AI Paper Atlas

Research map / Time series / Foundation models for time series

HyperLoad: A Cross-Modality Enhanced Large Language Model-Based Framework for Green Data Center Cooling Load Prediction

Haoyu Jiang, Boan Qu, Junjie Zhu, Fanjie Zeng, Xiaojie Lin, Wei Zhong

AAAI 2026 · Time series › Foundation models for time series · approach: LLM-based time series forecasting · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ADPT (text prior prefix tuning) (ablation)HyperLoad · DCData (100% train, PL=96)MSE: 0.0220 -> 0.0192Improved
add EGIA cross-variable attention (ablation)HyperLoad · DCData (100% train, PL=96)MSE: 0.0211 -> 0.0192Improved
add KARI cross-modal alignment loss (ablation)HyperLoad · DCData (100% train, PL=96)MSE: 0.0194 -> 0.0192Improved
use LLaMA backbone instead of BERT (ablation)HyperLoad · DCData (100% train, PL=96)MSE: 0.0217 -> 0.0192Improved
use LLaMA backbone instead of GPT-2 (ablation)HyperLoad · DCData (100% train, PL=96)MSE: 0.0224 -> 0.0192Improved
use LLaMA instead of GPT-2 costs more training time (ablation)HyperLoad · DCDataTime per iteration (s): 0.1943 -> 0.3928Hurt
HyperLoad vs best baseline with 50% training dataHyperLoad · DCData (50% train)MSE: MSE reduced by 6.5%, 3.3%, 4.5%, 3.9%Improved
HyperLoad vs best baseline with 25% training dataHyperLoad · DCData (25% train)MSE: MSE reductions 16.7%, 13.2%, 1.7%, 2.6%Improved
reduce training data from 100% to 25% (ablation)HyperLoad · DCData (PL=96)MSE: 0.0192 -> 0.0442Hurt
HyperLoad vs iTransformer at 100% data, PL=96HyperLoad · DCData (100% train, PL=96)MAE: 0.0977 -> 0.0931Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy