LLM Inference Efficiency

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

47 papers in the last four weeks, up 262% on the four weeks before. 0.5% of all new papers.

Jul 13Week of Sep 28

Latest papers 234

Oct 7, 2026cs.DS

Attention via Black-Box Vector Search

Sparse attention mechanisms estimate attention over

n

tokens using a small subset of keys. Many existing approaches use maximum inner product search (MIPS) to retrieve the heaviest keys, which motivates the following question: given black-box access to a MIPS oracle, how many keys must be retrieved to output an

\varepsilon

-accurate attention estimate? We answer this question by unifying prior approaches through the framework of priority sampling. With a single MIPS index, we show that

Θ(\sqrt{n}/\varepsilon)

retrieved keys are both sufficient and necessary. With

Θ(\log n)

indices, we give an algorithm that retrieves only

O(\log n+1/\varepsilon^2)

keys and prove that this is near-optimal. More generally, we design algorithms that establish a smooth tradeoff between the number of MIPS indices and number of retrieved keys. We then show that if we allow augmentation of keys and queries, we can bypass the above lower bounds: there exists a simple priority-sampling estimator using a single MIPS index and

O(1/\varepsilon^2)

retrieved keys. When integrated into LLM inference, our algorithms outperform top-

k

and sampling approaches used in prior work and yield attention approximation that scales favorably to long contexts.

Stepan Zharkov, Krish Singal, Ashwin Padaki +1

University of Pennsylvania

Efficient Attention Sparse Attention LLM Inference Efficiency

Oct 6, 2026cs.CL

ReFold: Training-Free Reversible Inter-Turn Context Folding for Long-Horizon Agents

Long-horizon LLM agents act on an append-only interaction history that is re-sent to the model at every step, so the context and its cost grow with steps until the sessions exceed the context window. Existing methods manage the context through context requirement prediction, relying on additional model calls, heuristic rules, or trained policies. However, these predictive approaches introduce runtime overhead, invalidate prefix caches, and permanently discard content with no guarantee of recovery. To overcome these limitations, we introduce ReFold: a training-free rendering layer that preserves the underlying interaction history while compressing only the model's rendered context. It removes two kinds of inter-turn redundancy without an auxiliary predictor: content an earlier turn already displayed, replaced by a stub, and turns the agent itself reports finished, folded into a one-line note. Both operators use chunked rendering, rewriting the cached prefix once every few steps rather than at every step. Every removal is strictly reversible, a wrong removal costs one restore from the history rather than permanent content loss. Because it operates at the rendering layer, ReFold is plug-and-play across standard ReAct-style harnesses. Evaluations across five long-horizon benchmarks and two frontier LLMs demonstrate that ReFold reduces token consumption by up to 2.5x and halves the KV-cache memory per session without degrading task success rates. Under capped context budgets, it avoids up to 92% of forced compactions. Under concurrent serving workloads, it reduces request queuing delays by up to 100%, accelerating inference by up to 1.7x, while cutting inference costs by up to 3.4x.

Yupeng Su, Jiayi Tian, Zheng Zhang +1

University of California, Santa Barbara · Intel

LLM Inference Efficiency Long-Context Language Model Inference Long-Horizon LLM Agents LLM Context Management Context Compression

Oct 6, 2026cs.AI

Persistent Memory in Multi-Agent LLM Inference: What It Costs, What It Buys, and When You Can Tell

Decomposing long-context inference across cooperating agents bounds the active KV cache per call rather than total evidence, which matters when KV-cache memory binds. Many such systems add a persistent tier storing and recalling reasoning traces, usually validated by an ablation reporting an accuracy gain. We measure both on one three-tier agent architecture. Decomposition delivers: peak KV working set of 14.3 MiB per query against 35.5 and 35.3 MiB for single-pass and retrieval-augmented baselines. The persistent tier does not: across eight controlled dataset pairs at n=100 per arm it costs +0.368 MiB [+0.167, +0.590] of peak cache and produces no detectable accuracy change (+0.015, 95% CI [-0.011, +0.046]). We argue the null is structural: single-question benchmarks supply each item with its own evidence and score it independently, and correctness requires resetting stored traces between conditions, so recall has nothing informative to retrieve. Reaching it took four measurement corrections -- three inflating the apparent benefit, the fourth making an effect that size look resolvable -- none visible in the results table. We give the conditions an agent-memory ablation must satisfy and detection procedures that need no knowledge of the specific defect.

Hochan Son, Kyungdoe Han, Jaehan Koh +3

University of California, Los Angeles · University of Wisconsin · HCLTech America

LLM Inference Efficiency Multi-Agent LLM Systems Persistent Memory for Language Models Memory Agent Benchmarks

Oct 6, 2026cs.CL

Monte Carlo Estimation for KV Cache Eviction

Most KV-cache eviction methods ask, in effect, which memory appeared important while reading the prompt? We instead ask, which memory will matter while answering? Since decoding queries are unavailable at eviction time, prior future-aware methods rely on pseudo-responses or synthetic future-query estimates. We cast fixed-budget future-aware eviction as distributional estimation over plausible model-conditional query trajectories and introduce LORE-KV (Lookahead Output-perturbation with Reliability-weighted Ensembles for Key-Value caches), a training-free method that samples short autoregressive continuations from the frozen target model and uses their response-side query states to estimate prompt-token utility. Tokens are scored by projected leave-one-out attention-output deletion cost and aggregated across sampled futures with optional trajectory weighting. The temporary continuations are discarded before final decoding, requiring no auxiliary model or training. Ablations isolate the mechanism: at B=128, a single response-side continuation recovers about 89% of the gain over the prompt-window control, while additional futures provide smaller improvements. At B=128, LORE-KV raises the LongBench average on Qwen2.5-14B from 45.49 to 48.24 (+2.75) and the 16K RULER average on Mistral-7B from 45.20 to 51.05 (+5.85). Gains diminish at larger cache budgets and coexist with task-level regressions. LORE-KV incurs 1.46-2.77x AnDPro's per-sample wall-clock time as a one-time compression overhead across six dense and hybrid-attention backbones.

Ahsan Bilal, Muhammad Ahmed Mohsin, Muhammad Umer +5

University of Oklahoma · Stanford University · University of Houston +3

LLM Inference Efficiency KV-Cache Eviction

Oct 5, 2026cs.AR

Evaluating Inference Compute for Generative AI: A Framework for Enterprise Workloads

LLM deployment is shifting from single-turn completion to agentic trajectories in which a model plans, calls tools, reads results and reasons at test time before acting. This inverts the economics of inference hardware: chat serving amortises weight reads across large batches, whereas agent trajectories are sequentially dependent, run at effective batch one, and make per-token decode latency (TPOT) the dominant term in task completion time. Using a roofline analysis and a closed-form episode-latency model, we show why this regime favours accelerators that keep weights in on-die SRAM (Cerebras WSE-3/3T, Groq/NVIDIA LPU) or compiler-managed tiered memory (SambaNova SN40L/SN50), and why three vendor ecosystems converged in 2026 on disaggregated prefill/decode serving. We show that per-step reliability compounds exponentially in trajectory length-a 2% per-step failure rate erases a 2x decode advantage for a 20-step agent-so determinism and tail latency are first-order performance variables. We then propose a four-layer evaluation framework (silicon, serving system, agent episode, enterprise) with a metric set built on goodput at an agentic SLO and cost per successful episode, a six-axis benchmark protocol over six task families, a paired-bootstrap statistical design, an attestation protocol for vendor-run benchmarks, and TCO, availability and adoption-timing models with explicit break-even conditions. All performance figures are public and labelled by evidence class; we state seven falsifiable hypotheses and the experiments that test them, and argue that the most likely original result is that token-throughput rankings diverge from cost-per-successful-task rankings on long-horizon work.

Abbas Raza Ali, Muhammad Ajmal Siddiqui, Moona Zahid

Citigroup Inc., London, United Kingdom · Ernst & Young LLP, London, United Kingdom · NVIDIA Corporation, London, United Kingdom

LLM Inference Efficiency Disaggregated LLM Serving LLM Agent Evaluation Cost-Aware Inference

Oct 4, 2026cs.LG

Expanding LLM Reasoning

Extra inference compute is usually spent on sampling more reasoning chains. We study where inside an existing chain an additional continuation should begin. We define expansion utility, the change in correctness from restarting a chain at a stored step, and measure it at every eligible step for nine models on six benchmarks (41 model and benchmark cells). Restart position matters: steps selected on one set of continuations beat uniform placement when scored on disjoint ones, in held-out audits on 5, 16, and 38 cells (+4.25 points [+2.51, +6.63] in a fresh five-cell audit). A fixed rule that restarts from the last eligible steps, always-last, is a strong baseline: our learned router beats uniform placement but shows no detected gain over it, and on DeepSeek-R1-Distill-Qwen-14B/MATH-500 always-last exceeds the exact self-consistency frontier at matched aggregate generated output by +0.052 [+0.008, +0.098], using 0.774x the aggregate generated output of four-sample self-consistency. Cross-fitted oracle selection still finds held-out headroom beyond declared positional classes, a target for future selectors. Finally, breaking step-label ties by earliest index flips the sign of a pointwise selector's gain over uniform placement in every seed of a five-seed diagnostic with four rollouts per step; randomized ties remove the bias.

Rian Atri, Evan Luo

Keiji AI · University of California, Berkeley

LLM Inference Efficiency Inference-Time Scaling Efficient Language Model Reasoning Inference-Time Compute Allocation LLM Reasoning

Oct 1, 2026cs.AI

Beyond State-of-the-Art: Standardising Environmental Impact Metrics for AI Research

As the capabilities and ubiquity of Large Language Models (LLMs) grow, so does their environmental footprint. Despite calls for responsible AI, the machine learning community lacks standardised practices for carbon accounting. Our automated literature review of the 5,285 papers accepted to NeurIPS 2025 reveals that reporting of environmental impact is nearly non-existent. To catalyse a shift toward sustainable AI, we define standardised sustainability metrics for evaluating model training efficiency, accompanied by simple heuristics to estimate the carbon cost of LLM inference. We implement these metrics in carbonbenchmark, a drop-in software solution for tracking and reporting emissions. Finally, to combat the pursuit of marginal accuracy gains at disproportionate environmental costs, we formalise the

Smallest Model that Achieves the Job' (SMAJ), a framework which challenges the field to prioritise computational efficiency and environmental accountability alongside traditional

State-of-the-Art' (SotA) accuracy.

