SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1906.05807 · 2019

Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index

arXiv · PDF · Open in the Atlas

Code that ran

16 of 19 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; 4 honoured the contract we drafted, 1 violated it and 8 have no contract verdict; 3 failures of our instrument; 3 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/1906.05807.json · badge: /api/ran/1906.05807.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
uwnlp/denspi canonical (classifier label)
classifier basis: census
16 ran · 3 no recorded run
FunctionStatusWhere it lives
adjust Ran
ran on a synthesized input
uwnlp/denspi/open/mips.py
code served (permissive licence) · get_code("805c7cf81b34cfb8")
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 dict. 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

torch.manual_seed(0)

# The function 'adjust' takes a dict 'each' with keys:
# - 'context': a string
# - 'start_pos': an integer
# - 'end_pos': an integer
# It searches for ' [PAR] ' markers in the context and adjusts positions

ARGS = {
    'each': {
        'context': 'hello world [PAR] this is a test [PAR] end of text',
        'start_pos': 20,
        'end_pos': 25
    }
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
convert_to_unicode Ran
ran; our contract draft was wrong, not the code
uwnlp/denspi/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.

encode_phrase Ran
ran on a synthesized input
uwnlp/denspi/phrase.py
code served (permissive licence) · get_code("b29057ff0bdfab27")
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 tuple. 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

torch.manual_seed(0)

# The function expects:
# - layer: a tensor with shape (batch, channels, sequence_length)
# - phrase_size: an integer
# - span_kq_size: an integer (default 64)
# - get_first_only: a boolean (default False)

# From the assertions and operations:
# - layer.size(2) must be >= phrase_size - 1 + span_kq_size
# - boundary_layer has shape (batch, channels, phrase_size - 1)
# - span_layer has shape (batch, channels, span_kq_size)
# - These are chunked along dim 2, so phrase_size - 1 and span_kq_size must be even

batch_size = 2
channels = 3
phrase_size = 10  # phrase_size - 1 = 9 (even)
span_kq_size = 8  # even number
sequence_length = phrase_size - 1 + span_kq_size  # 9 + 8 = 17

# Create a layer tensor with appropriate shape
layer = torch.randn(batch_size, channels, sequence_length)

ARGS = {
    "layer": layer,
    "phrase_size": phrase_size,
    "span_kq_size": span_kq_size,
    "get_first_only": False,
}
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 tuple — a tuple output has no array shape of its own; none is 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
exact_match_score Ran
ran, violated the contract we drafted
uwnlp/denspi/evaluate-v1.1.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
uwnlp/denspi/evaluate-v1.1.py
code served (permissive licence) · get_code("2112c433b9c6d343")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) 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.

ARGS = {
    "prediction": "The quick brown fox jumps over the lazy dog.",
    "ground_truth": "A quick brown fox jumped over a lazy dog"
}
Setup block drafted by
a language model, Claude Sonnet 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 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
filter_metadata Ran
ran on a synthesized input
uwnlp/denspi/post.py
code served (permissive licence) · get_code("2ce75402d89eabd7")
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 dict. 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 numpy as np

# Construct metadata dictionary with the required structure
metadata = {
    'filter_start': np.array([0.1, 0.5, 0.9, 0.3]),
    'filter_end': np.array([0.2, 0.6, 0.8, 0.4]),
    'word2char_start': np.array([0, 5, 12, 18]),
    'word2char_end': np.array([4, 11, 17, 22]),
    'start': np.array([0, 1, 2, 3]),
    'end': np.array([1, 2, 3, 4]),
    'span_logits': np.array([[0.1, 0.2, 0.3, 0.4],
                             [0.5, 0.6, 0.7, 0.8],
                             [0.9, 1.0, 1.1, 1.2],
                             [1.3, 1.4, 1.5, 1.6]]),
    'start2end': np.array([[0, 1, 2],
                           [1, 2, 3],
                           [2, 3, 1],
                           [0, 1, 3]])
}

threshold = 0.4

ARGS = {
    'metadata': metadata,
    'threshold': threshold
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
gelu Ran
ran, honoured the contract we drafted
uwnlp/denspi/bert.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from shmsw25/qa-hard-em, is pointer only (licence: NONE) · get_code("40e9fee2e0b7e278")
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 shmsw25/qa-hard-em, 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.

