SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.08751 · 2020

Knowledge-Based Video Question Answering with Unsupervised Scene Descriptions

arXiv · PDF · Open in the Atlas

Code that ran

7 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 7 have no contract verdict; 0 failures of our instrument; 2 no recorded run), measured 2026-11-21; 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/2007.08751.json · badge: /api/ran/2007.08751.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
noagarcia/ROLL-VideoQA canonical (classifier label)
classifier basis: census
7 ran · 2 no recorded run
FunctionStatusWhere it lives
accuracy Ran
ran on a synthesized input
noagarcia/ROLL-VideoQA/Source/utils.py
code served (permissive licence) · get_code("3b3e645279cd83cf")
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, 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 pandas as pd
import torch

torch.manual_seed(0)

# Create a DataFrame with QType column
df = pd.DataFrame({
    'QType': ['visual', 'textual', 'temporal', 'knowledge', 'visual', 'textual']
})

# Create output predictions (6 samples)
out = [0, 1, 2, 3, 0, 1]

# Create ground truth labels (6 samples)
label = [0, 1, 2, 3, 1, 1]

# Create indices into the dataframe
index = [0, 1, 2, 3, 4, 5]

ARGS = {
    'df': df,
    'out': out,
    'label': label,
    'index': index
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple 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
box_area Ran
ran on a synthesized input
noagarcia/ROLL-VideoQA/Source/utils_graphs.py
code served (permissive licence) · get_code("13fbfd3eb97edd3a")
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
import numpy as np

torch.manual_seed(0)

# Construct arguments for box_area function
# The function takes scalar or array-like coordinates and computes area
# Based on the formula: (right - left + 1) * (bottom - top + 1)

left = torch.tensor([1.0, 5.0])
top = torch.tensor([2.0, 3.0])
right = torch.tensor([10.0, 15.0])
bottom = torch.tensor([8.0, 12.0])

ARGS = {
    "left": left,
    "top": top,
    "right": right,
    "bottom": bottom
}
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 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
box_centroid Ran
ran on a synthesized input
noagarcia/ROLL-VideoQA/Source/utils_graphs.py
code served (permissive licence) · get_code("0050ace8c8d75386")
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, 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

torch.manual_seed(0)

ARGS = {
    "left": 10,
    "top": 20,
    "right": 50,
    "bottom": 80
}
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, ndim 1, dtype family float — a tuple 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
boxes_intersect Ran
ran on a synthesized input
noagarcia/ROLL-VideoQA/Source/utils_graphs.py
code served (permissive licence) · get_code("4ec85f637b3305fc")
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, 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.

import numpy as np

# The function boxes_intersect takes 8 parameters representing two bounding boxes
# Each bounding box is defined by left, top, right, bottom coordinates
# The function uses np.maximum and np.minimum, so these should be numpy arrays or scalars

# Create small numpy arrays for the bounding box coordinates
bbox1_left = np.array([1.0, 2.0])
bbox1_top = np.array([3.0, 4.0])
bbox1_right = np.array([5.0, 6.0])
bbox1_bottom = np.array([7.0, 8.0])

bbox2_left = np.array([1.5, 2.5])
bbox2_top = np.array([3.5, 4.5])
bbox2_right = np.array([4.5, 5.5])
bbox2_bottom = np.array([6.5, 7.5])

ARGS = {
    'bbox1_left': bbox1_left,
    'bbox1_top': bbox1_top,
    'bbox1_right': bbox1_right,
    'bbox1_bottom': bbox1_bottom,
    'bbox2_left': bbox2_left,
    'bbox2_top': bbox2_top,
    'bbox2_right': bbox2_right,
    'bbox2_bottom': bbox2_bottom,
}
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 tuple, ndim 2, dtype family float — a tuple 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
cleanhtml Ran
ran on a synthesized input
noagarcia/ROLL-VideoQA/Source/dataloader_knowit.py
code served (permissive licence) · get_code("c7368d407afb9277")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def cleanhtml(raw_html):
    cleanr = re.compile('<.*?>')
    cleantext = re.sub(cleanr, '', raw_html)
    return cleantext

