SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2001.08950 · ICML · 2020

PoWER-BERT: Accelerating BERT Inference via Progressive Word-vector Elimination

Saurabh Goyal, Anamitra Choudhury, Saurabh Raje, Venkatesan Chakaravarthy, Yogish Sabharwal, Ashish Verma

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-21; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2001.08950.json · badge: /api/ran/2001.08950.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
IBM/PoWER-BERT canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
Constraint Ran
ran on a synthesized input
IBM/PoWER-BERT/model/model.py
code served (permissive licence) · get_code("02abc68e03216355")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from scipy import linalg

INIT_ARGS = {}

ARGS = {
    "w": torch.randn(2, 4)
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
numpy, scipy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
CustomConstraint Ran
ran on a synthesized input · constructed an object; no result was computed
IBM/PoWER-BERT/model/model.py
code served (permissive licence) · get_code("3c320f6e8b748218")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, CustomConstraint: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

class Constraint(object):

    def __call__(self, w):
        return w

    def get_config(self):
        return {}

class CustomConstraint(Constraint):

    def __init__(self, min, max):
        self.min = min
        self.max = max

    def __call__(self, w):
        new_w = tf.clip_by_value(w, clip_value_min=self.min, clip_value_max=self.max)
        return new_w

    def get_config(self):
        return {'C': 0.0}


# Setup valid arguments for CustomConstraint
ARGS = {
    'min': -1.0,
    'max': 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type CustomConstraint — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a CustomConstraint output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
Soft_Extract Ran
ran on a synthesized input · constructed an object; no result was computed
IBM/PoWER-BERT/model/model.py
code served (permissive licence) · get_code("4749eae87ebd6765")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, Soft_Extract: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import keras
import keras.backend as K
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create a small attention tensor that the layer will use
# Shape: (batch_size * head_num, seq_len, feature_dim) = (24, 5, 4)
# This will be reshaped to (2, 12, 5, 4) inside calc_avg_atten
atten_tensor = tf.random.normal(shape=(24, 5, 4), dtype=tf.float32)

# Create input tensor for the layer
# Shape: (batch_size, seq_len, feature_dim) = (2, 5, 4)
input_tensor = tf.random.normal(shape=(2, 5, 4), dtype=tf.float32)

ARGS = {
    'atten': atten_tensor,
    'kernel_initializer': 'glorot_normal',
    'kernel_Regularizer': None,
    'kernel_constraint': None,
    'LAMBDA': 0.01,
    'name': 'soft_extract_layer'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
keras, numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Soft_Extract — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Soft_Extract output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We develop a novel method, called PoWER-BERT, for improving the inference time of the popular BERT model, while maintaining the accuracy. It works by: a) exploiting redundancy pertaining to word-vectors (intermediate encoder outputs) and eliminating the redundant vectors. b) determining which word-vectors to eliminate by developing a strategy for measuring their significance, based on the self-attention mechanism. c) learning how many word-vectors to eliminate by augmenting the BERT model and the loss function. Experiments on the standard GLUE benchmark shows that PoWER-BERT achieves up to 4.5x reduction in inference time over BERT with < 1% loss in accuracy. We show that PoWER-BERT offers significantly better trade-off between accuracy and inference time compared to prior methods. We demonstrate that our method attains up to 6.8x reduction in inference time with < 1% loss in accuracy when applied over ALBERT, a highly compressed version of BERT. The code for PoWER-BERT is publicly available at https://github.com/IBM/PoWER-BERT.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2001.08950")
get_code_for_paper("2001.08950")
have("2001.08950")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2001.08950.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2001.08950.svg)](https://syntology.ai/paper/2001.08950)

Connect an agent — have() is free.