# TherMapNet Attention-Guided Runtime Full-Chip Thermal Map Prediction from Performance Metrics

arXiv: https://arxiv.org/abs/2608.21887 · HTML: https://mlsanity.com/paper/2608.21887/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 19 references that are in the map.
- First posted on arXiv 22 Aug 2026.

## Builds on (most cited first, 6 of 6)

- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Self-Attention with Relative Position Representations](https://mlsanity.com/paper/1803.02155/) (2018)
- [DeepOHeat: Operator Learning-based Ultra-fast Thermal Simulation in 3D-IC Design](https://mlsanity.com/paper/2302.12949/) (2023)
- [Wafer-scale Computing: Advancements, Challenges, and Future Perspectives](https://mlsanity.com/paper/2310.09568/) (2023)
- [A Methodology for Thermal Simulation of Interconnects Enabled by Model Reduction with Material Property Variation](https://mlsanity.com/paper/2112.03023/) (2021)
- [PODTherm-GP: A Physics-based Data-Driven Approach for Effective Architecture-Level Thermal Simulation of Multi-Core CPUs](https://mlsanity.com/paper/2305.01911/) (2023)

## Abstract (arXiv)

Runtime thermal management of high-performance chips depends on fast and accurate full-chip thermal maps. Conventional simulators typically estimate power traces from performance metrics first, which adds overhead. This work proposes TherMapNet, an attention-guided thermal simulator that predicts full-chip thermal maps directly from performance metrics. A Transformer encoder captures temporal evolution by treating the time series of each metric as a token, improving modeling of dynamic workloads. A CNN then extracts fine-grained spatial features. For the CNN, a dual-branch channel-spatial attention convolution module (DACM) and a triplet loss are used to improve spatial learning and reconstruction accuracy. TherMapNet is applied to a multi-core CPU (AMD Ryzen 7 4800U) and a many-core GPU (NVIDIA GeForce RTX 4060). Experiments show that it outperforms prior thermal simulators, with RMSE below 0.26 C and inference under 2.4 ms on an NVIDIA GeForce RTX 3090 GPU. These results indicate that TherMapNet can support high-quality runtime thermal management of modern multi-core chips.