get_logits Ran
ran on a synthesized input
uwnlp/denspi/phrase.py
code served (permissive licence) · get_code("c3c09581ddf6a885")
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 Tensor, ndim 1, dtype family float, shape (2,) 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 torch

torch.manual_seed(0)

# Create small tensors for a and b
# Shape: (batch=2, dim=4) for both a and b
a = torch.randn(2, 4)
b = torch.randn(2, 4)

# metric is a string that determines the computation type
metric = 'ip'

ARGS = {
    'a': a,
    'b': b,
    'metric': metric
}
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 Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) 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
int8_to_float Ran
ran on a synthesized input
uwnlp/denspi/open/mips.py
code served (permissive licence) · get_code("892f4e6b7e006e15")
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 ndarray, ndim 1, dtype family float, shape (5,) 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 numpy as np

# Create a small int8 array
num = np.array([0, 50, 100, 127, -128], dtype=np.int8)

# Define offset and factor parameters
offset = 0.5
factor = 128.0

# Construct the ARGS dictionary
ARGS = {
    "num": num,
    "offset": offset,
    "factor": factor
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (5,) — shape (5,) 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
load_backward Ran
ran on a synthesized input
uwnlp/denspi/run_piqa.py
code served (permissive licence) · get_code("cba1d07a90a46468")
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 OrderedDict. 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 collections

torch.manual_seed(0)

# The function load_backward takes a state dict (OrderedDict or dict)
# and transforms keys by:
# 1. Checking for 'module.' prefix (DataParallel indicator)
# 2. Skipping 'true_help' key
# 3. Skipping keys starting with 'bert_q.'
# 4. Skipping keys starting with 'linear.'
# 5. Renaming keys starting with 'bert.' to 'encoder.'
# 6. Re-adding 'module.' prefix if it was present

# Create a realistic state dict with various key patterns
state = collections.OrderedDict([
    ('module.encoder.layer1.weight', torch.randn(4, 3)),
    ('module.encoder.layer1.bias', torch.randn(4)),
    ('module.bert.embeddings.weight', torch.randn(5, 6)),
    ('bert_q.attention.weight', torch.randn(4, 3)),
    ('linear.weight', torch.randn(2, 8)),
    ('true_help', torch.randn(3, 3)),
    ('module.other.param', torch.randn(2, 2)),
    ('bert.encoder.layer.weight', torch.randn(7, 8)),
])

ARGS = {
    'state': state
}
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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
load_vocab Ran
ran on a synthesized input
uwnlp/denspi/tokenization.py
code served (permissive licence) · get_code("c8956ce86812a6c2")
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 OrderedDict. 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 collections
import tempfile
import os

# Create a temporary vocabulary file
with tempfile.NamedTemporaryFile(mode='w', suffix='.txt', delete=False) as f:
    vocab_file_path = f.name
    f.write("hello\n")
    f.write("world\n")
    f.write("test\n")
    f.write("vocab\n")

try:
    ARGS = {
        'vocab_file': vocab_file_path
    }
finally:
    # Clean up the temporary file after defining ARGS
    pass
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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
uwnlp/denspi/evaluate-v1.1.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
overlap Ran
ran on a synthesized input
uwnlp/denspi/open/mips.py
code served (permissive licence) · get_code("df3944b07a180cef")
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 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 torch
import numpy as np

torch.manual_seed(0)

# The function overlap(t1, t2, a1, a2, b1, b2) accesses:
# - t1[b1], t1[a1] (indexing into t1 with integers b1 and a1)
# - t2[a2], t2[b2] (indexing into t2 with integers b2 and a2)
# So t1 and t2 need to be indexable (lists, tensors, or arrays)
# a1, a2, b1, b2 are used as indices, so they should be integers

