用图像代替文本处理时间序列，让大模型更省电还更准。
A Picture is Worth a Thousand Tokens: How Vision Language Models Cut AI Energy Costs While Improving Accuracy
- 把时间序列转成2D图表输入大模型，减少3.6到10.4倍的输入tokens
- 实测推理能耗降低1.8到2.5倍，每天省7.2兆焦耳能量
- 适合电信、工业等需低延迟高精度的数值时序分析场景
大模型推理占人工智能运营能耗超90%，且与输入token数量直接相关。在4G/5G基站多变量KPI分析中，原始数据会膨胀为数千个浮点数token。视觉语言模型（VLMs）通过将时间序列编码为二维图表，使Llama-3.2-90B、Qwen2.5-VL-72B和Pixtral-12B等架构的输入token减少3.6至10.4倍，实测推理能耗下降1.8至2.5倍，单次部署（监控200个小区/15分钟）每日节省约7.2兆焦耳。关键的是，准确率未下降：微调后的Llama-3.2-90B-Vision在电信异常检测中精度比纯文本模型高220.7%，优于LSTM与ARIMA超过144%；在公开基准上，Pixtral-12B的J/F1得分提升20.6倍，均值F1达0.82。当含24个KPI时，文本表示超出多数生产级LLM的128K上下文窗口，必须截断，而视觉表示仍符合标准限制。结果表明，VLM是数值时序任务中兼具能效与精度优势的新范式。
原文摘要 · Abstract (English)
LLM inference accounts for over 90% of AI operational energy, scaling directly with input token count---a critical inefficiency for telecom network analytics and numerical time-series data analysis (NTSDA), where raw multivariate KPI windows from 4G/5G cell sites expand into thousands of floating-point tokens. Vision-Language Models (VLMs) eliminate this mismatch by encoding time-series as 2D plots, achieving 3.6-10.4x input token reduction across Llama-3.2-90B, Qwen2.5-VL-72B, and Pixtral-12B architectures. This translates to 1.8-2.5x measured inference energy reduction, saving approximately 7.2 MJ/day at telecom edge deployments and CloudRAN that monitor 200 cells per 15-minute interval. Critically, efficiency gains do not sacrifice accuracy: a fine-tuned Llama-3.2-90B-Vision VLM achieves 220.7% higher precision than its text-only counterpart and outperforms LSTM and ARIMA baselines by over 144% on telecom anomaly detection. On public benchmarks, Pixtral-12B achieves a 20.6x improvement in J/F1 score at mean F1 = 0.82. At 24 KPIs, text representations exceed the 128K context window of most production LLMs, rendering text-only processing infeasible without truncation, while visual representations remain within standard limits. These results establish VLMs as an energy-efficient and accuracy-superior modality for numerical time-series workloads, providing empirical grounding for AI inference systems that treat energy consumption as a first-class engineering constraint.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。