SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.11929 · 2022

Word Segmentation on Discovered Phone Units with Dynamic Programming and Self-Supervised Scoring

arXiv · PDF · Open in the Atlas

Code that ran

17 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; 5 honoured the contract we drafted, 1 violated it and 5 have no contract verdict; 6 failures of our instrument; 2 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2202.11929.json · badge: /api/ran/2202.11929.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
kamperh/vqwordseg canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
kamperh/dpdp_aernn canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
jasonppy/word-discovery community (classifier label)
classifier basis not recorded on the link
12 ran · 2 no recorded run
FunctionStatusWhere it lives
swish Ran
ran; our contract draft was wrong, not the code
jasonppy/word-discovery/models/utils.py
code served (permissive licence) · get_code("0f786c407fb1ee4c")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-04T22:33:52Z). 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 swish takes a single parameter 'x' which is used with:
# - torch.sigmoid(x) - requires a tensor
# - x * ... - requires a tensor that supports multiplication
# So x should be a tensor

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
Margin_InfoNCE_loss Ran
ran on a synthesized input
jasonppy/word-discovery/models/utils.py
code served (permissive licence) · get_code("05ead5734e7a624a")
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 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)

# Create a similarity matrix S of shape (batch_size, batch_size)
# The function expects S to be a square matrix for the loss computation
batch_size = 2
S = torch.randn(batch_size, batch_size, dtype=torch.float32)

# margin parameter - a scalar value for the margin in the loss
margin = 0.1

# img_id parameter - optional numpy array of image IDs for masking
# Shape: (batch_size,) with integer IDs
img_id = np.array([0, 1], dtype=np.int64)

ARGS = {
    "S": S,
    "margin": margin,
    "img_id": img_id,
}
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 Tensor, 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
bool_flag Ran
ran, violated the contract we drafted
jasonppy/word-discovery/models/vit_utils.py
code served (permissive licence) · get_code("4d25be8ebf17fde1")
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 argparse

def bool_flag(s):
    """
    Parse boolean arguments from the command line.
    """
    FALSY_STRINGS = {'off', 'false', '0'}
    TRUTHY_STRINGS = {'on', 'true', '1'}
    if s.lower() in FALSY_STRINGS:
        return False
    elif s.lower() in TRUTHY_STRINGS:
        return True
    else:
        raise argparse.ArgumentTypeError('invalid value for a boolean flag')


ARGS = {
    's': 'true'
}
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 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
boundaries_to_intervals Ran
ran, honoured the contract we drafted
kamperh/vqwordseg/eval_segmentation.py
code served (permissive licence) · get_code("5cf9b5e522af96d5")
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 2, dtype family int; 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 numpy as np

# The function takes a 'boundaries' parameter which is used with np.where(boundaries)[0]
# This means it should be an array-like object that can be converted to a boolean array
# and indexed with np.where to find non-zero positions

# Create a simple boundaries array with some True/False values
boundaries = np.array([False, True, False, True, False, False, True, False])

ARGS = {
    "boundaries": boundaries
}
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 list, ndim 2, dtype family int — 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
clip_gradients Ran
ran, honoured the contract we drafted
jasonppy/word-discovery/models/vit_utils.py
code served (permissive licence) · get_code("2b5489053b222e00")
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 torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        return self.linear2(self.linear1(x))

model = SimpleModel()

# Create a simple forward pass and backward pass to populate gradients
x = torch.randn(2, 4)
y = torch.randn(2, 2)
output = model(x)
loss = nn.MSELoss()(output, y)
loss.backward()

clip = 1.0

ARGS = {
    "model": model,
    "clip": clip,
}
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 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
cls_attn_seg_feats Ran
ran on a synthesized input
jasonppy/word-discovery/save_seg_feats.py
code served (permissive licence) · get_code("8de801ce80b0728a")
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
from itertools import groupby
from operator import itemgetter

torch.manual_seed(0)

# Create test data
# feats: tensor of shape (T, D) where T is time steps, D is feature dimension
num_frames = 10
feat_dim = 4
feats = torch.randn(num_frames, feat_dim)

# cls_attn_weights: shape (num_heads, T) - attention weights over time
num_heads = 2
cls_attn_weights = torch.rand(num_heads, num_frames)

