SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1812.01880 · 2018

Learning to Compose Dynamic Tree Structures for Visual Contexts

arXiv · PDF · Open in the Atlas

Code that ran

10 of 14 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, 1 violated it and 4 have no contract verdict; 5 failures of our instrument; 4 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/1812.01880.json · badge: /api/ran/1812.01880.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
KaihuaTang/VCTree-Visual-Question-Answering canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
KaihuaTang/VCTree-Scene-Graph-Generation community (classifier label)
classifier basis not recorded on the link
8 ran · 4 no recorded run
FunctionStatusWhere it lives
apply_attention Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
KaihuaTang/VCTree-Visual-Question-Answering/model.py
code served (permissive licence) · get_code("fb40006c9cd123ac")
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, 12) 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 input tensor: (batch_size=2, channels=4, spatial_dims=3x5)
input = torch.randn(2, 4, 3, 5)

# Create attention tensor: (batch_size=2, glimpses=3, spatial_dims=3x5)
attention = torch.randn(2, 3, 3, 5)

# use_softmax parameter
use_softmax = True

ARGS = {
    "input": input,
    "attention": attention,
    "use_softmax": use_softmax
}
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, 12) — shape (2, 12) 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
arTree_to_biTree Ran
ran on a synthesized input
KaihuaTang/VCTree-Scene-Graph-Generation/lib/tree_lstm/graph_to_tree.py
code served (permissive licence) · get_code("bacd2c4e4708f3a1")
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 BiNode. 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 torch.nn as nn

class ArNode:
    def __init__(self, value=None):
        self.value = value
        self.children = []
    
    def add_child(self, child):
        self.children.append(child)
    
    def get_child_num(self):
        return len(self.children)
    
    def generate_bi_tree(self):
        return BiNode(self.value)


class BiNode:
    def __init__(self, value=None):
        self.value = value
        self.left_child = None
        self.right_child = None
    
    def add_left_child(self, child):
        self.left_child = child
    
    def add_right_child(self, child):
        self.right_child = child


class ArTree:
    def __init__(self, root):
        self.root = root
    
    def generate_bi_tree(self):
        return BiNode(self.root.value if hasattr(self.root, 'value') else None)
    
    def get_child_num(self):
        if hasattr(self.root, 'get_child_num'):
            return self.root.get_child_num()
        return 0
    
    @property
    def children(self):
        if hasattr(self.root, 'children'):
            return self.root.children
        return []


# Construct a simple n-ary tree
root_ar_node = ArNode(value=1)
child1 = ArNode(value=2)
child2 = ArNode(value=3)
child3 = ArNode(value=4)
root_ar_node.add_child(child1)
root_ar_node.add_child(child2)
root_ar_node.add_child(child3)

grandchild1 = ArNode(value=5)
child1.add_child(grandchild1)

ar_tree = ArTree(root_ar_node)

ARGS = {
    "arTree": ar_tree
}
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 BiNode — a BiNode output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a BiNode 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
filter_dets Ran
ran on a synthesized input
KaihuaTang/VCTree-Scene-Graph-Generation/lib/surgery.py
code served (permissive licence) · get_code("23725d7c2832de1b")
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 torch

torch.manual_seed(0)

# Define tensor arguments
num_box = 2
num_rel = 3
num_predicates = 4

# boxes: [num_box, 4]
boxes = torch.tensor([
    [10.0, 20.0, 30.0, 40.0],
    [50.0, 60.0, 70.0, 80.0]
], dtype=torch.float32)

# obj_scores: [num_box]
obj_scores = torch.tensor([0.9, 0.8], dtype=torch.float32)

# obj_classes: [num_box]
obj_classes = torch.tensor([1, 2], dtype=torch.long)

# rel_inds: [num_rel, 2] consisting of (box_ind0, box_ind1)
rel_inds = torch.tensor([
    [0, 1],
    [1, 0],
    [0, 0]
], dtype=torch.long)

# pred_scores: [num_rel, num_predicates]
pred_scores = torch.tensor([
    [0.1, 0.5, 0.3, 0.1],
    [0.2, 0.4, 0.2, 0.2],
    [0.3, 0.3, 0.2, 0.2]
], dtype=torch.float32)

# gt_boxes: ground truth boxes (can be numpy array or tensor)
gt_boxes = [[15.0, 25.0, 35.0, 45.0], [55.0, 65.0, 75.0, 85.0]]

# gt_classes: ground truth classes
gt_classes = [1, 2]

# gt_rels: ground truth relations
gt_rels = [[0, 1]]

ARGS = {
    'boxes': boxes,
    'obj_scores': obj_scores,
    'obj_classes': obj_classes,
    'rel_inds': rel_inds,
    'pred_scores': pred_scores,
    'gt_boxes': gt_boxes,
    'gt_classes': gt_classes,
    'gt_rels': gt_rels
}
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 — 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
get_ranking Ran
ran; our contract draft was wrong, not the code
KaihuaTang/VCTree-Scene-Graph-Generation/lib/pytorch_misc.py
pointer only (licence: MIT); the stored copy get_code serves, from declare-lab/lg-vqa, is pointer only (licence: NONE) · get_code("a1e587129dd2dfaf")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from declare-lab/lg-vqa, and GitHub detected no licence in that repository when it was checked. 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.

