Papers › Q-BERT: Hessian Based Ultra Low Precision Quantization of BERT

Q-BERT: Hessian Based Ultra Low Precision Quantization of BERT

12 Sep 2019arXiv:1909.05840archive 2025-07-28

Sheng Shen, Zhen Dong, Jiayu Ye, Linjian Ma, Zhewei Yao, Amir Gholami, Michael W. Mahoney, Kurt Keutzer

Transformer based architectures have become de-facto models used for a range of Natural Language Processing tasks. In particular, the BERT based models achieved significant accuracy gain for GLUE tasks, CoNLL-03 and SQuAD. However, BERT based models have a prohibitive memory footprint and latency. As a result, deploying BERT based models in resource constrained environments has become a challenging task. In this work, we perform an extensive analysis of fine-tuned BERT models using second order Hessian information, and we use our results to propose a novel method for quantizing BERT models to ultra low precision. In particular, we propose a new group-wise quantization scheme, and we use a Hessian based mix-precision method to compress the model further. We extensively test our proposed method on BERT downstream tasks of SST-2, MNLI, CoNLL-03, and SQuAD. We can achieve comparable performance to baseline with at most 2.3% performance degradation, even with ultra-low precision quantization down to 2 bits, corresponding up to 13× compression of the model parameters, and up to 4× compression of the embedding table as well as activations. Among all tasks, we observed the highest performance loss for BERT fine-tuned on SQuAD. By probing into the Hessian based analysis as well as visualization, we show that this is related to the fact that current training/fine-tuning strategy of BERT does not converge for SQuAD.

PaperPDF

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

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

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

Tasks

Linguistic AcceptabilityNatural Language InferenceQuantizationSemantic Textual SimilaritySentiment Analysis

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Linguistic Acceptability CoLA Q-BERT (Shen et al., 2020) Accuracy 65.1 #23 of 43 Archive leaderboard report
Natural Language Inference MultiNLI Q-BERT (Shen et al., 2020) Matched 87.8 #20 of 67 Archive leaderboard report
Natural Language Inference QNLI Q-BERT (Shen et al., 2020) Accuracy 93.0 #22 of 43 Archive leaderboard report
Natural Language Inference RTE Q-BERT (Shen et al., 2020) Accuracy 84.7 #28 of 90 Archive leaderboard report
Semantic Textual Similarity MRPC Q-BERT (Shen et al., 2020) Accuracy 88.2 #22 of 45 Archive leaderboard report
Semantic Textual Similarity STS Benchmark Q-BERT (Shen et al., 2020) Pearson Correlation 0.911 #13 of 66 Archive leaderboard report
Sentiment Analysis SST-2 Binary classification Q-BERT (Shen et al., 2020) Accuracy 94.8 #31 of 87 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

AdamAttentionAttention DropoutBERTDense ConnectionsDropoutLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionResidual ConnectionSoftmaxWeight DecayWordPiece

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