Lachlan McGinness, Dan Pagendam, Robert Offner

School of Computing, Australian National University · Commonwealth Scientific and Industrial Research Organisation

LLM Inference Efficiency Energy-Efficient ML Efficient Language Model Training

Oct 1, 2026cs.AI

From Discovery to Decision: Finite-Budget Recoverability in LLM Voting

Voting over multiple LLM responses is a common primitive in test-time scaling and ensemble inference. Collecting more responses can expand the candidate pool and increase the chance that a correct answer is discovered. Under a fixed call budget, a discovered answer still needs to accumulate enough support within the remaining calls to become the final plurality winner, creating a discovery-to-decision gap. In this work, we characterize this gap through the realized vote state and remaining call budget. We derive a sharp recoverability threshold and show that, as sampling proceeds, the observed candidate set can only expand while the set of reachable endpoint winners can only contract, inducing a candidate-level conversion window. Under a specified iid response law, the same state yields exact finite-horizon endpoint probabilities. We further show that merging wrong-answer identities preserves single-call correctness and cannot improve plurality accuracy, and that the effect of redistributing wrong-answer probability depends on the realized vote state. Singleton reachability yields a gold-free exact locking certificate. For a known answer universe, its first trigger is the earliest prefix at which all admissible continuations yield the same fixed-budget output. Empirically, most discovered-but-unselected correct answers lose reachability only after discovery. In a controlled Word16 study, input permutation improves raw-plurality accuracy by 21.1 points with essentially unchanged single-call correctness. Exact locking saves 28-30% of calls at a 16-call budget while preserving every fixed-budget output.

Shaoang Li, Jian Li

Stony Brook University

LLM Inference Efficiency LLM Inference LLM Decision-Making Language Model Ensembles

Oct 1, 2026cs.CL

Beyond Leaderboards: Tokenomics of Agentic Small Language Model Ensembles

As large language models (LLMs) move from standalone assistants into agentic workflows, evaluation must extend beyond scalar leaderboard accuracy to account for operational reliability, cost, latency, and token efficiency. We use an agentic ensemble of small language models (SLMs) with an SLM-judge-mediated feedback loop as a case study for such beyond-leaderboard evaluation. On the 541-prompt IFEval benchmark, the best ensemble achieves 97.34% strict prompt accuracy, exceeding the strongest standalone LLM baseline, gpt-5.4, by 5.81 percentage points while operating in a lower-cost regime. We then analyze the tokenomics and operational behavior behind this gain, including cost per sample, token composition, useful-output goodput, feedback-loop recovery, latency decomposition, and performance across instruction categories and constraint counts. Our results show that agentic SLM ensembles can trade additional test-time tokens and orchestration overhead for improved instruction-following fidelity, motivating multi-dimensional evaluation protocols for future agentic AI systems.

Alexei N. Skurikhin, Emily M. Taylor, Nathan A. DeBardeleben

Los Alamos National Laboratory Los Alamos, New Mexico, USA

LLM Inference Efficiency Language Model Generation Evaluation AI Agent Evaluation Small Language Models Instruction Following Language Model Ensembles

Sep 30, 2026cs.LG

Characterizing High Bandwidth Flash for LLM Serving

Large language model (LLM) serving requires substantial memory to store model weights and KV caches. As models grow larger and contexts become longer, memory capacity and bandwidth increasingly become bottlenecks for serving performance. Agentic workloads compound this pressure through repeated interactions over growing contexts, making it increasingly important to retain KV state for reuse. High-bandwidth flash (HBF) offers a way to expand accelerator memory capacity for large language model (LLM) serving, but its access costs and limited write endurance complicate its use. We evaluate HBF for high-throughput agentic serving across system design and scheduling choices to understand when additional capacity improves serving performance and energy efficiency. We introduce an HBM-HBF-host hierarchical storage system and buffered cache-aware scheduling, and use trace-driven simulations to analyze their effects on performance, energy consumption, and HBF write lifetime. Across the evaluated workloads, the fastest HBF-augmented systems reduce completion time by 36.1-87.7% relative to HBM-only systems. Modeled energy savings reach 59.1%, with benefits depending on the workload and weight placement. Buffered cache-aware scheduling extends estimated HBF write lifetime from 1.21 to 14.82 years in the evaluated configuration. These results demonstrate the importance of coordinating data placement and scheduling to improve serving efficiency while sustaining a practical HBF write lifetime.

Zack Yu, Chloe Wong, Coleman Hooper +7

University of California, Berkeley Berkeley, California, USA · FuriosaAI Seoul, South Korea · University of California, Berkeley; ICSI; LBNL Berkeley, California, USA

LLM Inference Efficiency LLM Serving LLM Inference Scheduling KV-Cache Management

Sep 30, 2026cs.LG

Adaptive Self-Consistency: From Black-Box Sampling to Distribution-Valued Feedback

Self-consistency samples many reasoning trajectories and aggregates their final answers, treating the LLM as a black box that returns one answer per trajectory. Yet the final answer of each trajectory is sampled from a softmax vector that is available from the model's log-probabilities. We refer to this as the grey-box setting in which each trajectory reveals this answer distribution rather than a single draw from it. We formulate efficient inference in this setting as sequential mode identification with distribution-valued observations: sample trajectories one at a time and stop as soon as the LLM's modal answer is identified at a prescribed confidence level. We characterize the asymptotic stopping rate of mode identification with distribution-valued observations exactly and show that it is never worse than the black-box rate. We then propose the ASC-D algorithm, a betting stopping rule that attains this asymptotic stopping rate. On MMLU-Redux, ASC-D uses

46.4

--

95.6\%

fewer trajectories than answer-only adaptive self-consistency baselines and achieves the highest fixed-budget correct-certification rate across three open-source models.