optimistic_restore Ran
ran on a synthesized input
KaihuaTang/VCTree-Scene-Graph-Generation/lib/pytorch_misc.py
code served (permissive licence) · get_code("6d0daf87a6ec7e39")
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 bool, 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 as nn

torch.manual_seed(0)

# Create a simple neural network
class SimpleNetwork(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

# Create network instance
network = SimpleNetwork()

# Create a state dict that will be partially compatible
# Include some matching keys, one unexpected key, and one missing key
state_dict = {
    'linear1.weight': torch.randn(3, 4),  # matches network's linear1.weight size
    'linear1.bias': torch.randn(3),       # matches network's linear1.bias size
    'linear2.weight': torch.randn(2, 3),  # matches network's linear2.weight size
    'linear2.bias': torch.randn(2),       # matches network's linear2.bias size
    'unexpected_key': torch.randn(5, 5),  # unexpected key not in network
}

# The network has: linear1.weight, linear1.bias, linear2.weight, linear2.bias
# The state_dict has all of those plus 'unexpected_key'
# This means linear2.weight and linear2.bias will be missing from the "missing" set
# and 'unexpected_key' will trigger the mismatch

ARGS = {
    'network': network,
    'state_dict': state_dict
}
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 bool, 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
pairwise Ran
ran; our contract draft was wrong, not the code
KaihuaTang/VCTree-Scene-Graph-Generation/lib/pytorch_misc.py
code served (permissive licence) · get_code("fef49730c9e9a2b8")
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 zip. 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
from itertools import tee

torch.manual_seed(0)

# The pairwise function takes an iterable and returns consecutive pairs
# It uses: tee() to create independent iterators, next() to advance one, and zip() to pair them
# The iterable parameter just needs to be any iterable object

# Create a simple iterable - a list of distinct values
iterable_input = [1, 2, 3, 4, 5]

ARGS = {
    "iterable": iterable_input
}
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 zip — a zip output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a zip 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
path Ran
ran; our contract draft was wrong, not the code
KaihuaTang/VCTree-Scene-Graph-Generation/config.py
code served (permissive licence) · get_code("5d64438de6614360")
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 os
import tempfile

# Create a temporary directory to simulate the module structure
temp_dir = tempfile.mkdtemp()
ROOT_PATH = temp_dir
DATA_PATH = os.path.join(ROOT_PATH, 'data')

# Ensure the data directory exists
os.makedirs(DATA_PATH, exist_ok=True)

# The function takes one parameter: fn (a filename string)
ARGS = {
    'fn': 'test_file.txt'
}
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
reporthook Ran
ran; our contract draft was wrong, not the code
KaihuaTang/VCTree-Scene-Graph-Generation/lib/word_vectors.py
code served (permissive licence) · get_code("b0ccc0da43f7ec20")
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
from tqdm import tqdm

# Create a tqdm instance with update and total attributes
t = tqdm(total=0)

# Define ARGS with the required parameter
ARGS = {
    "t": t
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, tqdm — 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
stanford_path Ran
ran on a synthesized input
KaihuaTang/VCTree-Scene-Graph-Generation/config.py
code served (permissive licence) · get_code("081b0b43ff7d9810")
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 os
import tempfile

# Create a temporary directory structure to simulate the expected paths
temp_dir = tempfile.mkdtemp()
data_dir = os.path.join(temp_dir, 'data')
stanford_dir = os.path.join(data_dir, 'stanford_filtered')
os.makedirs(stanford_dir, exist_ok=True)

# Mock the ROOT_PATH and DATA_PATH by patching the module
ROOT_PATH = temp_dir
DATA_PATH = os.path.join(ROOT_PATH, 'data')

ARGS = {
    'fn': 'example_file.txt'
}
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
tile_2d_over_nd Ran
ran, violated the contract we drafted
KaihuaTang/VCTree-Visual-Question-Answering/model.py
code served (permissive licence) · get_code("b3299f7858d7d2f2")
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, 4, 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)

# feature_vector: a 2D tensor of shape (batch_size, num_features)
# feature_map: a tensor with shape (batch_size, num_features, *spatial_sizes)
# The function will repeat feature_vector across all spatial positions

feature_vector = torch.randn(2, 4)  # batch_size=2, num_features=4

# feature_map should have batch_size=2, num_features=4, and some spatial dimensions
# Let's use spatial dimensions of 3x5 to be distinct
feature_map = torch.randn(2, 4, 3, 5)

ARGS = {
    "feature_vector": feature_vector,
    "feature_map": feature_map,
}
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, 4, 3, 5) — shape (2, 4, 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
arbitraryForest_to_biForest Not yet run
unverified (harvested, no recorded run)
KaihuaTang/VCTree-Scene-Graph-Generation/lib/tree_lstm/graph_to_tree.py
code served (permissive licence) · get_code("cd49722cb881389f")
How it ran

no invocation recorded.

