SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.06147 · 2021

NAAQA: A Neural Architecture for Acoustic Question Answering

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/2106.06147.json · badge: /api/ran/2106.06147.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
necotis/naaqa-acoustic-question-answering community (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
FunctionStatusWhere it lives
append_spatial_location Ran
ran on a synthesized input
necotis/naaqa-acoustic-question-answering/models/utils.py
code served (permissive licence) · get_code("f6f1f83d29ecd9f5")
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 4, dtype family float, shape (2, 6, 3, 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 torch

torch.manual_seed(0)

# Create a small feature tensor with shape (batch_size, channels, height, width)
# Using distinct sizes: batch=2, channels=4, height=3, width=5
features = torch.randn(2, 4, 3, 5)

ARGS = {
    "features": features,
    "start": -1,
    "end": 1,
    "axis": 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 Tensor, ndim 4, dtype family float, shape (2, 6, 3, 5) — shape (2, 6, 3, 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
calc_f1_score Ran
ran on a synthesized input
necotis/naaqa-acoustic-question-answering/models/metrics.py
code served (permissive licence) · get_code("a687fa3dac8a2e5c")
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 float64, ndim 0, 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 numpy as np

torch.manual_seed(0)

# Create predictions and ground truth labels
# Using small integer labels for multi-class classification
preds = torch.tensor([0, 1, 0, 1, 2, 2], dtype=torch.long)
ground_truths = torch.tensor([0, 1, 1, 1, 2, 1], dtype=torch.long)

# Define labels explicitly (optional, can be None to auto-detect)
labels = np.array([0, 1, 2])

ARGS = {
    'preds': preds,
    'ground_truths': ground_truths,
    'labels': labels
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
one_game_inference_by_id Ran
ran on a synthesized input
necotis/naaqa-acoustic-question-answering/runner.py
code served (permissive licence) · get_code("3fb6d124f1f1b4dc")
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.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Stand-in tokenizer with decode_answer method
class StandInTokenizer:
    def decode_answer(self, pred):
        return f"answer_{pred}"

# Stand-in model that takes question, seq_length, image and returns logits and softmax
class StandInModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 10)  # 10 possible answers
    
    def forward(self, question, seq_length, image, pack_sequence=False):
        # question shape: (batch, seq_len, 4)
        # image shape: (batch, 4)
        # Return: (logits, softmax_output)
        batch_size = question.shape[0]
        logits = self.linear(image)  # (batch, 10)
        softmax_output = torch.softmax(logits, dim=-1)  # (batch, 10)
        return logits, softmax_output

# Stand-in dataset
class StandInDataset:
    def __init__(self):
        self.tokenizer = StandInTokenizer()
        self.data = [
            {
                'image': torch.randn(4),
                'question': torch.randn(5, 4),
                'seq_length': torch.tensor(5)
            }
        ]
    
    def __getitem__(self, idx):
        return self.data[idx % len(self.data)]
    
    def __len__(self):
        return len(self.data)

# Collate function
def collate_fn(batch):
    images = torch.stack([item['image'] for item in batch])
    questions = torch.stack([item['question'] for item in batch])
    seq_lengths = torch.stack([item['seq_length'] for item in batch])
    return {
        'image': images,
        'question': questions,
        'seq_length': seq_lengths
    }

# Stand-in dataloader
class StandInDataLoader:
    def __init__(self):
        self.dataset = StandInDataset()
        self.collate_fn = collate_fn

device = torch.device('cpu')
model = StandInModel()
dataloader = StandInDataLoader()
game_id = 0
nb_top_pred = 10

ARGS = {
    'device': device,
    'model': model,
    'dataloader': dataloader,
    'game_id': game_id,
    'nb_top_pred': nb_top_pred
}
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 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
pad2d_and_cat_tensors Ran
ran on a synthesized input
necotis/naaqa-acoustic-question-answering/models/utils.py
code served (permissive licence) · get_code("25bcc2a8366b319c")
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 4, dtype family float, shape (2, 12, 6, 7) 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 torch.nn.functional as F

torch.manual_seed(0)