Jingkai Huang, Yunfan Zhang, Will Ma +2

Stern School of Business New York University · School of Management Zhejiang University

LLM Inference Efficiency Adaptive Inference Self-Consistency in Language Models

Sep 29, 2026cs.DC

Purlin: Separating Orchestration from the Datapath of Collectives

Distributed inference depends on GPU collective communication that must keep pace with evolving hardware and specialized workloads. However, existing collective implementations often couple semantics, orchestration (where and when data moves), and the datapath (how data moves). This coupling makes it costly to adopt new hardware mechanisms and customize communication for applications. We present Purlin, a scale-up communication framework that separates these concerns. At the top of Purlin, we specify collectives as a naming of an input and output layout and a copy or reduction operation. In the middle, we introduce a shared orchestration protocol, Stage, Notify, And Consume (SNAC), which derives coordination from these specifications. Below SNAC sits a hardware-specific datapath we call Atom, which implements two key data movement primitives for collectives: copy and reduce. This separation lets us customize collectives and adopt new hardware mechanisms while reusing orchestration via SNAC. We evaluate Purlin on A100, H200, and B200 GPUs. Across seven collectives, Purlin achieves latency speedups of up to 5.14x and bandwidth improvements of up to 4.50x over baselines. Integrated into SGLang, Purlin improves offline LLM serving throughput and interactivity by 1.13x on average and up to 1.37x over baselines. For online LLM inference, Purlin improves interactivity by 1.26x on average and up to 2.85x, with the largest gain occurring under overload. For diffusion image generation, Purlin reduces end-to-end latency by up to 1.13x.

Osayamen Jonathan Aimuyo, Swapnil Gandhi, Christos Kozyrakis

Stanford University · NVIDIA & Stanford University

LLM Inference Efficiency High-Performance Computing Distributed Inference

Sep 28, 2026cs.CY

Beyond Energy: When Sustainability Dimensions Reshape LLM Serving Decisions

Large language model (LLM) serving has environmental impacts across energy consumption, carbon emission, water consumption, and biodiversity loss. Yet these dimensions are largely evaluated in isolation, leaving it unclear when and how they lead to different optimization decisions. We present PRISM, a unified framework for characterizing and optimizing LLM serving across energy, carbon, water, and biodiversity impacts. Our analysis reveals a fundamental distinction: computing configurations determine energy consumption, whereas where and when LLM serving is deployed determine its carbon, water, and biodiversity impacts. Under a fixed deployment choice and operational-only accounting, all dimensions preserve the same energy-based configuration ranking. Deployment rankings can diverge across dimensions, while embodied impacts can break configuration invariance when they exceed a lifecycle crossover boundary. PRISM identifies these conditions, quantifies cross-dimensional regrets, and balances the four dimensions. In regional-routing experiments, PRISM reduces median worst-case regret by 50.2% relative to the strongest baseline.

Tianyao Shi, Xipeng Shen, Yi Ding

Elmore Family School of Electrical and Computer Engineering, Purdue University, USA

LLM Inference Efficiency LLM Serving Energy-Efficient ML

Sep 28, 2026cs.DC

WavePP: High-Throughput Pipeline Parallel LLM Prefill under Prefix Reuse

Pipeline parallelism can improve prefill throughput by processing multiple request chunks concurrently across different stages of the model. However, keeping the pipeline fully utilized requires efficient scheduling and request preparation. In systems where stages retain and evict cache state independently, a local cache hit does not guarantee that the same prefix can be reused across the pipeline. Here, coordination overhead can impede request admission cadence and thus reduce overall throughput. In this paper, we present WavePP, a prefill runtime built on top of TensorRT-LLM that addresses these challenges by overlapping request admission with pipeline execution. WavePP asynchronously finds a prefix that can be reused across all stages, protects the cached state, and reserves space for the remaining input while earlier requests continue to execute. It subsequently plans the chunk sizes of each request dynamically to maximize pipeline fill. Each stage then completes the local preparation before executing the request. In the same system and pipeline topology, WavePP improves TensorRT-LLM's prefill throughput in 37 of 40 tested settings on GLM 5.2 and MiniMax M2.7. At concurrency 128 with high cache reuse, these changes increase throughput by factors of 2.91 and 2.02, respectively. Across 28 Kimi K3 settings, WavePP also has the highest measured throughput in all 18 settings at concurrency eight or higher, compared with tensor/expert-parallel and pipeline-parallel baselines from TRT-LLM, SGLang, and vLLM.

Aaryam Sharma

Baseten

LLM Inference Efficiency LLM Serving LLM Inference Scheduling Prefix Caching Pipeline Parallelism

Sep 28, 2026cs.CL

SCBO: Semantically Coherent Batching and Ordering for LLM-Based Social Surveys

Large Language Models (LLMs) offer a scalable way to simulate survey respondents using demographic profiles and observed reference responses. However, the conventional approach of predicting one question per prompt repeatedly encodes the same context, limits each target to a narrow set of reference responses, and prevents later predictions from using information in earlier answers. Predicting multiple questions in one prompt can reduce these costs, share a broader pool of references, and let later predictions build on earlier ones. This requires forming coherent batches, selecting shared references, and ordering questions and references effectively. We propose Semantically Coherent Batching and Ordering (SCBO), a training-free framework that addresses these challenges. SCBO first uses an LLM to extract compact semantic representations from survey items and filter out template noise. It then groups related questions into batches and builds a shared reference bank using target-specific retrieval and centroid-based completion. Finally, it orders target questions from easy to hard and arranges references according to their semantic alignment with those questions. Experiments on four large-scale survey datasets and four LLMs show that SCBO substantially reduces token consumption and inference time while generally improving prediction accuracy over a non-batched baseline. Code is available at https://anonymous.4open.science/r/SCBO-41D8.

