SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.04823 · NeurIPS · 2024

BERTs are Generative In-Context Learners

David Samuel

arXiv · PDF · Open in the Atlas

Code that ran

13 of 26 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; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 11 failures of our instrument; 13 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/2406.04823.json · badge: /api/ran/2406.04823.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
ltgoslo/bert-in-context canonical (classifier label)
classifier basis: census
13 ran · 13 no recorded run
FunctionStatusWhere it lives
exact_match_score Ran
ran, violated the contract we drafted
ltgoslo/bert-in-context/glue/record.py
code served (permissive licence) · get_code("f6c275d6a18330a9")
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 bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 string
import re

def normalize_answer(s):
    """Lower text and remove punctuation, articles and extra whitespace."""

    def remove_articles(text):
        return re.sub('\\b(a|an|the)\\b', ' ', text)

    def white_space_fix(text):
        return ' '.join(text.split())

    def remove_punc(text):
        exclude = set(string.punctuation)
        return ''.join((ch for ch in text if ch not in exclude))

    def lower(text):
        return text.lower()
    return white_space_fix(remove_articles(remove_punc(lower(s))))

def exact_match_score(prediction, ground_truth):
    return normalize_answer(prediction) == normalize_answer(ground_truth)

ARGS = {
    "prediction": "The quick brown fox jumps over the lazy dog.",
    "ground_truth": "the quick brown fox jumps over the lazy dog"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, shape () — shape () as a later run recorded it
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: none by the harness; any seed is in the setup block
f1_score Ran
ran, honoured the contract we drafted
ltgoslo/bert-in-context/glue/record.py
code served (permissive licence) · get_code("d9296df040618579")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 string
from collections import Counter
import re

def normalize_answer(s):
    """Lower text and remove punctuation, articles and extra whitespace."""

    def remove_articles(text):
        return re.sub('\\b(a|an|the)\\b', ' ', text)

    def white_space_fix(text):
        return ' '.join(text.split())

    def remove_punc(text):
        exclude = set(string.punctuation)
        return ''.join((ch for ch in text if ch not in exclude))

    def lower(text):
        return text.lower()
    return white_space_fix(remove_articles(remove_punc(lower(s))))

def f1_score(prediction, ground_truth):
    prediction_tokens = normalize_answer(prediction).split()
    ground_truth_tokens = normalize_answer(ground_truth).split()
    common = Counter(prediction_tokens) & Counter(ground_truth_tokens)
    num_same = sum(common.values())
    if num_same == 0:
        return 0
    precision = 1.0 * num_same / len(prediction_tokens)
    recall = 1.0 * num_same / len(ground_truth_tokens)
    f1 = 2 * precision * recall / (precision + recall)
    return f1

ARGS = {
    "prediction": "The quick brown fox jumps over the lazy dog",
    "ground_truth": "A quick brown fox jumps over a lazy dog"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
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: none by the harness; any seed is in the setup block
format_prompt Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/language-modeling/hellaswag.py
code served (permissive licence) · get_code("28110ec97c757855")
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 torch
import numpy as np
from typing import List

# Create a stand-in args object with a separator attribute
class Args:
    def __init__(self):
        self.separator = ' | '

args = Args()

# Create the inputs list with the structure expected by format_prompt
inputs = [
    {
        'activity_label': 'reading',
        'ctx_a': 'The book is on the table',
        'ctx_b': 'and it is very interesting',
        'answer': 'yes'
    },
    {
        'activity_label': 'writing',
        'ctx_a': 'She writes every day',
        'ctx_b': '',
        'answer': 'no'
    },
    {
        'activity_label': 'thinking',
        'ctx_a': 'He thinks about',
        'ctx_b': 'the future',
        'answer': 'maybe'
    }
]

ARGS = {
    'args': args,
    'inputs': inputs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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
format_prompt Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/language-modeling/story_cloze.py
code served (permissive licence) · get_code("11065df7f8515818")
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 torch
import numpy as np
from typing import List

# Create a small stand-in class for args
class Args:
    def __init__(self):
        self.separator = ' | '

# Create the inputs list with dictionaries containing the required keys
inputs = [
    {
        'input_sentence_1': 'The cat sat',
        'input_sentence_2': 'on the mat',
        'input_sentence_3': 'and purred',
        'input_sentence_4': 'loudly',
        'answer': 'Yes'
    },
    {
        'input_sentence_1': 'Dogs are',
        'input_sentence_2': 'loyal animals',
        'input_sentence_3': 'that love',
        'input_sentence_4': 'their owners',
        'answer': 'True'
    }
]

args = Args()

