SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1812.08658 · 2018

nocaps: novel object captioning at scale

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 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 2 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/1812.08658.json · badge: /api/ran/1812.08658.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
nocaps-org/updown-baseline canonical (classifier label)
classifier basis: census
2 ran · 0 no recorded run
FunctionStatusWhere it lives
select_best_beam Ran
ran on a synthesized input
nocaps-org/updown-baseline/updown/utils/decoding.py
code served (permissive licence) · get_code("f6d8d3b14e31142a")
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 int, 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

torch.manual_seed(0)

# Create sample tensors matching the function signature
batch_size = 2
num_states = 3
max_decoding_steps = 4
beam_size = 5

# beams: shape (batch_size, num_states, max_decoding_steps)
beams = torch.randint(0, 100, (batch_size, num_states, max_decoding_steps))

# beam_log_probabilities: shape (batch_size, num_states, beam_size)
beam_log_probabilities = torch.randn(batch_size, num_states, beam_size)

ARGS = {
    "beams": beams,
    "beam_log_probabilities": beam_log_probabilities,
}
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 int, 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
select_best_beam_with_constraints Ran
ran on a synthesized input
nocaps-org/updown-baseline/updown/utils/decoding.py
code served (permissive licence) · get_code("35d87aa8053e1da3")
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 int, shape (2, 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)

# Parameters for select_best_beam_with_constraints
batch_size = 2
num_states = 8  # 2^3 for 3 constraints max
beam_size = 4
max_decoding_steps = 5

# Create beams tensor: (batch_size, num_states, beam_size, max_decoding_steps)
beams = torch.randint(0, 100, (batch_size, num_states, beam_size, max_decoding_steps))

# Create beam_log_probabilities tensor: (batch_size, num_states, beam_size)
# Log probabilities should be negative, descending within each state
beam_log_probabilities = torch.randn(batch_size, num_states, beam_size)
beam_log_probabilities = torch.sort(beam_log_probabilities, dim=2, descending=True)[0]

# Create given_constraints tensor: (batch_size, )
# Values should be between 1 and 3 (since max 3 constraints supported)
given_constraints = torch.tensor([2, 3])

# min_constraints_to_satisfy parameter
min_constraints_to_satisfy = 2

ARGS = {
    "beams": beams,
    "beam_log_probabilities": beam_log_probabilities,
    "given_constraints": given_constraints,
    "min_constraints_to_satisfy": min_constraints_to_satisfy,
}
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 int, shape (2, 5) — shape (2, 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

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 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 2 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

Image captioning models have achieved impressive results on datasets containing limited visual concepts and large amounts of paired image-caption training data. However, if these models are to ever function in the wild, a much larger variety of visual concepts must be learned, ideally from less supervision. To encourage the development of image captioning models that can learn visual concepts from alternative data sources, such as object detection datasets, we present the first large-scale benchmark for this task. Dubbed 'nocaps', for novel object captioning at scale, our benchmark consists of 166,100 human-generated captions describing 15,100 images from the OpenImages validation and test sets. The associated training data consists of COCO image-caption pairs, plus OpenImages image-level labels and object bounding boxes. Since OpenImages contains many more classes than COCO, nearly 400 object classes seen in test images have no or very few associated training captions (hence, nocaps). We extend existing novel object captioning models to establish strong baselines for this benchmark and provide analysis to guide future work on this task.

For agents

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

get_harvested_code_for_paper("1812.08658")
get_code_for_paper("1812.08658")
have("1812.08658")

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

curl https://syntology.ai/api/ran/1812.08658.json

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

[![Syntology run record](https://syntology.ai/api/ran/1812.08658.svg)](https://syntology.ai/paper/1812.08658)

Connect an agent — have() is free.