Yuanzi Li, Lingjie Wang, Zihang Tian +2

Renmin University of China

LLM Inference Efficiency LLM Prompting Large Language Model-Based Social Simulation

Sep 28, 2026cs.AI

Beneath the Tokens: A Performance Engineering Study of Multi-Token Prediction in GPU-Accelerated LLM Inference

Autoregressive large language model inference repeatedly invokes the target model to generate one token at a time, making generation sensitive to GPU memory movement and sequential execution. This study evaluates two-token multi-token prediction (MTP) against autoregressive decoding in a controlled single-request deployment on an NVIDIA A10G GPU. A 360-request benchmark covered plain-text, reasoning-intensive, and tool-calling workloads, while runtime telemetry, Nsight Systems, PyTorch Profiler, and selected Nsight Compute measurements were used to explain the observed performance. MTP increased output throughput by

1.91\times

to

2.19\times

across all prompts and reduced time to first output by 10.0--14.2%. Median mean acceptance length ranged from 2.370 to 2.595 tokens per verification iteration. Profiling showed that MTP introduced a longer and more complex execution path, including proposal, sampling, attention, gathering, and reduction operations. However, it required 56.4--78.1% fewer executions of the selected repeating CUDA Graph per generated token. The dominant MTP GEMM kernel was not faster than the dominant autoregressive GEMV kernel, and selected instances of both approached the A10G memory-bandwidth limit. These results show that MTP improved inference through amortization: greater token progress reduced repeated GPU execution sufficiently to outweigh the additional speculative-execution cost.

Suwesh Prasad Sah

LLM Inference Efficiency Multi-Token Prediction GPU Acceleration Speculative Decoding LLM Inference Acceleration

Sep 28, 2026cs.LG

AgentPerfBench: A Benchmarking and Evaluation Suite for Inference Performance of Agentic LLMs

The optimization of LLM serving engines, such as vLLM and SGLang, is largely benchmark-driven: optimizations, scheduling policies, hardware and system designs are all selected based on representative workloads. However, a significant mismatch has emerged in the agentic era. Existing benchmarks primarily focus on simple single-turn chatbot workloads. LLM applications are increasingly agentic: coding agents, terminal execution systems, and tool-use agents issue multi-turn requests with growing context lengths. We introduce AgentPerfBench, a benchmark suite for agentic inference. It uses real traces from agentic benchmarks, such as SWE-Bench and TerminalBench, alongside standard chat baselines. This enables benchmarking of models on multi-turn tasks involving tool calling, skill utilization, and increasing context lengths. AgentPerfBench also samples from empirical distributions of input length, output length, and turn count derived from the real traces, generating representative synthetic profiles for cheap and accurate measurements on new hardware. In addition, we further find that several existing benchmarks fail to accurately reflect real hardware performance for two key reasons: 1) they do not account for realistic context-length growth, and 2) they measure inference performance without operating at hardware saturation. We discuss these issues in detail and provide rich kernel-level Nsight Compute (NCU) traces to construct a new multi-dimensional roofline model that captures hardware-system limitations in both memory bandwidth and memory capacity footprint. The benchmarking suite then includes automated scripts to identify potential bottleneck conditions on emerging hardware when evaluated with diverse agentic traces. Together, these contributions quantify the chat-to-agentic gap in current inference benchmarks and characterise per-kernel GPU resource utilisation via roofline analysis.

Cheuk Hang Lau, Zeyu Cao, Kevin Wong Cheuk Yin +6

Imperial College London · University of Cambridge · University of Oxford

LLM Inference Efficiency LLM Inference AI Agent Benchmarks

Sep 28, 2026cs.LG

Spexis: Speculative Lookahead Scheduling for LLM Inference

Spexis is a multi-GPU LLM inference framework that improves the efficiency of pipeline and tensor parallelism through speculative parallelism. Rather than using speculative decoding only to accelerate token generation, Spexis runs speculation in parallel with normal execution, introducing a new parallelism axis without increasing KV-cache memory usage. This improves memory efficiency and helps mitigate the bottlenecks of multi-GPU inference. Spexis further uses lookahead scheduling to predict speculation quality and future memory pressure, allowing it to reduce wasted speculation, KV-cache eviction, and recomputation. Built on top of vLLM, Spexis largely improves serving performance across a range of GPU configurations, achieving speedups of up to 34% over a baseline that uses the optimal combination of pipeline and tensor parallelism. Spexis's source code is publicly available at https://github.com/mlsys-seo/spexis.

Hyungyu Jung, Jaehyeok Yu, Hoonseo Choi +3

Seoul National University · Hanyang University

LLM Inference Efficiency LLM Inference Scheduling LLM Inference Serving Speculative Decoding Pipeline Parallelism

Sep 27, 2026cs.CY

Greenpixie's AI Token Methodology: Assessing the Energy, Water and CO2-eq Impact of AI Tokens for Open and Closed Weight Models

We describe a methodology for estimating the per-token energy cost of cloud-hosted large language model (LLM) inference, separating between input (prefill) and output (decode) tokens. Graphics processing unit (GPU) energy usage is measured during inference benchmarking with open-weights models on a wide range of text-based tasks. The remaining server energy contribution from non-GPU hardware is estimated from the inference wall time. Bayesian linear regression is used to model the relationship between energy per token and LLM size, request traffic, and hardware deployment configuration. Proprietary frontier LLMs of unknown size and deployment are binned into size buckets based on naming conventions and performance priors, and the space of possible LLM configurations is sampled with Monte-Carlo methods to give a representative average energy per token and uncertainty. We also describe how these energy measurements can be used to estimate the carbon-dioxide equivalent (

\mathrm{CO_2\text{-}eq}

) emissions, both usage and embodied, and water consumed per token of AI inference. This methodology provides actionable data that enables reductions in cost, electricity usage,

\mathrm{CO_2\text{-}eq}

emitted and water consumed in cloud and Software as a Service (SaaS).