gather_res Not yet run
unverified (harvested, no recorded run)
KaihuaTang/VCTree-Scene-Graph-Generation/lib/object_detector.py
code served (permissive licence) · get_code("013f03d1624b019d")
How it ran

no invocation recorded.

load_word_vectors Not yet run
unverified (harvested, no recorded run)
KaihuaTang/VCTree-Scene-Graph-Generation/lib/word_vectors.py
code served (permissive licence) · get_code("da7576453008a270")
How it ran

no invocation recorded.

vgg_fc Not yet run
unverified (harvested, no recorded run)
KaihuaTang/VCTree-Scene-Graph-Generation/lib/resnet.py
code served (permissive licence) · get_code("16d7f949c9b0c390")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 9 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 8 later) and 5 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
pairwise 9 other papers
9 later
1903.03326 (2019-03)
Knowledge-Embedded Routing Network for Scene Graph Generation
2003.12962 (2020-03)
GPS-Net: Graph Property Sensing Network for Scene Graph Generation
2006.09623 (2020-06)
Learning Visual Commonsense for Robust Scene Graph Generation
and 6 more not named here
get_ranking 5 other papers
5 later
1903.03326 (2019-03)
Knowledge-Embedded Routing Network for Scene Graph Generation
2003.12962 (2020-03)
GPS-Net: Graph Property Sensing Network for Scene Graph Generation
2006.09623 (2020-06)
Learning Visual Commonsense for Robust Scene Graph Generation
and 2 more not named here
reporthook 5 other papers
2 dated earlier, 3 later · 1 only as an assembly brick
1606.05250 (2016-06)
SQuAD: 100,000+ Questions for Machine Comprehension of Text
1611.01603 (2016-11)
Bidirectional Attention Flow for Machine Comprehension
2107.12309 (2021-07)
Spatial-Temporal Transformer for Dynamic Scene Graph Generation
and 2 more not named here
optimistic_restore 4 other papers
4 later
1903.03326 (2019-03)
Knowledge-Embedded Routing Network for Scene Graph Generation
2003.12962 (2020-03)
GPS-Net: Graph Property Sensing Network for Scene Graph Generation
2006.09623 (2020-06)
Learning Visual Commonsense for Robust Scene Graph Generation
and 1 more not named here
gather_res 3 other papers
3 later
1903.03326 (2019-03)
Knowledge-Embedded Routing Network for Scene Graph Generation
2006.09623 (2020-06)
Learning Visual Commonsense for Robust Scene Graph Generation
2403.12033 (2024-03)
HiKER-SGG: Hierarchical Knowledge Enhanced Robust Scene Graph Generation
vgg_fc 3 other papers
3 later
1903.03326 (2019-03)
Knowledge-Embedded Routing Network for Scene Graph Generation
2006.09623 (2020-06)
Learning Visual Commonsense for Robust Scene Graph Generation
2403.12033 (2024-03)
HiKER-SGG: Hierarchical Knowledge Enhanced Robust Scene Graph Generation
path 2 other papers
2 later
1903.03326 (2019-03)
Knowledge-Embedded Routing Network for Scene Graph Generation
2006.09623 (2020-06)
Learning Visual Commonsense for Robust Scene Graph Generation
stanford_path 1 other paper
1 later
1903.03326 (2019-03)
Knowledge-Embedded Routing Network for Scene Graph Generation
tile_2d_over_nd 1 other paper
1 dated earlier
1805.07932 (2018-05)
Bilinear Attention Networks

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 to compose dynamic tree structures that place the objects in an image into a visual context, helping visual reasoning tasks such as scene graph generation and visual Q&A. Our visual context tree model, dubbed VCTree, has two key advantages over existing structured object representations including chains and fully-connected graphs: 1) The efficient and expressive binary tree encodes the inherent parallel/hierarchical relationships among objects, e.g., "clothes" and "pants" are usually co-occur and belong to "person"; 2) the dynamic structure varies from image to image and task to task, allowing more content-/task-specific message passing among objects. To construct a VCTree, we design a score function that calculates the task-dependent validity between each object pair, and the tree is the binary version of the maximum spanning tree from the score matrix. Then, visual contexts are encoded by bidirectional TreeLSTM and decoded by task-specific models. We develop a hybrid learning procedure which integrates end-task supervised learning and the tree structure reinforcement learning, where the former's evaluation result serves as a self-critic for the latter's structure exploration. Experimental results on two benchmarks, which require reasoning over contexts: Visual Genome for scene graph generation and VQA2.0 for visual Q&A, show that VCTree outperforms state-of-the-art results while discovering interpretable visual context structures.

For agents

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

get_harvested_code_for_paper("1812.01880")
get_code_for_paper("1812.01880")
have("1812.01880")

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

curl https://syntology.ai/api/ran/1812.01880.json

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

[![Syntology run record](https://syntology.ai/api/ran/1812.01880.svg)](https://syntology.ai/paper/1812.01880)

Connect an agent — have() is free.