ARGS = {
    'args': args,
    'inputs': inputs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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
format_prompt Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/language-modeling/winograd.py
code served (permissive licence) · get_code("522de10ce348cb0c")
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 torch

class Args:
    def __init__(self):
        self.separator = ' | '

ARGS = {
    'args': Args(),
    'inputs': [
        {'answer': 'First answer with\nnewlines'},
        {'answer': 'Second answer\nwith multiple\nlines'}
    ]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — 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
format_prompt Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/glue/glue.py
code served (permissive licence) · get_code("8131636afc6ecc7d")
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 re
from typing import List

class Args:
    def __init__(self, task_name, separator):
        self.task_name = task_name
        self.separator = separator

# Test with 'boolq' task
ARGS = {
    'args': Args(task_name='boolq', separator=' | '),
    'inputs': [
        {
            'passage': 'The sky is blue.',
            'question': 'Is the sky blue',
            'answer': 'yes'
        },
        {
            'passage': 'Water is wet.',
            'question': 'Is water wet',
            'answer': 'yes'
        }
    ]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
format_prompt Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/glue/multirc.py
code served (permissive licence) · get_code("1edc73f4d0b284d9")
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.

class Args:
    def __init__(self):
        self.separator = '\n'

ARGS = {
    'args': Args(),
    'inputs': [
        {
            'paragraph': 'The quick brown fox jumps over the lazy dog.',
            'question': 'What color is the fox?',
            'answers': ['brown', 'reddish-brown']
        },
        {
            'paragraph': 'Paris is the capital of France and is known for the Eiffel Tower.',
            'question': 'What is the capital of France?',
            'answers': ['Paris']
        }
    ]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
format_prompt Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/needle-in-a-haystack/haystack.py
code served (permissive licence) · get_code("67264be51677507e")
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 torch
import numpy as np

class Args:
    def __init__(self):
        self.separator = " | "
        self.model_name_or_path = "facebook/opt-350m"

ARGS = {
    'args': Args(),
    'inputs': {
        'lines_1': 'This is the first part of the article with some context.',
        'lines_2': 'This is the second part of the article.',
        'needle': 'The magic number is 42'
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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
format_prompt Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/question-answering/arc.py
code served (permissive licence) · get_code("f423eb95fad23611")
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 torch
import torch.nn as nn
from typing import List

torch.manual_seed(0)

# Create a simple args object with a separator attribute
class Args:
    def __init__(self, separator):
        self.separator = separator

args = Args(separator=' | ')

# Create the inputs list with question-answer pairs
inputs = [
    {
        'question': 'What is 2+2?',
        'answer': '4'
    },
    {
        'question': 'What is the capital of France?',
        'answer': 'Paris'
    }
]

ARGS = {
    'args': args,
    'inputs': inputs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — 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
format_prompt Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/question-answering/natural_questions.py
code served (permissive licence) · get_code("4f852be96a94479f")
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 string

torch_imported = False
try:
    import torch
    torch_imported = True
except ImportError:
    pass

# Construct valid arguments for format_prompt
questions = [
    "what is the capital of france",
    "how many continents are there"
]

answers = [
    "Paris",
    "Seven"
]

ARGS = {
    "questions": questions,
    "answers": answers
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — 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
format_prompt Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/question-answering/openbookqa.py
code served (permissive licence) · get_code("41aac6a681c0af69")
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 torch
import numpy as np
from typing import List

# Create a simple args object with a separator attribute
class Args:
    def __init__(self, separator: str):
        self.separator = separator

# Create sample inputs - list of dicts with 'question_stem' and 'answer' keys
inputs = [
    {
        'question_stem': 'What is the capital of France?',
        'answer': 'Paris'
    },
    {
        'question_stem': 'What is 2 + 2?',
        'answer': '4'
    }
]

# Create args object with a separator
args = Args(separator=' | ')

# Define ARGS mapping
ARGS = {
    'args': args,
    'inputs': inputs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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
normalize_answer Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/glue/record.py
code served (permissive licence) · get_code("c6a80c065d2e4851")
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 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 string
import re

def normalize_answer(s):
    """Lower text and remove punctuation, articles and extra whitespace."""

    def remove_articles(text):
        return re.sub('\\b(a|an|the)\\b', ' ', text)

    def white_space_fix(text):
        return ' '.join(text.split())

    def remove_punc(text):
        exclude = set(string.punctuation)
        return ''.join((ch for ch in text if ch not in exclude))

    def lower(text):
        return text.lower()
    return white_space_fix(remove_articles(remove_punc(lower(s))))