Joshua Horswill, Ross Hunter, Matt Clifford +1

Greenpixie Ltd., London, UK.

LLM Inference Efficiency LLM Inference Responsible AI

Sep 27, 2026cs.LG

You Only Edit Once: Incentivizing In-Context Capability of LLMs via Local Demonstration Refinement

In-context learning (ICL) is crucial for boosting the inference performance of large language models (LLMs). However, the effectiveness of ICL in LLMs is greatly influenced by the choice of demonstration sets. Exhaustive searches over these sets are combinatorial, and existing selectors often rely on relevance or likelihood proxies to implicitly assess ICL quality. Making repeated queries to the target LLM with these strategies can incur substantial costs. This work simplifies selection by framing it as a constrained local search problem and presents local demonstration editing (LDE). Starting with an initially retrieved set of demonstrations, LDE employs a single structured edit to explore its surrounding neighborhood while balancing performance gains with search costs. Technically, LDE is reduced to a policy search problem, for which we train a small LLM, referred to as Jev-LDE. This model as the System-1 modifies the retrieved demonstration set by performing actions such as \texttt{Keep}, \texttt{Delete}, or \texttt{Replace} elements, all within a framework of reinforcement learning with verifiable rewards. At test time, Jev-LDE executes a single edit of the retrieved demonstration set, followed by one inference from the target LLM, avoiding the need for iterative context scoring or subset searches. Across standard classification benchmarks, various target LLMs with Jev-LDE as the plug-and-play module consistently improve ICL performance, and Jev-LDE shows transferability to held-out benchmarks and models without retraining. These findings indicate that the LDE approach offers an efficient and adaptable method for harnessing the ICL capabilities of target LLMs.

Jiarong Wen, Qi Wang, Yun Qu +8

Department of Automation, Tsinghua University

LLM Inference Efficiency In-Context Example Selection In-Context Learning Retrieval-Augmented ICL

Sep 24, 2026cs.CL

JEV vs. LLMs as Rubric Judges: Cheaper, Faster, and Wrong in the Same Places

We ask whether Jev, a typed classifier that returns probabilities over permitted answers without generating text, can replace an LLM rubric judge. We compare it with three flash-tier LLM judges on nine panels drawn from seven benchmarks, giving every judge identical criterion texts. Jev's accuracy differs significantly from an LLM judge's in only 8 of 27 paired comparisons, ahead mostly on binary criteria and behind only on graded ones, and most of the other comparisons are inconclusive. Summed over the nine panels, the LLM judges, called once per criterion, cost 29 to 325 times as much as Jev and took 30 to 220 times as long. On graded criteria all four judges agree more with one another than with the labels and mostly assign lower levels than the raters. One of several observational accounts is that raters followed scale conventions our criterion texts omit. Jev's confidence ranks its own errors on most panels, which should make a cheap classifier the ideal first stage of a cascade that defers its uncertain verdicts to an LLM judge. Correlated errors undo that advantage. The LLM judges repeat nearly all of Jev's most confident errors, so a cascade replayed on the recorded verdicts lowers cost but gains at most 1.5 points over the best single judge with cross-fitted thresholds, and at most 2.0 even with oracle thresholds.

Delip Rao, Chris Callison-Burch

University of Pennsylvania

LLM Inference Efficiency LLM Evaluation LLM-as-a-Judge Selective Prediction Classification Rubric-Based Evaluation

Sep 24, 2026cs.DC

When Fancy Eviction Fails: Rethinking Cache Replacement For LLM Prefix Reuse

Long-running LLM applications repeatedly send growing context, making prefix caching critical for reducing prefill cost. Yet prefix-cache behavior under agentic workloads remains poorly understood. We study production traces from two companies and evaluate 14 eviction algorithms across HBM-constrained and large memory-pool settings. Despite a large gap to Belady, sophisticated policies designed for traditional caches provide little benefit over LRU. The reason is structural: prefix reuse is dominated by the regular pacing of active sessions, making recency unusually predictive. Prefix caching nevertheless introduces new challenges, including heavy-tailed session footprints and highly variable miss costs as attention computation grows with sequence length. We introduce the compute-savings ratio and two offline oracles to quantify these effects. Our results show that effective prefix-cache management should retain recency as its foundation while selectively adding quick demotion for one-hit prefixes, compute-aware partial eviction for expensive misses, and capacity-dependent eviction granularity. We will release the traces and simulator to support future research.

