SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.16010 · 2020

Loss re-scaling VQA: Revisiting the LanguagePrior Problem from a Class-imbalance View

arXiv · PDF · Open in the Atlas

Code that ran

5 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 3 have no contract verdict; 2 failures of our instrument; 1 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/2010.16010.json · badge: /api/ran/2010.16010.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
guoyang9/class-imbalance-VQA canonical (classifier label)
classifier basis: census
5 ran · 1 no recorded run
FunctionStatusWhere it lives
elementwise_logsumexp Ran
ran; our contract draft was wrong, not the code
guoyang9/class-imbalance-VQA/utils/losses.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from chrisc36/bottom-up-attention-vqa, is pointer only (licence: GPL-3.0) · get_code("4bd23f3562cfbd1c")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from chrisc36/bottom-up-attention-vqa, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

compute_score_with_logits Ran
ran on a synthesized input
guoyang9/class-imbalance-VQA/train-css.py
code served (permissive licence) · get_code("5250520f3498ea70")
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, 1) 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 small tensors for the function
# logits: shape (batch_size, num_classes) - used with max(dim=1, keepdim=True)
# labels: shape (batch_size, num_classes) - used with gather(dim=1, index=log_index)

batch_size = 2
num_classes = 4

logits = torch.randn(batch_size, num_classes)
labels = torch.randint(0, 2, (batch_size, num_classes)).float()

ARGS = {
    "logits": logits,
    "labels": labels
}
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 2, dtype family float, shape (2, 1) — shape (2, 1) 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
convert_sigmoid_logits_to_binary_logprobs Ran
ran; our contract draft was wrong, not the code
guoyang9/class-imbalance-VQA/utils/losses.py
code served (permissive licence) · get_code("c0ea72864a818321")
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 3, 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.functional as F

torch.manual_seed(0)

# The function takes logits as input and applies sigmoid-related operations
# logits should be a tensor of any shape
logits = torch.randn(2, 4)

ARGS = {
    "logits": logits
}
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 3, 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
cross_entropy_loss Ran
ran on a synthesized input
guoyang9/class-imbalance-VQA/utils/losses.py
code served (permissive licence) · get_code("bd13bf1e7cf15819")
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 torch.nn.functional as F

torch.manual_seed(0)

# Define the arguments for cross_entropy_loss
logits = torch.randn(2, 4)  # batch_size=2, num_classes=4
labels = torch.randn(2, 4)  # batch_size=2, num_classes=4 (soft labels for cross entropy)
miu = torch.randn(2, 4)     # batch_size=2, num_classes=4
mask = torch.randint(0, 2, (2, 4)).float()  # binary mask with same shape as miu

ARGS = {
    'logits': logits,
    'labels': labels,
    'miu': miu,
    'mask': mask
}
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
mask_softmax Ran
ran on a synthesized input
guoyang9/class-imbalance-VQA/modules/attention.py
code served (permissive licence) · get_code("b82f764aadc77b83")
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 3, dtype family float, shape (2, 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

torch.manual_seed(0)

# Construct x: shape (batch=2, seq_len=3, dim=4)
# This will be exponentiated and summed, so needs proper shape
x = torch.randn(2, 3, 4)

# Construct mask: shape (batch=2, seq_len=3)
# The function unsqueezes it at dim=2, so we need 2D input
mask = torch.ones(2, 3)

ARGS = {
    'x': x,
    'mask': mask
}
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 3, dtype family float, shape (2, 3, 4) — shape (2, 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
saved_for_eval Not yet run
unverified (harvested, no recorded run)
guoyang9/class-imbalance-VQA/train-css.py
code served (permissive licence) · get_code("1c77d8d458f0dcaa")
How it ran

no invocation recorded.

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, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 2 later) and 4 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
convert_sigmoid_logits_to_binary_logprobs 4 other papers
3 dated earlier, 1 later
1707.07998 (2017-07)
Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering
1909.03683 (2019-09)
Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases
2003.06576 (2020-03)
Counterfactual Samples Synthesizing for Robust Visual Question Answering
and 1 more not named here
elementwise_logsumexp 4 other papers
3 dated earlier, 1 later
1707.07998 (2017-07)
Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering
1909.03683 (2019-09)
Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases
2003.06576 (2020-03)
Counterfactual Samples Synthesizing for Robust Visual Question Answering
and 1 more not named here

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 studies have pointed out that many well-developed Visual Question Answering (VQA) models are heavily affected by the language prior problem, which refers to making predictions based on the co-occurrence pattern between textual questions and answers instead of reasoning visual contents. To tackle it, most existing methods focus on enhancing visual feature learning to reduce this superficial textual shortcut influence on VQA model decisions. However, limited effort has been devoted to providing an explicit interpretation for its inherent cause. It thus lacks a good guidance for the research community to move forward in a purposeful way, resulting in model construction perplexity in overcoming this non-trivial problem. In this paper, we propose to interpret the language prior problem in VQA from a class-imbalance view. Concretely, we design a novel interpretation scheme whereby the loss of mis-predicted frequent and sparse answers of the same question type is distinctly exhibited during the late training phase. It explicitly reveals why the VQA model tends to produce a frequent yet obviously wrong answer, to a given question whose right answer is sparse in the training set. Based upon this observation, we further develop a novel loss re-scaling approach to assign different weights to each answer based on the training data statistics for computing the final loss. We apply our approach into three baselines and the experimental results on two VQA-CP benchmark datasets evidently demonstrate its effectiveness. In addition, we also justify the validity of the class imbalance interpretation scheme on other computer vision tasks, such as face recognition and image classification.

For agents

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

get_harvested_code_for_paper("2010.16010")
get_code_for_paper("2010.16010")
have("2010.16010")

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

curl https://syntology.ai/api/ran/2010.16010.json

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

[![Syntology run record](https://syntology.ai/api/ran/2010.16010.svg)](https://syntology.ai/paper/2010.16010)

Connect an agent — have() is free.