# threshold: quantile threshold (float between 0 and 1)
threshold = 0.5

# pool: pooling strategy
pool = 'mean'

# spf: seconds per frame (float)
spf = 0.1

# level2: boolean flag
level2 = False

# no_cls: not used in function body, but is a parameter
no_cls = False

ARGS = {
    'feats': feats,
    'cls_attn_weights': cls_attn_weights,
    'threshold': threshold,
    'pool': pool,
    'spf': spf,
    'level2': level2,
    'no_cls': no_cls,
}
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
cls_attn_seg_feats Ran
ran on a synthesized input
jasonppy/word-discovery/save_seg_feats_buckeye.py
code served (permissive licence) · get_code("f2fd30826872a75d")
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)

# Construct valid arguments for cls_attn_seg_feats
# feats: not actually used in the function body, but should be a tensor
feats = torch.randn(2, 4)

# cls_attn_weights: used for quantile, sum, comparison, shape[-1] access
# Shape: (batch_size, num_frames) - let's use (3, 5) for distinct dimensions
cls_attn_weights = torch.randn(3, 5)

# threshold: scalar value between 0 and 1 for quantile
threshold = 0.5

# spf: seconds per frame, a scalar float
spf = 0.5

# level2: boolean flag
level2 = False

# start_sec: start time in seconds
start_sec = 0.0

# end_sec: end time in seconds
end_sec = 5.0

ARGS = {
    'feats': feats,
    'cls_attn_weights': cls_attn_weights,
    'threshold': threshold,
    'spf': spf,
    'level2': level2,
    'start_sec': start_sec,
    'end_sec': end_sec,
}
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
cosine_scheduler Ran
ran, honoured the contract we drafted
jasonppy/word-discovery/models/vit_utils.py
code served (permissive licence) · get_code("361a6b24f11fc50a")
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 ndarray, ndim 1, dtype family float, shape (500,) 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

ARGS = {
    "base_value": 0.1,
    "final_value": 0.001,
    "epochs": 10,
    "niter_per_ep": 50,
    "warmup_epochs": 2,
    "start_warmup_value": 0.0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 (500,) — shape (500,) 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
drop_path Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
jasonppy/word-discovery/models/vision_transformer.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from gladzhang/Xformer, is pointer only (licence: NONE) · get_code("55120f2026b56aa2")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from gladzhang/Xformer, 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.

force_align_seg_feats Ran
ran on a synthesized input
jasonppy/word-discovery/save_seg_feats.py
code served (permissive licence) · get_code("ae600d378be4e580")
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
import torch.nn as nn

torch.manual_seed(0)

# Construct feats tensor: shape (total_frames, feat_dim)
# Using 30 frames with feature dimension 4
feats = torch.randn(30, 4)

# Construct text_alignment string
# Format: "start__token__end start__token__end ..."
# Each segment defined by start time (s), token name, end time (e)
text_alignment = "0.0__word1__1.0 1.0__word2__2.0 2.0__word3__3.0"

# FPS: frames per second (used to convert time to frame indices)
fps = 10

# Pool method: one of 'mean', 'max', or 'median'
pool = 'mean'

ARGS = {
    'feats': feats,
    'text_alignment': text_alignment,
    'fps': fps,
    'pool': pool
}
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
gelu Ran
ran, honoured the contract we drafted
jasonppy/word-discovery/models/utils.py
code served (permissive licence) · get_code("c2f0705ed8e6f7e6")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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 math

torch.manual_seed(0)

# The gelu function takes a single parameter 'x' which should be a tensor
# The function applies element-wise operations: multiplication, addition, torch.erf, and division
# So x should be a torch.Tensor

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
get_intervals_from_dir Ran
ran; our contract draft was wrong, not the code
kamperh/vqwordseg/eval_segmentation.py
code served (permissive licence) · get_code("5d99453f8b60341d")
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.

from pathlib import Path
import tempfile
import os

# Create a temporary directory with test files
temp_dir = tempfile.mkdtemp()
test_dir = Path(temp_dir)

# Create a few test text files with interval data
file1_path = test_dir / "file1.txt"
file1_path.write_text("0 10 label_a\n20 30 label_b\n")

file2_path = test_dir / "file2.txt"
file2_path.write_text("5 15 label_c\n25 35 label_d\n")

# Create a file that will be referenced but not exist (for testing missing files)
missing_file = "missing_file"

ARGS = {
    "directory": test_dir,
    "filenames": ["file1", "file2", missing_file]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tqdm — 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
get_intervals_from_dir Ran
ran; our contract draft was wrong, not the code
jasonppy/word-discovery/buckeye_eval.py
code served (permissive licence) · get_code("54dbd7b3ecbb6448")
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 tempfile
from pathlib import Path
from tqdm import tqdm