Yiyu Liu, Minlan Yu, Juncheng Yang

Harvard University

LLM Inference Efficiency KV Caching KV-Cache Eviction Prefix Caching Agent Memory Management

Sep 22, 2026cs.LG

Scaling of Capability and Efficiency at Inference Time in Large Reasoning Models

Capability and efficiency are two key dimensions of reasoning in large language models (LLMs). Capability refers to the ability to solve a given problem correctly, whereas efficiency refers to the ability to do so with limited resources. When LLMs use Chain-of-Thought (CoT) reasoning to solve problems of controlled hardness, both the number of problems solved correctly and the number of tokens required to reach a correct answer depend on problem hardness and model size. However, how these factors jointly shape capability and efficiency remains poorly understood. Here, we use hierarchical Bayesian models to evaluate the capability and efficiency of LLMs from the DeepSeek-R1-Distill model family across four classes of arithmetic and algorithmic reasoning problems. At a fixed model size, the probability of correctly solving an instance decays approximately exponentially with instance size, our proxy for problem hardness. The decay scale grows sublinearly with model size, indicating that larger models are more capable, but that capability gains diminish with scale. Output length grows as a power law with instance size, which serves as a proxy for difficulty. However, the parameters of this power law do not vary systematically with model size, suggesting that larger models do not become more efficient. Together, these findings reveal potential limitations of naive scaling as a strategy for developing more capable AI systems: capability improves with diminishing returns, while efficiency shows little to no improvement.

Moritz Laber, Zohair Shafi, Germans Savcisens +6

Northeastern University, Boston, Massachusetts, USA · Complexity Science Hub Vienna, Vienna, Austria · Santa Fe Institute, Santa Fe, New Mexico, USA

LLM Inference Efficiency Language Model Scaling Laws Inference-Time Scaling Algorithmic Reasoning Large Reasoning Models

Sep 22, 2026cs.CL

COMED: The Missing Middle Between Routing and Collaboration in Multi-LLM Inference

No single Large Language Model (LLM) is uniformly reliable across queries, motivating multi-model inference systems that either route among models or combine their outputs. However, routing stops after selecting an initial model, while dense collaboration invokes peers on every query. We show that collaboration is non-monotonic: peers can recover failures that no model solves alone, but can also corrupt initially correct answers. We introduce COMED (Controlled Model Escalation for Multi-LLM Deliberation), a post-anchor controller for selective cross-model collaboration. COMED uses anchor self-consistency, router margin, and a lightweight peer probe to accept confident answers, verify ambiguous cases, and escalate only when collaboration is likely beneficial. We formalize this trade-off with a rescue-harm decomposition showing that selective collaboration improves when rescued errors outweigh collaboration-induced harms. Across medical, scientific, and general reasoning benchmarks, COMED improves fixed and routed anchors in all 16 open-weight settings, with gains up to +10.7 percentage points on MedQA while invoking fewer models and using fewer decoded tokens than dense collaboration. On HLE with frontier models, COMED improves GPT-5.5 from 23.1% to 28.1%, outperforming dense collaboration and achieving the best results.

Norah Alballa, Wenxuan Zhang, Salma Kharrat +4

KAUST · LUMS

LLM Inference Efficiency LLM Routing Multi-LLM Collaboration

Sep 22, 2026cs.AI

REFLEX with Jev for Efficient Selective Control in LLM Agents

LLM agents often use generative models for bounded decisions, raising the question of when these decisions can be handled more efficiently without reducing task success. We study REFLEX, an agent architecture that uses Jev as a fast, typed decision layer and calls a strong LLM when confidence is low, or generation is required. On a frozen 100-task benchmark, REFLEX achieves 95% success with 72.7% fewer strong-model calls than a strong-only agent, with reductions persisting across three fallback families. Controlled interventions show that reliability depends on action-set size and near-valid alternatives near authorization boundaries. External BFCL and

τ

-style evaluations reveal limited advantages over a cheap generative cascade when ordinary routing is already highly accurate. These findings identify when selective control with Jev can reduce computation and where its benefits are limited.

Tiantong Wu, Wei Yang Bryan Lim

Nanyang Technological University Singapore

LLM Inference Efficiency LLM Agent Evaluation LLM Decision-Making LLM Agents LLM Agent Reliability

Sep 22, 2026cs.CL

Compressing Long Context into Answer-Aligned Memory Embeddings for LLM Inference

Large language model (LLM) inference is constrained by the quadratic scaling of self-attention and the linear scaling of the KV cache, increasing latency, energy consumption, and GPU memory demand as context length scales. Existing soft-compression methods either lack query-guided memory selection at inference time, train without answer-targeted supervision, or couple compression tightly to a specific decoder architecture. We propose a Context-to-Answer-Aligned Memory Compression (CMC) framework, which compresses long input contexts into compact Context Memory Embeddings (CMEs) aligned to any frozen decoder's embedding space, reducing inference costs without modifying decoder weights. CMC introduces a two-tier KV cache that combines question-guided CME selection with a local context window, and trains the compressor with answer-targeted distillation from a frozen LLM. Experiments across nine encoder-decoder combinations and four QA benchmarks show that CMC consistently outperforms the baseline, achieving up to 7.3 EM and 4.0 F1 point gains on SQuAD, while reducing inference time and energy consumption by up to 20% and peak reserved GPU memory by up to 50% at 3,000 generation tokens. Ablation studies confirm that each architectural component and training objective contributes to the performance.