# Create two tensors that can be indexed
t1 = torch.tensor([1.0, 2.0, 3.0, 4.0])
t2 = torch.tensor([0.5, 1.5, 2.5, 3.5])

# Define integer indices for the parameters
a1 = 0
a2 = 1
b1 = 2
b2 = 3

ARGS = {
    't1': t1,
    't2': t2,
    'a1': a1,
    'a2': a2,
    'b1': b1,
    'b2': b2
}
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 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
printable_text Ran
ran; our contract draft was wrong, not the code
uwnlp/denspi/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
warmup_constant Ran
ran, honoured the contract we drafted
uwnlp/denspi/optimization.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Yueeeeeeee/Self-Supervised-QA, is pointer only (licence: NONE) · get_code("e7d542062316094a")
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 Yueeeeeeee/Self-Supervised-QA, 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.

warmup_linear Ran
ran, honoured the contract we drafted
uwnlp/denspi/optimization.py
code served (permissive licence) · get_code("c58d57224530d17e")
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 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 torch

torch.manual_seed(0)

# The function warmup_linear takes two parameters:
# - x: a numeric value (float) representing a position/step in training
# - warmup: a float representing the warmup threshold (default 0.002)
# 
# The function uses x and warmup in simple arithmetic comparisons and operations,
# so we just need numeric values.

ARGS = {
    "x": 0.001,
    "warmup": 0.002
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — 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
compress_metadata Not yet run
unverified (harvested, no recorded run)
uwnlp/denspi/post.py
code served (permissive licence) · get_code("13b8c556b20f1352")
How it ran

no invocation recorded.

get_sparse_logits Not yet run
unverified (harvested, no recorded run)
uwnlp/denspi/phrase.py
code served (permissive licence) · get_code("ad8fbb8cb7f3dc0a")
How it ran

no invocation recorded.

warmup_cosine Not yet run
unverified (harvested, no recorded run)
uwnlp/denspi/optimization.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Yueeeeeeee/Self-Supervised-QA, is pointer only (licence: NONE) · get_code("35f7cddf90dd05d4")
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 Yueeeeeeee/Self-Supervised-QA, 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.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 distinct functions with an identity, 9 are also present in other papers whose code we hold (9 with at least one paper dated earlier, 6 the same month, 9 later) and 10 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
gelu 198 other papers
8 dated earlier, 4 the same month, 186 later · 32 only as an assembly brick
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.10125 (2019-01)
Glyce: Glyph-vectors for Chinese Character Representations
and 195 more not named here
normalize_answer 133 other papers
11 dated earlier, 121 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
12 dated earlier, 1 the same month, 43 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
10 dated earlier, 1 the same month, 30 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
convert_to_unicode 36 other papers
8 dated earlier, 28 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
8 dated earlier, 25 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
warmup_constant 28 other papers
2 dated earlier, 1 the same month, 25 later
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1903.09588 (2019-03)
Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence
1906.01698 (2019-06)
Open Sesame: Getting Inside BERT's Linguistic Knowledge
and 25 more not named here
warmup_linear 19 other papers
6 dated earlier, 1 the same month, 12 later
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1803.03376 (2018-03)
Learning Approximate Inference Networks for Structured Prediction
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
and 16 more not named here
warmup_cosine 17 other papers
2 dated earlier, 1 the same month, 14 later
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1903.09588 (2019-03)
Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence
1906.01698 (2019-06)
Open Sesame: Getting Inside BERT's Linguistic Knowledge
and 14 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

Existing open-domain question answering (QA) models are not suitable for real-time usage because they need to process several long documents on-demand for every input query. In this paper, we introduce the query-agnostic indexable representation of document phrases that can drastically speed up open-domain QA and also allows us to reach long-tail targets. In particular, our dense-sparse phrase encoding effectively captures syntactic, semantic, and lexical information of the phrases and eliminates the pipeline filtering of context documents. Leveraging optimization strategies, our model can be trained in a single 4-GPU server and serve entire Wikipedia (up to 60 billion phrases) under 2TB with CPUs only. Our experiments on SQuAD-Open show that our model is more accurate than DrQA (Chen et al., 2017) with 6000x reduced computational cost, which translates into at least 58x faster end-to-end inference benchmark on CPUs.

For agents

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

get_harvested_code_for_paper("1906.05807")
get_code_for_paper("1906.05807")
have("1906.05807")

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

curl https://syntology.ai/api/ran/1906.05807.json

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

[![Syntology run record](https://syntology.ai/api/ran/1906.05807.svg)](https://syntology.ai/paper/1906.05807)

Connect an agent — have() is free.