SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2005.11787 · 2020

Common Sense or World Knowledge? Investigating Adapter-Based Knowledge Injection into Pretrained Transformers

arXiv · PDF · Open in the Atlas

Code that ran

5 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 6 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; 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/2005.11787.json · badge: /api/ran/2005.11787.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
wluper/retrograph canonical (classifier label)
classifier basis: census
5 ran · 6 no recorded run
FunctionStatusWhere it lives
convert_to_unicode Ran
ran; our contract draft was wrong, not the code
wluper/retrograph/retrograph/modeling/tokenization.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from huawei-noah/Pretrained-Language-Model, is pointer only (licence: NONE) · get_code("1923fc05163d207d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from huawei-noah/Pretrained-Language-Model, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

create_float_feature Ran
ran; our contract draft was wrong, not the code
wluper/retrograph/archive/create_pretraining_data.py
code served (permissive licence) · get_code("b24b3b9c700da839")
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 Feature. 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

def create_float_feature(values):
    feature = tf.train.Feature(float_list=tf.train.FloatList(value=list(values)))
    return feature

# The function takes 'values' parameter and converts it to a list, then creates a FloatList
# We need to provide an iterable of numeric values

values = [1.0, 2.5, 3.7, 4.2]

ARGS = {
    "values": values
}
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 Feature — a Feature output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Feature 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
create_int_feature Ran
ran; our contract draft was wrong, not the code
wluper/retrograph/archive/create_pretraining_data.py
code served (permissive licence) · get_code("dd3ef48a15aec0cd")
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 Feature. 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

# The function create_int_feature takes 'values' as a parameter
# It converts values to a list and passes it to tf.train.Int64List
# So values should be an iterable of integers

ARGS = {
    "values": [1, 2, 3, 4, 5]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Feature — a Feature output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Feature 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
get_activation Ran
ran; our contract draft was wrong, not the code
wluper/retrograph/retrograph/modeling/modeling.py
code served (permissive licence) · get_code("5681f7dd9b6e3679")
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) and returned function. 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 six

# Test cases for get_activation function
# The function takes activation_string as parameter and returns either None or a callable

ARGS = {
    "activation_string": "relu"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
six, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
printable_text Ran
ran; our contract draft was wrong, not the code
wluper/retrograph/retrograph/modeling/tokenization.py
code served (permissive licence) · get_code("0e5615f8994003cf")
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 str. 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 six

# The function takes a single parameter 'text' which can be:
# - str (works in both Python 2 and 3)
# - bytes (in Python 3, decoded to utf-8)
# - unicode (in Python 2, encoded to utf-8)

# Since we're running on Python 3, we'll test with a string
text_input = "Hello, World!"

ARGS = {
    "text": text_input
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
six — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
feedforward_adapter Not yet run
unverified (harvested, no recorded run)
wluper/retrograph/retrograph/modeling/modeling_adapter.py
code served (permissive licence) · get_code("b42f7bbad5547ecf")
How it ran

no invocation recorded.

file_based_input_fn_builder Not yet run
unverified (harvested, no recorded run)
wluper/retrograph/archive/run_classifier_adapter.py
code served (permissive licence) · get_code("f105653fd006d566")
How it ran

no invocation recorded.

file_based_input_fn_builder Not yet run
unverified (harvested, no recorded run)
wluper/retrograph/archive/run_regression_adapter.py
code served (permissive licence) · get_code("9df709f11c75e63a")
How it ran

no invocation recorded.

gelu Not yet run
unverified (harvested, no recorded run)
wluper/retrograph/retrograph/modeling/modeling.py
code served (permissive licence) · get_code("ecab128238ebf253")
How it ran

no invocation recorded.

get_assignment_map_from_checkpoint Not yet run
unverified (harvested, no recorded run)
wluper/retrograph/retrograph/modeling/modeling.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from toggle1995/RIS-DMMI, is pointer only (licence: NONE) · get_code("50958618b65e514e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from toggle1995/RIS-DMMI, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

load_vocab Not yet run
unverified (harvested, no recorded run)
wluper/retrograph/retrograph/modeling/tokenization.py
code served (permissive licence) · get_code("ff83ccc8b0b6462d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 10 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 6 the same month, 8 later) and 1 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.

