparse Attention

The quadratic cost of self-attention makes long-context inference prohibitively expensive, and proxy-based block-sparse attention has become a practical remedy. Existing methods typically rely on a proxy to predict a binary sparse mask and a kernel to consume this mask and perform sparse attention computation. Such an approach is effective under moderate budgets. However, as the budget tightens, the estimated proxy inevitably drops some salient blocks, while the kernel can only apply the sparse mask mechanically, leading to an evident drop in model accuracy. We propose CoSA, a two-stage training-free Sparse Attention under proxy-kernel CO-design, which couples a Kernel-Aware Proxy (KAP) with an Ordered-Skipping Kernel (OSK). In the first stage, the KAP selects blocks under a moderate budget and produces an ordered mask that prescribes the order in which KV pages are visited in the kernel inner loop. In the second stage, the OSK applies this mask and skips more blocks under a tightened budget given online-softmax statistics. Across mainstream LLM backbones and long-context benchmarks, CoSA attains higher accuracy at lower budgets. Impressively, CoSA achieves a 4.93

\times

attention speedup and reduces end-to-end Time-to-First-Token by 2.53

\times

under a context length of 128K with negligible performance degradation. Code is available at https://github.com/Tencent/AngelSlim.

Yufei Xue, Lin Niu, Hong Liu +6

1Tencent · 2The Hong Kong University of Science and Technology

Long-Context Language Model Inference LLM Inference Acceleration Sparse Attention Block-Sparse Attention

Jul 27, 2026cs.CL

PIVOT: Efficient Query-Group Indexing for Token-Level Sparse Attention

Token-level sparse attention, as implemented by DeepSeek Sparse Attention (DSA) in production systems, makes the downstream attention efficient but shifts the bottleneck to the indexer that feeds it. To select the top-k tokens for each query, the indexer must still score every preceding token, incurring a cost of O(L^2) per layer for a sequence of length L. We observe that this per-query scan is largely redundant: nearby queries select highly overlapping top-k tokens, and the indexer scores are long-tailed along the key axis. We exploit these properties in PIVOT, Proxy Indexing Via One full-prefix Traversal, a training-free, drop-in replacement for the DSA indexer that shares one prefix scan across a group of nearby queries. PIVOT aggregates a group into a single proxy query, performs one shared full-prefix scan to obtain a candidate set, and then selects a top-k for each query from that set. Two variants trade speed for fidelity: PIVOT-Reuse shares the proxy top-k across the group for maximum speed, whereas PIVOT-Refine re-scores the candidate set with the indexer of each query and then selects an individual top-k, matching the dense indexer at a small additional cost. A single algorithm covers both inference phases, differing only in how groups are formed: fixed-size groups of consecutive queries in prefill, and the queries decoded together in one multi-token prediction (MTP) step in decode. On DeepSeek-V3.2 and GLM-5.1 across LongBench and RULER, PIVOT matches the accuracy of the dense DSA indexer while accelerating it by up to 4x and reducing end-to-end latency by up to 1.6x at long context.

Hong Liu, Yuan Cheng, Lin Niu +5

Tencent

LLM Inference Acceleration Efficient Attention Sparse Attention

Jul 27, 2026cs.LG

LOCKS: Page-Local Compact Key Summaries for Efficient Long-Context Decoding

Serving large language models at long context is bottlenecked by the key-value (KV) cache, which is read at every decode step. We find that attention keys are approximately low-rank within pages. A single low-rank projection shared across pages can miss page-specific directions; fitting a basis to each page better identifies the pages receiving the most attention at comparable stored selector cost. LOCKS stores a rank-

r

spectral summary per page, reconstructs its within-page logits, and selects pages by log-sum-exp mass without reading candidate keys or values. It stays within about a point of FullKV on LongBench-v1, tracks the read-every-key exact-LSE oracle on RULER down to the smallest budgets, and retains quality furthest under tight budgets on AIME26 and MATH-500. At a

2048

-token budget it matches FullKV aggregate quality beyond

100

K context while attending about

2\%

of tokens. Across ranks

2

-

8

, summaries use

4

-

10\%

of full-KV bytes. On GH200 with GPU-resident KV, LOCKS reduces complete decode-step time by

1.8\times

at

512

K context. With full KV offloaded to Grace memory, it reaches

3.82

-

4.22\times

the faster dense backend's aggregate throughput at

64

K-

256

K by serving larger batches.

Junsung Hwang

Low-Rank Attention KV Caching Long-Context Language Model Inference LLM Inference Acceleration Sparse Attention Low-Rank Approximation

Jul 27, 2026cs.LG

DraftExpert: Expansion-Aware Self-Speculative Decoding for End-Device MoE Inference

Large Mixture-of-Experts (MoE) language models are attractive for end-device deployment because only a small subset of experts is active per token, but their routed expert weights often exceed accelerator memory. We target latency-critical single-user settings where routed experts are staged on demand from CPU memory to a GPU or from Flash to a mobile NPU. In this setting, self-speculative decoding faces a new bottleneck: increasing the draft expert set improves accuracy but triggers extra expert loading, while cheap small-footprint drafts have low acceptance; moreover, verifying a multi-token block activates the union of target experts and is no longer close to one target step. We propose DraftExpert, an expansion-aware self-speculative decoding framework for expert-offloaded MoE inference. DraftExpert trains one lightweight accelerator-resident draft expert per layer by self-distilling residual, logit/token, and router-agreement signals from the frozen target MoE. At inference time, it uses a fixed-footprint shared+top-1+draft-expert drafter together with confidence--expansion truncation and target-expert prefetching, while final tokens are still exactly verified by the target model. On DeepSeek-V2-Lite and Moonlight-16B-A3B across CPU-GPU and Flash-NPU offload, DraftExpert improves decode throughput by 1.45x on average, raises draft acceptance to 84~~87%, and achieves 86~~88% prefetch hit rates.