ARGS = {
    'raw_html': '<div class="container"><p>Hello World</p></div>'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_knowit_data Ran
ran on a synthesized input
noagarcia/ROLL-VideoQA/Source/dataloader_knowit.py
code served (permissive licence) · get_code("cc48625c00cb4f6f")
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 DataFrame. 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 pandas as pd
import logging

# Set up logging
logging.basicConfig(level=logging.INFO)

# Create a temporary directory and CSV files for the function to read
temp_dir = tempfile.mkdtemp()
knowit_data_dir = os.path.join(temp_dir, 'knowit_data')
os.makedirs(knowit_data_dir, exist_ok=True)

# Create minimal CSV files with tab delimiter
train_df = pd.DataFrame({
    'col1': ['a', 'b'],
    'col2': [1, 2]
})
val_df = pd.DataFrame({
    'col1': ['c', 'd'],
    'col2': [3, 4]
})
test_df = pd.DataFrame({
    'col1': ['e', 'f'],
    'col2': [5, 6]
})

train_df.to_csv(os.path.join(knowit_data_dir, 'knowit_data_train.csv'), sep='\t', index=False)
val_df.to_csv(os.path.join(knowit_data_dir, 'knowit_data_val.csv'), sep='\t', index=False)
test_df.to_csv(os.path.join(knowit_data_dir, 'knowit_data_test.csv'), sep='\t', index=False)

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

args = Args(data_dir=temp_dir)

ARGS = {
    'args': args,
    'split_name': 'train'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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_obj Ran
ran on a synthesized input
noagarcia/ROLL-VideoQA/Source/utils.py
code served (permissive licence) · get_code("ae9474aa34099c9c")
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 pickle
import tempfile
import os

# Create a temporary pickle file with a simple object
temp_dir = tempfile.gettempdir()
test_obj = {"key": "value", "data": [1, 2, 3]}
temp_filename = os.path.join(temp_dir, "test_object.pkl")

with open(temp_filename, 'wb') as f:
    pickle.dump(test_obj, f)

ARGS = {
    "filename": temp_filename,
    "verbose": 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 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_all_scenes_dataset Not yet run
unverified (harvested, no recorded run)
noagarcia/ROLL-VideoQA/Source/generate_scene_description.py
code served (permissive licence) · get_code("0bb71d7f1ef98525")
How it ran

no invocation recorded.

load_visual_features Not yet run
unverified (harvested, no recorded run)
noagarcia/ROLL-VideoQA/Source/generate_scene_description.py
code served (permissive licence) · get_code("504538556ae9be87")
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, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 8 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
cleanhtml 1 other paper
1 later
2210.03849 (2022-10)
CONVFINQA: Exploring the Chain of Numerical Reasoning in Conversational Finance Question Answering

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 2026-11-21.

Repositories linked to this paper

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

Abstract

To understand movies, humans constantly reason over the dialogues and actions shown in specific scenes and relate them to the overall storyline already seen. Inspired by this behaviour, we design ROLL, a model for knowledge-based video story question answering that leverages three crucial aspects of movie understanding: dialog comprehension, scene reasoning, and storyline recalling. In ROLL, each of these tasks is in charge of extracting rich and diverse information by 1) processing scene dialogues, 2) generating unsupervised video scene descriptions, and 3) obtaining external knowledge in a weakly supervised fashion. To answer a given question correctly, the information generated by each inspired-cognitive task is encoded via Transformers and fused through a modality weighting mechanism, which balances the information from the different sources. Exhaustive evaluation demonstrates the effectiveness of our approach, which yields a new state-of-the-art on two challenging video question answering datasets: KnowIT VQA and TVQA+.

For agents

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

get_harvested_code_for_paper("2007.08751")
get_code_for_paper("2007.08751")
have("2007.08751")

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

curl https://syntology.ai/api/ran/2007.08751.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.08751.svg)](https://syntology.ai/paper/2007.08751)

Connect an agent — have() is free.