SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.04831 · 2020

Improving Local Identifiability in Probabilistic Box Embeddings

arXiv · PDF · Open in the Atlas

Code that ran

5 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 0 failures of our instrument; 4 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/2010.04831.json · badge: /api/ran/2010.04831.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
iesl/geometric_graph_embedding community (classifier label)
classifier basis not recorded on the link
5 ran · 4 no recorded run
FunctionStatusWhere it lives
from_linkage_matrix Ran
ran on a synthesized input
iesl/geometric_graph_embedding/src/graph_modeling/generate/hac.py
code served (permissive licence) · get_code("ead45e312b908abd")
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 DiGraph. 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

# Create a small linkage matrix Z for the hierarchical clustering function
# The linkage matrix has shape (n-1, 4) where n is the number of samples
# For this example, we'll use n=3, so Z.shape = (2, 4)
# Columns: [idx1, idx2, distance, sample_count]
Z = np.array([
    [0, 1, 0.5, 2],
    [2, 3, 1.5, 3]
], dtype=np.float64)

ARGS = {
    "Z": Z
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
networkx, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DiGraph — a DiGraph output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DiGraph output the recorded ndim 1 and dtype family int 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
generate Ran
ran on a synthesized input
iesl/geometric_graph_embedding/src/graph_modeling/generate/hac.py
code served (permissive licence) · get_code("32e8c69cc44844db")
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 DiGraph. 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
import os
import numpy as np

# Create a temporary vector file with test data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_path = temp_file.name

# Generate simple test vectors
np.random.seed(0)
num_samples = 16  # 2^4 samples for log_num_nodes=4
vector_dim = 5

for i in range(num_samples):
    # Create a random vector
    vec = np.random.randn(vector_dim)
    # Normalize it
    vec = vec / (np.linalg.norm(vec) + 1e-8)
    # Write: id, label, vector components
    label = f"label_{i}"
    vec_str = '\t'.join(str(v) for v in vec)
    temp_file.write(f"{i}\t{label}\t{vec_str}\n")

temp_file.close()

ARGS = {
    'log_num_nodes': 4,
    'seed': 42,
    'vector_file': temp_path,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
loguru, networkx, numpy, scipy, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DiGraph — a DiGraph output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DiGraph output the recorded ndim 1 and dtype family int 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
generate Ran
ran on a synthesized input
iesl/geometric_graph_embedding/src/graph_modeling/generate/knn_graph.py
code served (permissive licence) · get_code("6b17a38b61efff88")
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 DiGraph. 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
import os
import numpy as np

# Create a temporary vector file with sample data
temp_dir = tempfile.mkdtemp()
vector_file = os.path.join(temp_dir, "vectors.txt")

# Generate sample vector data
# Format: id\tlabel\tvec1\tvec2\tvec3...
np.random.seed(0)
num_vectors = 32  # 2^5, so we can use log_num_nodes=5
vec_dim = 8

with open(vector_file, 'w') as f:
    for i in range(num_vectors):
        vec = np.random.randn(vec_dim)
        # Normalize
        vec = vec / np.linalg.norm(vec)
        vec_str = '\t'.join([str(x) for x in vec])
        f.write(f"id_{i}\tlabel_{i%3}\t{vec_str}\n")

ARGS = {
    'log_num_nodes': 5,
    'seed': 42,
    'vector_file': vector_file,
    'k': 3,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
loguru, networkx, numpy, scipy, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DiGraph — a DiGraph output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DiGraph output the recorded ndim 1 and dtype family int 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_text Ran
ran on a synthesized input
iesl/geometric_graph_embedding/src/graph_modeling/generate/hac.py
code served (permissive licence) · get_code("f09239998b1749d1")
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 tempfile
import os
import numpy as np

# Create a temporary file with valid input data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
try:
    # Write sample data: id \t label \t vector_values
    temp_file.write("0\tclass_a\t0.5\t0.3\t0.2\t0.9\n")
    temp_file.write("1\tclass_b\t0.1\t0.4\t0.7\t0.2\n")
    temp_file.write("2\tclass_a\t0.6\t0.1\t0.3\t0.5\n")
    temp_file.close()
    
    ARGS = {
        'infile': temp_file.name
    }
finally:
    pass
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
loguru, numpy — 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
remove_self_loops Ran
ran on a synthesized input
iesl/geometric_graph_embedding/src/graph_modeling/generate/generic.py
code served (permissive licence) · get_code("30dc892e2be52489")
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 Graph. 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 networkx as nx

# Create a small graph with self-loops and regular edges
G = nx.Graph()
G.add_edges_from([(0, 1), (1, 2), (2, 2), (0, 0), (1, 3)])

ARGS = {
    'G': G
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
networkx — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Graph — a Graph output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Graph output the recorded ndim 1 and dtype family int 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_outtree Not yet run
unverified (harvested, no recorded run)
iesl/geometric_graph_embedding/src/graph_modeling/generate/generic.py
code served (permissive licence) · get_code("74b56dee8af9c94d")
How it ran

no invocation recorded.

generate Not yet run
unverified (harvested, no recorded run)
iesl/geometric_graph_embedding/src/graph_modeling/generate/kronecker_graph.py
code served (permissive licence) · get_code("9c877d5a9cfe2a1d")
How it ran

no invocation recorded.

generate Not yet run
unverified (harvested, no recorded run)
iesl/geometric_graph_embedding/src/graph_modeling/generate/price.py
code served (permissive licence) · get_code("3e86800f7a81f156")
How it ran

no invocation recorded.

generate Not yet run
unverified (harvested, no recorded run)
iesl/geometric_graph_embedding/src/graph_modeling/generate/random_tree.py
code served (permissive licence) · get_code("dc0c718b09c9a0c8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 9 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

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

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

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

Abstract

Geometric embeddings have recently received attention for their natural ability to represent transitive asymmetric relations via containment. Box embeddings, where objects are represented by n-dimensional hyperrectangles, are a particularly promising example of such an embedding as they are closed under intersection and their volume can be calculated easily, allowing them to naturally represent calibrated probability distributions. The benefits of geometric embeddings also introduce a problem of local identifiability, however, where whole neighborhoods of parameters result in equivalent loss which impedes learning. Prior work addressed some of these issues by using an approximation to Gaussian convolution over the box parameters, however, this intersection operation also increases the sparsity of the gradient. In this work, we model the box parameters with min and max Gumbel distributions, which were chosen such that space is still closed under the operation of the intersection. The calculation of the expected intersection volume involves all parameters, and we demonstrate experimentally that this drastically improves the ability of such models to learn.

For agents

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

get_harvested_code_for_paper("2010.04831")
get_code_for_paper("2010.04831")
have("2010.04831")

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

curl https://syntology.ai/api/ran/2010.04831.json

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

[![Syntology run record](https://syntology.ai/api/ran/2010.04831.svg)](https://syntology.ai/paper/2010.04831)

Connect an agent — have() is free.