SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.07097 · ECCV · 2022

ReAct: Temporal Action Detection with Relational Queries

Dacheng Tao, Lin Ma, Jia Li, Yujie Zhong, Jing Zhang, Dingfeng Shi, Qiong Cao

arXiv · PDF · Open in the Atlas

Code that ran

7 of 12 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 5 have no contract verdict; 2 failures of our instrument; 5 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/2207.07097.json · badge: /api/ran/2207.07097.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
sssste/React canonical (classifier label)
classifier basis not recorded on the link
7 ran · 5 no recorded run
FunctionStatusWhere it lives
all_gather Ran
ran; our contract draft was wrong, not the code
sssste/React/React/utill/misc.py
pointer only (licence: MIT); the stored copy get_code serves, from kellyiss/situformer, is pointer only (licence: NONE) · get_code("aa6da43d05cfaf65")
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 kellyiss/situformer, 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.

collate_fn Ran
ran on a synthesized input
sssste/React/React/utill/misc.py
code served (permissive licence) · get_code("8afa631a846ef588")
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
import numpy as np

torch.manual_seed(0)

# Build a batch for collate_fn
# The batch parameter should be a list of tuples where each tuple contains:
# - batch[0]: first element (will be combined with batch[1])
# - batch[1]: second element (will be combined with batch[0])
# - batch[2]: third element (will be combined with batch[3])
# - batch[3]: fourth element (will be combined with batch[2])

# Looking at nested_tensor_from_tensor_list, it expects a list of 2D tensors
# when tensor_list[0].ndim == 2

# Create small sample tensors
tensor1_a = torch.randn(3, 4)  # 3 samples, 4 features
tensor1_b = torch.randn(2, 4)  # 2 samples, 4 features
tensor1_c = torch.randn(5, 4)  # 5 samples, 4 features
tensor1_d = torch.randn(4, 4)  # 4 samples, 4 features

# Create labels/targets
label1_a = [0, 1, 2]
label1_b = [1, 0]
label1_c = [0, 2, 1, 0, 2]
label1_d = [1, 1, 0, 1]

# Create batch items
batch_item1 = (tensor1_a, tensor1_c, label1_a, label1_c)
batch_item2 = (tensor1_b, tensor1_d, label1_b, label1_d)

batch = [batch_item1, batch_item2]

ARGS = {
    'batch': batch
}
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 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
find_match_gt Ran
ran on a synthesized input
sssste/React/React/utill/temporal_box_producess.py
code served (permissive licence) · get_code("5334c4d83d9da129")
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 1, dtype family float, shape (2,) 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)

# Analyzing the function:
# - seg_instance: tensor that gets unsqueezed to shape (N, 1, 3) where dim 2 has indices [0, 1, 2]
# - gt: tensor that gets unsqueezed to shape (1, M, 3) where dim 2 has indices [0, 1, 2]
# - The function performs IoU computation on interval ranges (dim 0 and 1 are bounds, dim 2 is class)
# - seg_instance needs shape (N, 3) before unsqueeze
# - gt needs shape (M, 3) before unsqueeze
# Where dim 2 contains: [start, end, class_id]

# Create seg_instance: shape (2, 3) 
# Each row is [start, end, class_id]
seg_instance = torch.tensor([
    [1.0, 5.0, 0.0],
    [2.0, 6.0, 1.0]
], dtype=torch.float32)

# Create gt (ground truth): shape (4, 3)
# Each row is [start, end, class_id]
gt = torch.tensor([
    [1.0, 4.0, 0.0],
    [2.0, 7.0, 1.0],
    [3.0, 8.0, 2.0],
    [0.5, 5.5, 0.0]
], dtype=torch.float32)

ARGS = {
    'seg_instance': seg_instance,
    'gt': gt
}
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 1, dtype family float, shape (2,) — shape (2,) 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
mid_dis2start_end Ran
ran on a synthesized input
sssste/React/React/utill/temporal_box_producess.py
code served (permissive licence) · get_code("533a2025d10b811b")
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, 2) 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)

# Test with torch.Tensor input
x_torch = torch.tensor([
    [10.0, 4.0],
    [20.0, 8.0]
])

ARGS = {
    "x": x_torch
}
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 int, shape (2, 2) — shape (2, 2) 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
reduce_dict Ran
ran; our contract draft was wrong, not the code
sssste/React/React/utill/misc.py
pointer only (licence: MIT); the stored copy get_code serves, from kellyiss/situformer, is pointer only (licence: NONE) · get_code("05992505df9ef0cd")
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 kellyiss/situformer, 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.

