Papers › HAT: Hardware-Aware Transformers for Efficient Natural Language Processing

HAT: Hardware-Aware Transformers for Efficient Natural Language Processing

28 May 2020ACL 2020 6arXiv:2005.14187archive 2025-07-28

Hanrui Wang, Zhanghao Wu, Zhijian Liu, Han Cai, Ligeng Zhu, Chuang Gan, Song Han

Transformers are ubiquitous in Natural Language Processing (NLP) tasks, but they are difficult to be deployed on hardware due to the intensive computation. To enable low-latency inference on resource-constrained hardware platforms, we propose to design Hardware-Aware Transformers (HAT) with neural architecture search. We first construct a large design space with arbitrary encoder-decoder attention and heterogeneous layers. Then we train a SuperTransformer that covers all candidates in the design space, and efficiently produces many SubTransformers with weight sharing. Finally, we perform an evolutionary search with a hardware latency constraint to find a specialized SubTransformer dedicated to run fast on the target hardware. Extensive experiments on four machine translation tasks demonstrate that HAT can discover efficient models for different hardware (CPU, GPU, IoT device). When running WMT'14 translation task on Raspberry Pi-4, HAT can achieve 3× speedup, 3.7× smaller size over baseline Transformer; 2.7× speedup, 3.6× smaller size over Evolved Transformer with 12,041× less search cost and no performance loss. HAT code is https://github.com/mit-han-lab/hardware-aware-transformers.git

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

mit-han-lab/hardware-aware-transformers officialmentioned in papermentioned on GitHubpytorchNOASSERTION report
Luccadoremi/Model-Compression-DAQ mentioned on GitHubpytorchNOASSERTION report
aaditkapoor/PDFExtract mentioned on GitHubApache-2.0 report
mlatsjsu/PDFextract mentioned on GitHubMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

DecoderMachine TranslationNeural Architecture SearchRaspberry Pi 4Translation

2 archive task tags without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Machine Translation WMT2014 English-French Hardware Aware Transformer BLEU score 41.8 #21 of 57 Archive leaderboard report
Machine Translation WMT2014 English-German Hardware Aware Transformer BLEU score 28.4 #43 of 91 Archive leaderboard report
Machine Translation WMT2014 English-German Hardware Aware Transformer Number of Params 48M #43 of 91 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections