How Do Prompt Variations Affect Energy Consumption in On-Device LLMs?

• URL: http://arxiv.org/abs/2609.01798v1
• Date: Tue, 01 Sep 2026 19:08:04 GMT
• Title: How Do Prompt Variations Affect Energy Consumption in On-Device LLMs?
• Abstract summary: Large language models (LLMs) are increasingly deployed on mobile devices, making energy efficiency a key deployment constraint.<n>This paper aims to understand how two prompt properties, cognitive load and phrasing pattern, shape the energy behavior of on-device LLM inference.
• Score: 15.386321055792209
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) are increasingly deployed on mobile devices, making energy efficiency a key deployment constraint, yet the energy impact of prompt design remains underexplored. This paper aims to understand how two prompt properties, cognitive load and phrasing pattern, shape the energy behavior of on-device LLM inference. We conduct a broad empirical study covering prompt properties, datasets, models, and devices, with phase-level profiling that separates prefill and decode energy. We find that cognitive load primarily affects the energy cost per token, while phrasing pattern affects energy largely through token usage. Our energy-quality analysis further shows that prompt design reshapes the attainable frontier differently across models, highlighting the need for model-aware prompt design in energy-efficient on-device LLM inference. Code, datasets, and scripts are available at https://amai-gsu.github.io/PromptProperty/.

Related papers

• Understanding the Energy Scaling of Large Language Model Inference Across Context Lengths and Attention Architectures [0.2519906683279152]
  We study how attention architecture influences decode-phase energy consumption under varying inference workloads.<n>We show that attention is the primary factor governing how decode energy scales with context length.<n>Model size determines absolute energy consumption, while reducing both energy per generated token and latency by up to 87%.
  arXiv  Detail & Related papers  (2026-08-25T19:45:30Z)
• From Tokens to Watt-hours: Analytical Energy Estimation for LLM Inference on Modern GPUs [6.041312734387813]
  This report presents an analytically structured, empirically calibrated, GPU-level methodology for estimating large language model (LLM) inference energy on NVIDIA H100-class accelerators without direct measurement.<n>The proposed estimator combines parameter-scaled transformer FLOP accounting, memory-traffic factors, and hardware-specific energy coefficients for FP16/BF16 tensor-core computation and high-bandwidth-memory movement.<n>The methodology further decomposes total energy into compute, parameter-access, key-value-cache write, and attention-read components, allowing the scaling behavior with model size, context length, and generated-token count to
  arXiv  Detail & Related papers  (2026-07-29T07:50:55Z)
• Keyword Matters: Unveiling the Energy Sensitivity of On-Device LLM Prompting [2.1921490318031656]
  Large Language Models (LLMs) are increasingly deployed on mobile and embedded devices to improve privacy and reduce network latency.<n>Yet on-device inference faces a fundamental constraint: high energy consumption on battery-powered, resource-limited hardware.<n>This paper presents an empirical study of the relationship between prompt wording and energy consumption for on-device LLMs.
  arXiv  Detail & Related papers  (2026-05-31T18:29:16Z)
• Beyond Test-Time Compute Strategies: Advocating Energy-per-Token in LLM Inference [1.522424334864671]
  Large Language Models (LLMs) demonstrate exceptional performance across diverse tasks but come with substantial energy and computational costs.<n>Small Language Models (SLMs) can provide accurate responses for simpler text generation tasks.<n>These strategies can also introduce additional energy costs, creating an energy-accuracy trade-off.
  arXiv  Detail & Related papers  (2026-03-04T15:20:55Z)
• Towards Green AI: Decoding the Energy of LLM Inference in Software Development [46.879983975894135]
  AI-assisted tools are increasingly integrated into software development, but their reliance on large language models (LLMs) introduces substantial computational and energy costs.<n>We conduct a phase-level analysis of LLM inference energy consumption, distinguishing between the (1) prefill, where the model processes the input and builds internal representations, and (2) decoding, where output tokens are generated using the stored state.
  arXiv  Detail & Related papers  (2026-02-05T14:38:19Z)
• Understanding Efficiency: Quantization, Batching, and Serving Strategies in LLM Energy Use [4.513690948889834]
  Large Language Models (LLMs) are increasingly deployed in production, contributing towards shifting the burden in terms of computational resources and energy demands from training to inference.<n>We show how emphsystem-level design choices can lead to orders-of-magnitude differences in energy consumption for the same model.<n>Our findings motivate phase-aware energy profiling and system-level optimizations for greener AI services.
  arXiv  Detail & Related papers  (2026-01-29T22:16:25Z)
• Green Prompting [2.377736044164336]
  Large Language Models (LLMs) have become widely used across various domains spanning search engines, code generation, and text creation.<n>This study empirically study how different prompt and response characteristics directly impact LLM inference energy cost.
  arXiv  Detail & Related papers  (2025-03-09T19:49:31Z)
• Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding [58.364933651703524]
  We show that concentrated massive values consistently emerge in specific regions of attention queries.<n>These massive values play a critical role in interpreting contextual knowledge.<n>We trace the emergence of massive values and find that such concentration is caused by Rotary Positional.
  arXiv  Detail & Related papers  (2025-02-03T17:47:03Z)
• Prompt Highlighter: Interactive Control for Multi-Modal LLMs [50.830448437285355]
  This study targets a critical aspect of multi-modal LLMs' (LLMs&VLMs) inference: explicit controllable text generation. We introduce a novel inference method, Prompt Highlighter, which enables users to highlight specific prompt spans to interactively control the focus during generation. We find that, during inference, guiding the models with highlighted tokens through the attention weights leads to more desired outputs.
  arXiv  Detail & Related papers  (2023-12-07T13:53:29Z)
• On Feature Diversity in Energy-based Models [98.78384185493624]
  An energy-based model (EBM) is typically formed of inner-model(s) that learn a combination of the different features to generate an energy mapping for each input configuration. We extend the probably approximately correct (PAC) theory of EBMs and analyze the effect of redundancy reduction on the performance of EBMs.
  arXiv  Detail & Related papers  (2023-06-02T12:30:42Z)
• Energy Transformer [64.22957136952725]
  Our work combines aspects of three promising paradigms in machine learning, namely, attention mechanism, energy-based models, and associative memory. We propose a novel architecture, called the Energy Transformer (or ET for short), that uses a sequence of attention layers that are purposely designed to minimize a specifically engineered energy function.
  arXiv  Detail & Related papers  (2023-02-14T18:51:22Z)
• Pre-Training Transformers as Energy-Based Cloze Models [95.04748595976811]
  We introduce Electric, an energy-based cloze model for representation learning over text. Electric does not use masking or output a full distribution over tokens that could occur in a context. We train Electric using an algorithm based on noise-contrastive estimation and elucidate how this learning objective is closely related to the recently proposed ELECTRA pre-training method.
  arXiv  Detail & Related papers  (2020-12-15T19:17:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.