# Create a temporary directory with test files
temp_dir = tempfile.TemporaryDirectory()
directory = Path(temp_dir.name)

# Create test .txt files with interval data
file1_path = directory / "file1.txt"
file1_path.write_text("0 100 label_a\n50 150 label_b\n")

file2_path = directory / "file2.txt"
file2_path.write_text("10 50 label_c\n")

file3_path = directory / "file3.txt"
file3_path.write_text("")

# Set up arguments
ARGS = {
    "directory": directory,
    "filenames": None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tqdm — 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
get_segmented_sentence Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kamperh/dpdp_aernn/run_brent.py
code served (permissive licence) · get_code("6697d0657e11e88f")
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

torch.manual_seed(0)

# Analyze the function:
# - ids: indexed sequence (accessed as ids[i_symbol])
# - boundaries: sequence of booleans (accessed in loop, truthy check with "if boundary")
# - id_to_symbol: dict-like mapping from id to symbol string (accessed as id_to_symbol[ids[i_symbol]])
# - join_char: string parameter (default '')

# Create small test data
ids = [0, 1, 2, 1, 0]  # sequence of 5 ids
boundaries = [False, True, False, True, False]  # word boundaries
id_to_symbol = {
    0: 'hel',
    1: 'lo',
    2: 'wor'
}
join_char = ''

ARGS = {
    'ids': ids,
    'boundaries': boundaries,
    'id_to_symbol': id_to_symbol,
    'join_char': join_char
}
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
intervals_to_boundaries Ran
ran; our contract draft was wrong, not the code
kamperh/vqwordseg/eval_segmentation.py
code served (permissive licence) · get_code("bdd3717607f29767")
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 (8,) 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

# The function takes a list of intervals (tuples/lists with at least 2 elements)
# It accesses intervals[-1][1] to get the max boundary value
# It accesses i[1] for each interval i to index into the boundaries array

intervals = [
    (0, 3),
    (1, 5),
    (2, 8)
]

ARGS = {
    "intervals": intervals
}
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 (8,) — shape (8,) 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
neg_log_psuedo_geometric Ran
ran, honoured the contract we drafted
kamperh/dpdp_aernn/run_brent.py
code served (permissive licence) · get_code("77e3e9a881b530dc")
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 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 neg_log_psuedo_geometric takes a single parameter 'dur'
# and performs arithmetic: -(dur - 1)
# 'dur' can be a scalar, tensor, or any numeric type that supports subtraction

# Create a small tensor to use as 'dur'
dur = torch.tensor(5.0)

ARGS = {
    "dur": dur
}
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 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
parse_faiss_specs Ran
ran on a synthesized input
jasonppy/word-discovery/run_kmeans.py
code served (permissive licence) · get_code("35070d2f37830ca3")
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 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.

from collections import namedtuple

faiss_spec = namedtuple('faiss_spec', ['pca', 'norm', 'n_clus', 'sphere', 'spec_str'])

def parse_faiss_specs(specs_str):
    specs = []
    for ss in specs_str.split():
        comps = ss.split('_')
        pca = 0
        norm = False
        n_clus = 0
        sphere = False
        for c in comps:
            if c.startswith('PCA'):
                pca = int(c[3:])
            elif c == 'NORM':
                norm = True
            elif c.startswith('CLUS'):
                n_clus = int(c[4:])
            elif c == 'SPHERICAL':
                sphere = True
        assert n_clus > 0
        specs.append(faiss_spec(pca=pca, norm=norm, n_clus=n_clus, sphere=sphere, spec_str=ss))
    return specs

ARGS = {
    'specs_str': 'PCA128_NORM_CLUS256_SPHERICAL PCA64_CLUS128'
}
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 2, 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
vit_small Not yet run
unverified (harvested, no recorded run)
jasonppy/word-discovery/models/vision_transformer.py
code served (permissive licence) · get_code("3aae9623741abc40")
How it ran

no invocation recorded.

vit_tiny Not yet run
unverified (harvested, no recorded run)
jasonppy/word-discovery/models/vision_transformer.py
code served (permissive licence) · get_code("7d962f3bf8a0e816")
How it ran

