SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.15871 · 2022

VLT: Vision-Language Transformer and Query Generation for Referring Segmentation

arXiv · PDF · Open in the Atlas

Code that ran

3 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; 0 failures of our instrument; 3 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/2210.15871.json · badge: /api/ran/2210.15871.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
henghuiding/Vision-Language-Transformer canonical (classifier label)
classifier basis not recorded on the link
3 ran · 3 no recorded run
FunctionStatusWhere it lives
get_random_data Ran
ran on a synthesized input
henghuiding/Vision-Language-Transformer/loader/loader.py
code served (permissive licence) · get_code("19e932448eddc911")
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. 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 numpy as np
import cv2
import os
import tempfile
from pathlib import Path

# Create a minimal stand-in for the embed object
class MinimalEmbed:
    def __call__(self, word):
        # Return an object with a .vector attribute
        class WordVector:
            def __init__(self):
                self.vector = np.random.randn(300)
        return WordVector()

# Create a minimal config object
class Config:
    def __init__(self):
        self.seg_gt_path = "/tmp/seg_gt"
        self.image_path = "/tmp/images"
        self.word_len = 20
        self.seg_out_stride = 4

# Create temporary directories and files for testing
temp_dir = tempfile.mkdtemp()
image_dir = os.path.join(temp_dir, "images")
seg_dir = os.path.join(temp_dir, "seg_gt")
os.makedirs(image_dir, exist_ok=True)
os.makedirs(seg_dir, exist_ok=True)

# Create a dummy image file
dummy_image = np.random.randint(0, 256, (100, 120, 3), dtype=np.uint8)
cv2.imwrite(os.path.join(image_dir, "test_img.jpg"), dummy_image)

# Create a dummy segmentation map
dummy_seg = np.random.randint(0, 256, (100, 120), dtype=np.uint8)
cv2.imwrite(os.path.join(seg_dir, "1.png"), dummy_seg)

# Build the ref dictionary
ref = {
    'segment_id': 1,
    'img_name': 'test_img.jpg',
    'sentences': [
        {'sent': 'the quick brown fox'},
        {'sent': 'jumps over the lazy dog'}
    ],
    'sentences_num': 2
}

# Build config
config = Config()
config.seg_gt_path = seg_dir
config.image_path = image_dir

# Build embed
embed = MinimalEmbed()

# Define input shape
input_shape = (128, 128)

# Define ARGS
ARGS = {
    'ref': ref,
    'input_shape': input_shape,
    'embed': embed,
    'config': config,
    'train_mode': True,
    'max_boxes': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple 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
lr_step_decay Ran
ran on a synthesized input
henghuiding/Vision-Language-Transformer/callbacks/learning_scheduler.py
code served (permissive licence) · get_code("6f691e355615b0f3")
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 function. 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)

# Analyzing the function:
# - lr_start: a float representing initial learning rate
# - steps: a list of integers representing epoch thresholds for decay
# The function returns a closure that takes an epoch and returns a learning rate

ARGS = {
    "lr_start": 0.001,
    "steps": [30, 40]
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
qlist_to_vec Ran
ran on a synthesized input
henghuiding/Vision-Language-Transformer/loader/loader.py
code served (permissive licence) · get_code("00b7240d09897b4f")
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 2, dtype family float, shape (5, 300) 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

class MockEmbedding:
    """Mock embedding object that returns a vector for any word."""
    def __init__(self, emb_size=300):
        self.emb_size = emb_size
    
    def __call__(self, word):
        """Return a mock vector object for the given word."""
        vector = np.random.randn(self.emb_size).astype(float)
        return MockVector(vector)

class MockVector:
    """Mock vector object with a .vector attribute."""
    def __init__(self, vector):
        self.vector = vector

ARGS = {
    'max_length': 5,
    'q_list': 'hello world this is a test',
    'embed': MockEmbedding(emb_size=300),
    'emb_size': 300
}
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 2, dtype family float, shape (5, 300) — shape (5, 300) 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
expand_and_tile Not yet run
unverified (harvested, no recorded run)
henghuiding/Vision-Language-Transformer/model/utils.py
code served (permissive licence) · get_code("47337bf91f204990")
How it ran

no invocation recorded.

expand_and_tile_1 Not yet run
unverified (harvested, no recorded run)
henghuiding/Vision-Language-Transformer/model/utils.py
code served (permissive licence) · get_code("a6d7706f7a9a3eca")
How it ran

no invocation recorded.

normalize_by_dim Not yet run
unverified (harvested, no recorded run)
henghuiding/Vision-Language-Transformer/model/utils.py
code served (permissive licence) · get_code("364718e6449b4023")
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, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 0 later) and 3 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
expand_and_tile 1 other paper
1 dated earlier
2003.08813 (2020-03)
Multi-task Collaborative Network for Joint Referring Expression Comprehension and Segmentation
expand_and_tile_1 1 other paper
1 dated earlier
2003.08813 (2020-03)
Multi-task Collaborative Network for Joint Referring Expression Comprehension and Segmentation
normalize_by_dim 1 other paper
1 dated earlier
2003.08813 (2020-03)
Multi-task Collaborative Network for Joint Referring Expression Comprehension and Segmentation

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

We propose a Vision-Language Transformer (VLT) framework for referring segmentation to facilitate deep interactions among multi-modal information and enhance the holistic understanding to vision-language features. There are different ways to understand the dynamic emphasis of a language expression, especially when interacting with the image. However, the learned queries in existing transformer works are fixed after training, which cannot cope with the randomness and huge diversity of the language expressions. To address this issue, we propose a Query Generation Module, which dynamically produces multiple sets of input-specific queries to represent the diverse comprehensions of language expression. To find the best among these diverse comprehensions, so as to generate a better mask, we propose a Query Balance Module to selectively fuse the corresponding responses of the set of queries. Furthermore, to enhance the model's ability in dealing with diverse language expressions, we consider inter-sample learning to explicitly endow the model with knowledge of understanding different language expressions to the same object. We introduce masked contrastive learning to narrow down the features of different expressions for the same target object while distinguishing the features of different objects. The proposed approach is lightweight and achieves new state-of-the-art referring segmentation results consistently on five datasets.

For agents

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

get_harvested_code_for_paper("2210.15871")
get_code_for_paper("2210.15871")
have("2210.15871")

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

curl https://syntology.ai/api/ran/2210.15871.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.15871.svg)](https://syntology.ai/paper/2210.15871)

Connect an agent — have() is free.