Methods › Natural Language Processing › Autoencoding Transformers › I-BERT
I-BERT
Introduced by Sehoon Kim et al. in I-BERT: Integer-only BERT Quantization
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
I-BERT is a quantized version of BERT that quantizes the entire inference with integer-only arithmetic. Based on lightweight integer only approximation methods for nonlinear operations, e.g., GELU, Softmax, and Layer Normalization, it performs an end-to-end integer-only BERT inference without any floating point calculation.
In particular, GELU and Softmax are approximated with lightweight second-order polynomials, which can be evaluated with integer-only arithmetic. For LayerNorm, integer-only computation is performed by leveraging a known algorithm for integer calculation of square root.
Papers archive 2025-07-28
3 shown of 3, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.
-
Mixed Non-linear Quantization for Vision Transformers 26 Jul 2024 · 1 repository · arXiv:2407.18437
-
The Feasibility of Implementing Large-Scale Transformers on Multi-FPGA Platforms 24 Apr 2024 · 0 repositories · arXiv:2404.16158
-
I-BERT: Integer-only BERT Quantization 5 Jan 2021 · 7 repositories · arXiv:2101.01321
Tasks archive 2025-07-28
4 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
| Task | Papers |
|---|---|
| Quantization | 2 |
| GPU | 1 |
| Natural Language Inference | 1 |
| Natural Language Understanding | 1 |
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
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