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, 13 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 3 the same month, 13 later) and 6 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
drop_path 224 other papers
23 dated earlier, 2 the same month, 196 later, 2 same year, order not determinable, 1 undated · 38 only as an assembly brick
2003.12039 (2020-03)
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
2007.09384 (2020-07)
Multi-Scale Positive Sample Refinement for Few-Shot Object Detection
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
and 221 more not named here
swish 206 other papers
92 dated earlier, 2 the same month, 108 later, 4 same year, order not determinable · 44 only as an assembly brick
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1805.12114 (2018-05)
Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
1809.01357 (2018-09)
Zero Shot Learning for Code Education: Rubric Sampling with Deep Learning Inference
and 203 more not named here
gelu 198 other papers
98 dated earlier, 2 the same month, 94 later, 4 same year, order not determinable · 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
cosine_scheduler 37 other papers
6 dated earlier, 30 later, 1 same year, order not determinable
2005.10242 (2020-05)
Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere
2103.15375 (2021-03)
AlignMixup: Improving Representations By Interpolating Aligned Features
2106.09681 (2021-06)
XCiT: Cross-Covariance Image Transformers
and 34 more not named here
vit_tiny 33 other papers
1 dated earlier, 31 later, 1 undated
2201.02609 (2022-01)
Generalized Category Discovery
2203.08414 (2022-03)
Unsupervised Semantic Segmentation by Distilling Feature Correspondences
2203.14415 (2022-03)
Mugs: A Multi-Granular Self-Supervised Learning Framework
and 30 more not named here
bool_flag 31 other papers
3 dated earlier, 27 later, 1 same year, order not determinable
2106.09681 (2021-06)
XCiT: Cross-Covariance Image Transformers
2109.13228 (2021-09)
PASS: An ImageNet replacement for self-supervised pretraining without humans
2111.12958 (2021-11)
Self-Distilled Self-Supervised Representation Learning
and 28 more not named here
clip_gradients 28 other papers
3 dated earlier, 24 later, 1 same year, order not determinable
2106.09681 (2021-06)
XCiT: Cross-Covariance Image Transformers
2109.13228 (2021-09)
PASS: An ImageNet replacement for self-supervised pretraining without humans
2111.12958 (2021-11)
Self-Distilled Self-Supervised Representation Learning
and 25 more not named here
vit_small 22 other papers
1 dated earlier, 20 later, 1 undated
2201.02609 (2022-01)
Generalized Category Discovery
2203.08414 (2022-03)
Unsupervised Semantic Segmentation by Distilling Feature Correspondences
2204.07305 (2022-04)
Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference
and 19 more not named here
boundaries_to_intervals 2 other papers
2 later
2204.13094 (2022-04)
Unsupervised Word Segmentation using K Nearest Neighbors
2406.08380 (2024-06)
Towards Unsupervised Speech Recognition Without Pronunciation Models
get_intervals_from_dir 2 other papers
2 later
2204.13094 (2022-04)
Unsupervised Word Segmentation using K Nearest Neighbors
2406.08380 (2024-06)
Towards Unsupervised Speech Recognition Without Pronunciation Models
intervals_to_boundaries 2 other papers
2 later
2204.13094 (2022-04)
Unsupervised Word Segmentation using K Nearest Neighbors
2406.08380 (2024-06)
Towards Unsupervised Speech Recognition Without Pronunciation Models
Margin_InfoNCE_loss 1 other paper
1 later
2305.11435 (2023-05)
Syllable Discovery and Cross-Lingual Generalization in a Visually Grounded, Self-Supervised Speech Model
parse_faiss_specs 1 other paper
1 later
2305.11435 (2023-05)
Syllable Discovery and Cross-Lingual Generalization in a Visually Grounded, Self-Supervised Speech Model

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

Recent work on unsupervised speech segmentation has used self-supervised models with phone and word segmentation modules that are trained jointly. This paper instead revisits an older approach to word segmentation: bottom-up phone-like unit discovery is performed first, and symbolic word segmentation is then performed on top of the discovered units (without influencing the lower level). To do this, I propose a new unit discovery model, a new symbolic word segmentation model, and then chain the two models to segment speech. Both models use dynamic programming to minimize segment costs from a self-supervised network with an additional duration penalty that encourages longer units. Concretely, for acoustic unit discovery, duration-penalized dynamic programming (DPDP) is used with a contrastive predictive coding model as the scoring network. For word segmentation, DPDP is applied with an autoencoding recurrent neural as the scoring network. The two models are chained in order to segment speech. This approach gives comparable word segmentation results to state-of-the-art joint self-supervised segmentation models on an English benchmark. On French, Mandarin, German and Wolof data, it outperforms previous systems on the ZeroSpeech benchmarks. Analysis shows that the chained DPDP system segments shorter filler words well, but longer words might require some external top-down signal.

For agents

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

get_harvested_code_for_paper("2202.11929")
get_code_for_paper("2202.11929")
have("2202.11929")

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

curl https://syntology.ai/api/ran/2202.11929.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.11929.svg)](https://syntology.ai/paper/2202.11929)

Connect an agent — have() is free.