Dengke Han

Expert Offloading On-Device Language Model Inference Mixture-of-Experts Inference Speculative Decoding LLM Inference Acceleration

Jul 27, 2026cs.LG

KAP: Bridging the Knowledge Selection-Runtime Consumption Gap in LLM Systems

Modern LLM systems increasingly rely on knowledge-selection processes that produce high-value structured priors, such as ranked evidence, graph topology, multimodal alignment, and confidence signals. Yet LLM serving remains fundamentally oblivious to this rich structure: once such signals are serialized into a prompt, the backend observes only a flat token sequence, forcing dense and uniform consumption of the full key-value (KV) state during decoding. We term this architectural mismatch the Knowledge Selection-Runtime Consumption (KSRC) gap: richer contexts enlarge the full-prompt KV footprint and decode-time memory traffic, increasing latency and degrading throughput even when reasoning depends on only a small fraction of the context. To bridge the gap, we propose Knowledge Access Planning (KAP), a paradigm-shifting execution abstraction that elevates structured knowledge priors from passive prompt-construction hints into first-class physical execution artifacts. KAP establishes a universal intermediate representation (IR)-the runtime access plan-which compiles structured knowledge signals to govern physical KV access without altering logical prompt semantics, model weights, or training procedures. Through this IR, KAP shifts LLM serving from token-aware context consumption to plan-driven, knowledge-aware runtime consumption. We instantiate KAP with GraphSpec, a compiler-executor realization connecting structured knowledge selection to an LLM serving backend. We derive a phase-boundary model for the positive-speedup regime of plan-guided execution. Across 4K-128K long-context QA workloads, GraphSpec maintains answer quality comparable to full-context decoding while decoupling physical KV consumption from prompt length, reducing proposal-time KV access to 5.5% of source KV state at 128K, and fundamentally shifting the scaling trajectory of long-context generation.

Shuo Wang, Fang Xi, Wenyuan Huang +2

QiYuanLab Beijing, China

LLM Serving KV-Cache Management LLM Inference Acceleration Long-Context LLM Inference

Jul 26, 2026cs.CL

A Frozen 12B Beats Frontier Models on Verified Work: 100% Accuracy, 0 Tokens, Bit-Exact, Forever

Improving a language model today means retraining it: enormous compute, a new opaque model each cycle, non-deterministic output. We take the opposite path: the model stays frozen, and a persistent memory of verified solutions grows beside it. Once a problem family is solved and has passed an independent verification step that never consults the answer key, every new instance of that family is answered at zero generation tokens, bit-exact, deterministically. Across 180 fresh instances spanning nine problem families, four architectures from four vendors - dense and mixture-of-experts - each score 180/180 at zero generation tokens per answer: execution-bound capability decoupled from parameter scaling. A negative control attributes the capability fully to the memory: emptied, it solves nothing. The same verify-before-store contract holds for open-ended reasoning: 88/88 consistency-gated acceptances across all four models, machine-checked formal proof, and reasoning-method transfer at 77/80. Memory selection takes 1.4 microseconds; a full reuse completes in 6-23 ms at 36 mWh. Approximate similarity retrieval selects the wrong item 94.3% of the time on a 4,500-item verified store where exact addressing makes zero errors. The store also serves as working context at a scale no shipped engine matches: a 6,000,000-token movable window on a single 46 GB GPU at flat memory, where vLLM stops at 30,399 tokens and SGLang silently truncates past 32,000. On published benchmarks, frontier models remain far ahead of any 12B at raw from-scratch reasoning; on everything this system has solved and verified, the comparison inverts: a frontier API call pays a fresh generation pass on every query, forever, while verified reuse costs zero tokens and returns the identical bits every time. A public testbench with free, rate-limited access accompanies this report: https://corbenic-galahad-bench.hf.space

Sietse Schelpe

Corbenic AI

Persistent Memory for Language Models LLM Memory LLM Inference Acceleration

Jul 24, 2026cs.AR

HiKV: Hierarchical Importance-Aware KV Cache with Hardware Acceleration for LLM Decoding

With the rapid adoption of long-context large language models (LLMs), the continuously growing KV cache during decoding has become the critical memory bottleneck. To tackle this challenge, we propose HiKV, a novel algorithm-hardware co-design that exploits KV cache redundancy through hierarchical importance awareness. Algorithmically, HiKV compresses the KV cache at two granularities: Stage I evicts unimportant tokens within a fixed budget, and Stage II further loads only the significant elements of each retained token, reaching compression ratios unattainable at a single granularity. Architecturally, we develop a dedicated accelerator centered on a reconfigurable importance sorter that switches between the distinct sorting datapaths each stage requires, unifying the two-stage acceleration in one circuit with minimal overhead. Evaluated on representative LLMs, HiKV achieves up to 7.95x speedup and 90% energy reduction in the attention computation over the vanilla KV cache baseline within negligible 1% accuracy loss. Under iso-accuracy constraints, HiKV outperforms state-of-the-art importance-based methods by achieving an additional 1.82~4.87x reduction in external memory accesses. These benefits are enabled by specialized hardware components that add only 8% to the system area.

Chao Fang, Jun Yin, Man Shi +1

ESAT-MICAS, KU Leuven, Leuven, Belgium · UC Berkley, CA, USA

KV Caching Memory-Efficient Optimization KV-Cache Eviction LLM Inference Acceleration KV-Cache Compression KV-Cache Pruning