SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2305.17852 · CVPR · 2023

Hierarchical Neural Memory Network for Low Latency Event Processing

Yasutaka Furukawa, Masaki Onishi, Ken Sakurada, Ryuhei Hamaguchi

arXiv · PDF · Open in the Atlas

Code that ran

6 of 6 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 6 have no contract verdict; 0 failures of our instrument; 0 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/2305.17852.json · badge: /api/ran/2305.17852.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
hamarh/HMNet_pth canonical (classifier label)
classifier basis: census
6 ran · 0 no recorded run
FunctionStatusWhere it lives
collate_keep_dict Ran
ran on a synthesized input
hamarh/HMNet_pth/hmnet/dataset/custom_collate_fn.py
code served (permissive licence) · get_code("f57f4b757d63b350")
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 2, dtype family float, shape (3, 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 numpy as np
import collections

torch.manual_seed(0)

# Test case 1: batch of tensors
batch_tensors = [torch.randn(4), torch.randn(4), torch.randn(4)]

# Test case 2: batch of dicts
batch_dicts = [
    {'a': torch.randn(4), 'b': 1},
    {'a': torch.randn(4), 'b': 2}
]

# Test case 3: batch of floats
batch_floats = [1.5, 2.5, 3.5]

# Test case 4: batch of ints
batch_ints = [1, 2, 3]

# Test case 5: batch of strings
batch_strings = ["hello", "world", "test"]

# Test case 6: batch of lists
batch_lists = [
    [torch.randn(4), torch.randn(4)],
    [torch.randn(4), torch.randn(4)]
]

# Test case 7: batch of numpy arrays
batch_numpy = [np.array([1.0, 2.0]), np.array([3.0, 4.0])]

# Use the simplest case - batch of tensors
ARGS = {
    'batch': batch_tensors
}
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 2, dtype family float, shape (3, 4) — shape (3, 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
depth_to_nlogdepth Ran
ran on a synthesized input
hamarh/HMNet_pth/hmnet/dataset/eventscape.py
code served (permissive licence) · get_code("272d5490165e6c7c")
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 (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 numpy as np

# The function takes a single parameter 'depth' which should be a numeric value or array
# The function computes: ndepth = np.log(depth / DEPTH_MAX) / ALPHA + 1
# So depth needs to be a positive number or array of positive numbers

# Create a simple depth value or array
depth_value = np.array([100.0, 500.0])

ARGS = {
    "depth": depth_value
}
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 (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
depth_to_nlogdepth Ran
ran on a synthesized input
hamarh/HMNet_pth/hmnet/dataset/mvsec.py
code served (permissive licence) · get_code("e0536a1be34d350d")
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 (3,) 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 single parameter: depth
# It performs a logarithmic transformation on depth values
# depth can be a scalar or numpy array

ARGS = {
    "depth": np.array([10.0, 40.0, 80.0]),
}
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 (3,) — shape (3,) 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_chunk Ran
ran on a synthesized input
hamarh/HMNet_pth/hmnet/utils/common.py
code served (permissive licence) · get_code("0343b784110cc40e")
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 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 torch

torch.manual_seed(0)

# The function get_chunk takes a target_list and either:
# 1. chunk_id and num_chunk parameters, or
# 2. a chunk_str parameter in format "chunk_id/num_chunks"
# It splits the target_list into num_chunk parts and returns the chunk_id-th part

# Create a simple target list
target_list = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]

# Define ARGS with all parameters
ARGS = {
    "target_list": target_list,
    "chunk_id": 0,
    "num_chunk": 2,
    "chunk_str": None
}
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 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
get_list Ran
ran on a synthesized input
hamarh/HMNet_pth/hmnet/utils/common.py
code served (permissive licence) · get_code("dbdb23b280c73c47")
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 os
import glob
import tempfile
import shutil

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

# Create some test files with different extensions
with open(os.path.join(temp_dir, 'file1.txt'), 'w') as f:
    f.write('content1')
with open(os.path.join(temp_dir, 'file2.txt'), 'w') as f:
    f.write('content2')
with open(os.path.join(temp_dir, 'file3.csv'), 'w') as f:
    f.write('content3')

# Create a test file list file
list_file = os.path.join(temp_dir, 'filelist.txt')
with open(list_file, 'w') as f:
    f.write('./test1.txt\n./test2.txt\n')

ARGS = {
    'dpath': temp_dir,
    'ext': 'txt',
    'root': None
}
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
split_list Ran
ran on a synthesized input
hamarh/HMNet_pth/hmnet/utils/common.py
code served (permissive licence) · get_code("e9b970a4bb929871")
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. 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:
# - split_list takes two parameters: target_list and num
# - target_list: needs to be a list (indexable with slicing)
# - num: needs to be an integer (used in division, range, and list multiplication)
# - The function splits target_list into num roughly equal parts

# Create a simple list to split
target_list = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]

# Number of splits
num = 3

ARGS = {
    "target_list": target_list,
    "num": num
}
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 list — a list 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

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 5 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 5 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

This paper proposes a low latency neural network architecture for event-based dense prediction tasks. Conventional architectures encode entire scene contents at a fixed rate regardless of their temporal characteristics. Instead, the proposed network encodes contents at a proper temporal scale depending on its movement speed. We achieve this by constructing temporal hierarchy using stacked latent memories that operate at different rates. Given low latency event steams, the multi-level memories gradually extract dynamic to static scene contents by propagating information from the fast to the slow memory modules. The architecture not only reduces the redundancy of conventional architectures but also exploits long-term dependencies. Furthermore, an attention-based event representation efficiently encodes sparse event streams into the memory cells. We conduct extensive evaluations on three event-based dense prediction tasks, where the proposed approach outperforms the existing methods on accuracy and latency, while demonstrating effective event and image fusion capabilities. The code is available at https://hamarh.github.io/ hmnet/.

For agents

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

get_harvested_code_for_paper("2305.17852")
get_code_for_paper("2305.17852")
have("2305.17852")

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

curl https://syntology.ai/api/ran/2305.17852.json

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

[![Syntology run record](https://syntology.ai/api/ran/2305.17852.svg)](https://syntology.ai/paper/2305.17852)

Connect an agent — have() is free.