Md Mostafizer Rahman, Md Faizul Ibne Amin, Md Shahajada Mia +2

Lucy Family Institute for Data & Society, University of Notre Dame, IN, USA · The University of Aizu, Aizuwakamatsu, Japan · Department of Applied and Computational Mathematics and Statistics, University of Notre Dame, IN, USA

LLM Inference Efficiency Long-Context Language Model Inference Memory-Efficient Inference KV-Cache Compression Context Compression

Sep 21, 2026cs.AI

Rollout Efficiency in Reinforcement Learning for Reasoning Large Language Models: A Taxonomy and Future Directions

Reasoning-oriented reinforcement learning enables large language models to solve mathematical, coding, and other multi-step tasks, but shifts a substantial portion of the training cost to rollout, where trajectories are generated for policy updates. Efficient rollout mechanisms are therefore essential to reduce this cost while maintaining the freshness, consistency, and statistical validity of training data. This survey provides a systematic taxonomy of recent research on rollout efficiency for reasoning-oriented reinforcement learning, classifying existing approaches from both mechanism and bottleneck perspectives. Based on this taxonomy, we analyze how different technique families address distinct sources of rollout inefficiency, examine opportunities and potential conflicts for combining them, identify gaps in the evaluation and reporting of efficiency gains, and discuss open challenges and future research directions.

Niloofar Gholipour, Marcos Assuncao, Gursimran Singh +8

École de technologie supérieure, Univ. of Québec, Canada · Huawei Technologies, Canada · The Univ. of Melbourne, Australia +1

LLM Inference Efficiency Reinforcement Learning RL for Language Model Reasoning Efficient Language Model Reasoning

Sep 21, 2026cs.CL

Open-Jev Judgments on CallScreenBench: Calibrated One-Pass Scam Screening with a Small Language Model

Screening a phone call for fraud needs a trustworthy probability after every caller turn, in milliseconds. Jev-style typed decisions promise exactly that: declared options go in, one calibrated probability per option comes out of a single forward pass, with no generated text. We test an open implementation of this readout, JevLite, on scam-call screening: Qwen3-4B is LoRA-tuned so that the temperature-scaled softmax over two answer-label logits is P(scam). On 41 held-out CallScreenBench scenarios (577 per-turn decisions) a three-seed ensemble reaches AUROC .974 with calibration error .052, non-inferior to an LLM judge (MiniMax-M3) at a pre-registered .02 margin, with no false alarms on legitimate calls, decisions 1.14 turns earlier under the same hang-up rule, and 64.5 ms per decision on one consumer GPU, 4.9x lower than the same backbone fine-tuned to generate its answer. The gain is in the readout and calibration, not accuracy: a fine-tuned ModernBERT encoder is not significantly worse, the recipe was selected with test-set exposure, and all callers are synthetic. We claim no architectural novelty; the contribution is the application and an evaluation reporting calibration, false alarms and decision timing alongside AUROC.

Simiao Ren, Kidus Zewde, Xingyu Shen +6

Scam.ai (Reality Inc.)

LLM Inference Efficiency LLM Evaluation Language Model Calibration

Sep 20, 2026cs.LG

ValueDiff: Value-Geometric KV Cache Eviction for Sink-Suppressed LLMs

Modern LLMs with QK-normalization, gated attention, learned attention sinks, or logit softcapping exhibit weaker persistent attention sinks, on which existing KV cache eviction methods primarily rely. We observe that across these models, weaker sinks co-occur with greater value-vector dispersion relative to key-vector dispersion. Motivated by this value-side dispersion, we present ValueDiff, a value-geometric eviction that ranks tokens by the L2 deviation of their value vectors from the cache mean. The same score arises as the minimal-disturbance eviction under a max-entropy assumption about future attention. We evaluate under fixed cache budgets, with eviction at every block boundary during prefill and at every decoding step during generation. On RULER at a tight 2k token budget, ValueDiff retains 88-99% of dense across seven sink-suppressed models (best on 6 out of 7). On LongBench at the 4k budget, ValueDiff averages 92% retention across sink-suppressed models versus 83% for the strongest prior baseline. On MATH-500, ValueDiff is the strongest non-dense method on every sink-suppressed model tested at the 25% cache budget, outperforming prior methods by up to ~20 points on gated-attention models. Across all three benchmarks, value geometry emerges as the more reliable query-invariant eviction signal for sink-suppressed models.

Junyoung Park, Jungwook Choi, Mingu Lee

Qualcomm AI Research · Hanyang University

LLM Inference Efficiency KV-Cache Eviction Attention Sinks

Sep 17, 2026cs.AI

Not All AI Agents Are Equal: Characterizing Resource and Performance Dynamics

LLM-based AI agents process user requests through iterative reasoning and tool execution, often involving the invocation of remote LLM APIs with local tool containers. This execution model can make the optimization of agent serving difficult because latency, local resource demand, and container bottlenecks inter-mix across requests. However, the current agent ecosystem runs without much consideration of resource dynamics, which results in significant waste of the precious resources. This paper analyzes the resource inter-mix of AI agents for three representative tasks: retrieval-augmented question answering, web search, and software coding. To this end, we characterize the latency with respect to the resource dynamics of processing multiple requests and tasks concurrently. Our measurements show that agents have a wide range of behaviors depending on tasks, so that even the same tool can differ substantially in resource dynamics. We also find that running multiple requests concurrently exposes task-dependent bottlenecks in resource dynamics such as CPU, disk I/O, and memory. Furthermore, we uncover that faster LLM responses or more CPU cores do not always accelerate agents. Based on these observations, we demonstrate new optimization opportunities that exploit the resource dynamics of tasks: CPU-aware tool admission and task-aware CPU allocation. Our results show that the latency of CPU-sensitive agent tasks improves

\sim

5.4

\times

, and the average latency across multiple tasks is reduced

\sim

32% compared to native agents.

Wonmi Choi, Minuk Park, Zhixiong Niu +3

Department of Computer Science and Engineering, Korea University · Microsoft Research Asia

LLM Inference Efficiency LLM Agent Workflow Optimization LLM Agent Serving