# Create a list of tensors with different spatial dimensions
# Shape: [Batch_size, Channels, Height, Width]
tensor1 = torch.randn(2, 3, 4, 5)  # batch=2, channels=3, height=4, width=5
tensor2 = torch.randn(2, 4, 6, 7)  # batch=2, channels=4, height=6, width=7
tensor3 = torch.randn(2, 5, 3, 4)  # batch=2, channels=5, height=3, width=4

tensors = [tensor1, tensor2, tensor3]

ARGS = {
    'tensors': tensors,
    'pad_mode': 'end'
}
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 4, dtype family float, shape (2, 12, 6, 7) — shape (2, 12, 6, 7) 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
reproducible_uniform_ Ran
ran on a synthesized input
necotis/naaqa-acoustic-question-answering/models/utils.py
code served (permissive licence) · get_code("91c937aeca0ab343")
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 (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 numpy as np

torch.manual_seed(0)
np.random.seed(0)

# Create a small tensor to be filled with uniform values
tensor = torch.empty(2, 4, dtype=torch.float32)

# Define bounds for uniform distribution
lower_bound = 0.0
higher_bound = 1.0

ARGS = {
    "tensor": tensor,
    "lower_bound": lower_bound,
    "higher_bound": higher_bound,
}
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 (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
summary Ran
ran on a synthesized input
necotis/naaqa-acoustic-question-answering/models/torchsummary.py
code served (permissive licence) · get_code("07b0f0bc2092e0fb")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

# Create a simple model that accepts the expected input format
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(3, 16, kernel_size=3)
        self.fc1 = nn.Linear(16 * 2 * 2, 10)
    
    def forward(self, x, pack_sequence=False):
        x = self.conv1(x)
        x = x.view(x.size(0), -1)
        x = self.fc1(x)
        return x

model = SimpleModel()

# input_infos: list of tuples (input_size, input_type)
# Each input_size is a tuple of dimensions (excluding batch)
# The function creates tensors with shape (2, *input_size)
input_infos = [
    ((3, 4, 4), torch.float32),
]

ARGS = {
    'model': model,
    'input_infos': input_infos,
    'batch_size': -1,
    'device': 'cpu',
    'print_output': True,
}
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

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 6 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 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.

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

The goal of the Acoustic Question Answering (AQA) task is to answer a free-form text question about the content of an acoustic scene. It was inspired by the Visual Question Answering (VQA) task. In this paper, based on the previously introduced CLEAR dataset, we propose a new benchmark for AQA, namely CLEAR2, that emphasizes the specific challenges of acoustic inputs. These include handling of variable duration scenes, and scenes built with elementary sounds that differ between training and test set. We also introduce NAAQA, a neural architecture that leverages specific properties of acoustic inputs. The use of 1D convolutions in time and frequency to process 2D spectro-temporal representations of acoustic content shows promising results and enables reductions in model complexity. We show that time coordinate maps augment temporal localization capabilities which enhance performance of the network by ~17 percentage points. On the other hand, frequency coordinate maps have little influence on this task. NAAQA achieves 79.5% of accuracy on the AQA task with ~4 times fewer parameters than the previously explored VQA model. We evaluate the perfomance of NAAQA on an independent data set reconstructed from DAQA. We also test the addition of a MALiMo module in our model on both CLEAR2 and DAQA. We provide a detailed analysis of the results for the different question types. We release the code to produce CLEAR2 as well as NAAQA to foster research in this newly emerging machine learning task.

For agents

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

get_harvested_code_for_paper("2106.06147")
get_code_for_paper("2106.06147")
have("2106.06147")

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

curl https://syntology.ai/api/ran/2106.06147.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.06147.svg)](https://syntology.ai/paper/2106.06147)

Connect an agent — have() is free.