register_hooks Ran
ran on a synthesized input
sssste/React/React/utill/bad_grad_viz.py
code served (permissive licence) · get_code("55db86fcc85cb96d")
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 torch.nn as nn
from graphviz import Digraph

torch.manual_seed(0)

# Create a simple tensor that requires gradients
x = torch.randn(2, 4, requires_grad=True)

# Create a simple computation graph
y = x * 2
z = y.sum()

# z.grad_fn will be the root of the computation graph
var = z

ARGS = {
    "var": var
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
graphviz, 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
sigmoid_focal_loss Ran
ran on a synthesized input
sssste/React/React/model/React.py
code served (permissive licence) · get_code("2043ee96aadac29d")
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)

# Create small test tensors
# inputs: predictions (logits) - shape (batch_size, num_classes)
inputs = torch.randn(2, 4)

# targets: binary labels - same shape as inputs
targets = torch.randint(0, 2, (2, 4)).float()

# num_boxes: optional normalization factor
num_boxes = 2

# alpha: weighting factor for positive vs negative examples
alpha = 0.25

# gamma: focusing parameter
gamma = 2.0

ARGS = {
    'inputs': inputs,
    'targets': targets,
    'num_boxes': num_boxes,
    'alpha': alpha,
    'gamma': gamma,
}
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
embeddingto2d Not yet run
unverified (harvested, no recorded run)
sssste/React/React/utill/viz.py
code served (permissive licence) · get_code("958ce3808938e30f")
How it ran

no invocation recorded.

load_proposal_file Not yet run
unverified (harvested, no recorded run)
sssste/React/React/thumos_dataset.py
code served (permissive licence) · get_code("45fddd2562b43511")
How it ran

no invocation recorded.

start_end2mid_dis Not yet run
unverified (harvested, no recorded run)
sssste/React/React/utill/temporal_box_producess.py
code served (permissive licence) · get_code("6e92295b5b0a88ac")
How it ran

no invocation recorded.

vis_1d_box Not yet run
unverified (harvested, no recorded run)
sssste/React/React/utill/viz.py
code served (permissive licence) · get_code("f0b9c93da8e4c2bb")
How it ran

no invocation recorded.

vis_2d_scatter Not yet run
unverified (harvested, no recorded run)
sssste/React/React/utill/viz.py
code served (permissive licence) · get_code("ea5b8dccd6161c45")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 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 10 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
reduce_dict 33 other papers
10 dated earlier, 22 later, 1 same year, order not determinable
1811.07628 (2018-11)
ATOM: Accurate Tracking by Overlap Maximization
1811.10200 (2018-11)
IDD: A Dataset for Exploring Problems of Autonomous Navigation in Unconstrained Environments
2007.02419 (2020-07)
Attention-based Joint Detection of Object and Semantic Part
and 30 more not named here
all_gather 30 other papers
8 dated earlier, 21 later, 1 same year, order not determinable
1811.07628 (2018-11)
ATOM: Accurate Tracking by Overlap Maximization
1811.10200 (2018-11)
IDD: A Dataset for Exploring Problems of Autonomous Navigation in Unconstrained Environments
2007.02419 (2020-07)
Attention-based Joint Detection of Object and Semantic Part
and 27 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

This work aims at advancing temporal action detection (TAD) using an encoder-decoder framework with action queries, similar to DETR, which has shown great success in object detection. However, the framework suffers from several problems if directly applied to TAD: the insufficient exploration of inter-query relation in the decoder, the inadequate classification training due to a limited number of training samples, and the unreliable classification scores at inference. To this end, we first propose a relational attention mechanism in the decoder, which guides the attention among queries based on their relations. Moreover, we propose two losses to facilitate and stabilize the training of action classification. Lastly, we propose to predict the localization quality of each action query at inference in order to distinguish high-quality queries. The proposed method, named ReAct, achieves the state-of-the-art performance on THUMOS14, with much lower computational costs than previous methods. Besides, extensive ablation studies are conducted to verify the effectiveness of each proposed component. The code is available at https://github.com/sssste/React.

For agents

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

get_harvested_code_for_paper("2207.07097")
get_code_for_paper("2207.07097")
have("2207.07097")

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

curl https://syntology.ai/api/ran/2207.07097.json

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

[![Syntology run record](https://syntology.ai/api/ran/2207.07097.svg)](https://syntology.ai/paper/2207.07097)

Connect an agent — have() is free.