FunctionAlso present inEarliest-datedOthers named
convert_to_unicode 36 other papers
15 dated earlier, 1 the same month, 20 later
1609.08144 (2016-09)
Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.04085 (2019-01)
Passage Re-ranking with BERT
and 33 more not named here
printable_text 33 other papers
14 dated earlier, 1 the same month, 18 later
1609.08144 (2016-09)
Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.04085 (2019-01)
Passage Re-ranking with BERT
and 30 more not named here
get_activation 32 other papers
19 dated earlier, 1 the same month, 12 later · 1 only as an assembly brick
1804.05392 (2018-04)
Higher-order Coreference Resolution with Coarse-to-fine Inference
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.04085 (2019-01)
Passage Re-ranking with BERT
and 29 more not named here
get_assignment_map_from_checkpoint 26 other papers
17 dated earlier, 1 the same month, 8 later
1804.05392 (2018-04)
Higher-order Coreference Resolution with Coarse-to-fine Inference
1808.09781 (2018-08)
Self-Attentive Sequential Recommendation
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
and 23 more not named here
gelu 16 other papers
10 dated earlier, 1 the same month, 5 later · 2 only as an assembly brick
1706.03762 (2017-06)
Attention Is All You Need
1804.05392 (2018-04)
Higher-order Coreference Resolution with Coarse-to-fine Inference
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
and 13 more not named here
load_vocab 16 other papers
9 dated earlier, 1 the same month, 6 later
1609.08144 (2016-09)
Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.08746 (2019-01)
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
and 13 more not named here
create_int_feature 13 other papers
9 dated earlier, 4 later
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1906.05474 (2019-06)
Transfer Learning in Biomedical Natural Language Processing: An Evaluation of BERT and ELMo on Ten Benchmarking Datasets
1908.08962 (2019-08)
Well-Read Students Learn Better: On the Importance of Pre-training Compact Models
and 10 more not named here
create_float_feature 9 other papers
6 dated earlier, 3 later
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1908.08962 (2019-08)
Well-Read Students Learn Better: On the Importance of Pre-training Compact Models
1909.06356 (2019-09)
Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering
and 6 more not named here
feedforward_adapter 1 other paper
1 dated earlier
1902.00751 (2019-02)
Parameter-Efficient Transfer Learning for NLP
file_based_input_fn_builder 1 other paper
1 dated earlier
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

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 2026-11-22.

Repositories linked to this paper

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

Abstract

Following the major success of neural language models (LMs) such as BERT or GPT-2 on a variety of language understanding tasks, recent work focused on injecting (structured) knowledge from external resources into these models. While on the one hand, joint pretraining (i.e., training from scratch, adding objectives based on external knowledge to the primary LM objective) may be prohibitively computationally expensive, post-hoc fine-tuning on external knowledge, on the other hand, may lead to the catastrophic forgetting of distributional knowledge. In this work, we investigate models for complementing the distributional knowledge of BERT with conceptual knowledge from ConceptNet and its corresponding Open Mind Common Sense (OMCS) corpus, respectively, using adapter training. While overall results on the GLUE benchmark paint an inconclusive picture, a deeper analysis reveals that our adapter-based models substantially outperform BERT (up to 15-20 performance points) on inference tasks that require the type of conceptual knowledge explicitly present in ConceptNet and OMCS. All code and experiments are open sourced under: https://github.com/wluper/retrograph .

For agents

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

get_harvested_code_for_paper("2005.11787")
get_code_for_paper("2005.11787")
have("2005.11787")

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

curl https://syntology.ai/api/ran/2005.11787.json

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

[![Syntology run record](https://syntology.ai/api/ran/2005.11787.svg)](https://syntology.ai/paper/2005.11787)

Connect an agent — have() is free.