ARGS = {
    "s": "The Quick Brown Fox, jumps over the lazy dog!"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — 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
sample_random_examples Ran
ran; our contract draft was wrong, not the code
ltgoslo/bert-in-context/question-answering/natural_questions.py
code served (permissive licence) · get_code("a4e00754dd4bcfaa")
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 list, ndim 1, 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 random
from typing import List

# Create a small dataset with dict entries
dataset = [
    {"id": 0, "text": "example zero", "label": 1},
    {"id": 1, "text": "example one", "label": 0},
    {"id": 2, "text": "example two", "label": 1},
    {"id": 3, "text": "example three", "label": 0},
    {"id": 4, "text": "example four", "label": 1},
]

ARGS = {
    "dataset": dataset,
    "example_index": 0,
    "n_shots": 3,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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: none by the harness; any seed is in the setup block
get_matrix Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/needle-in-a-haystack/haystack_plot.py
code served (permissive licence) · get_code("86c6f499c0a026ce")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/language-modeling/hellaswag.py
code served (permissive licence) · get_code("3c0f2aa1a66632c5")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/language-modeling/story_cloze.py
code served (permissive licence) · get_code("3bd35418b3afc655")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/language-modeling/winograd.py
code served (permissive licence) · get_code("ea61f50fcaa11638")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/language-modeling/winogrande.py
code served (permissive licence) · get_code("414bac79226f48f5")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/glue/glue.py
code served (permissive licence) · get_code("dd674d997564c62d")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/glue/multirc.py
code served (permissive licence) · get_code("79d613cc4f9f531b")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/needle-in-a-haystack/haystack.py
code served (permissive licence) · get_code("1866651164e5bad9")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/question-answering/arc.py
code served (permissive licence) · get_code("7fbad6dad736d00b")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/question-answering/openbookqa.py
code served (permissive licence) · get_code("97a40c18e9c30df4")
How it ran

no invocation recorded.

load_model Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/language-modeling/hellaswag.py
code served (permissive licence) · get_code("081ead69959a12d1")
How it ran

no invocation recorded.

load_model Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/needle-in-a-haystack/haystack.py
code served (permissive licence) · get_code("3585e698dbdb0fc1")
How it ran

no invocation recorded.

load_model Not yet run
unverified (harvested, no recorded run)
ltgoslo/bert-in-context/question-answering/natural_questions.py
code served (permissive licence) · get_code("03857b5d2df81f28")
How it ran

no invocation recorded.

Also present in other papers' code

Of 26 samples lifted from this paper's linked repositories, 26 carry a function identity and 0 do not. Of the 26 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 3 the same month, 3 later) and 23 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
normalize_answer 133 other papers
77 dated earlier, 8 the same month, 47 later, 1 undated · 1 only as an assembly brick
1508.06615 (2015-08)
Character-Aware Neural Language Models
1606.05250 (2016-06)
SQuAD: 100,000+ Questions for Machine Comprehension of Text
1608.07905 (2016-08)
Machine Comprehension Using Match-LSTM and Answer Pointer
and 130 more not named here
exact_match_score 56 other papers
48 dated earlier, 1 the same month, 7 later
1508.06615 (2015-08)
Character-Aware Neural Language Models
1606.05250 (2016-06)
SQuAD: 100,000+ Questions for Machine Comprehension of Text
1608.07905 (2016-08)
Machine Comprehension Using Match-LSTM and Answer Pointer
and 53 more not named here
f1_score 41 other papers
39 dated earlier, 1 the same month, 1 later
1508.06615 (2015-08)
Character-Aware Neural Language Models
1606.05250 (2016-06)
SQuAD: 100,000+ Questions for Machine Comprehension of Text
1608.07905 (2016-08)
Machine Comprehension Using Match-LSTM and Answer Pointer
and 38 more not named here

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

While in-context learning is commonly associated with causal language models, such as GPT, we demonstrate that this capability also 'emerges' in masked language models. Through an embarrassingly simple inference technique, we enable an existing masked model, DeBERTa, to perform generative tasks without additional training or architectural changes. Our evaluation reveals that the masked and causal language models behave very differently, as they clearly outperform each other on different categories of tasks. These complementary strengths suggest that the field's focus on causal models for in-context learning may be limiting -both architectures can develop these capabilities, but with distinct advantages; pointing toward promising hybrid approaches that combine the strengths of both objectives.

For agents

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

get_harvested_code_for_paper("2406.04823")
get_code_for_paper("2406.04823")
have("2406.04823")

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

curl https://syntology.ai/api/ran/2406.04823.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.04823.svg)](https://syntology.ai/paper/2406.04823)

Connect an agent — have() is free.