SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.16182 · CVPR · 2024

EgoExoLearn: A Dataset for Bridging Asynchronous Ego-and Exo-centric View of Procedural Activities in Real World

Yu Qiao, Yali Wang, Limin Wang, Guo Chen, Yifei Huang, Mingfang Zhang, Baoqi Pei, Jilan Xu, Lijin Yang, Lu Dong, Hongjie Zhang

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 6 failures of our instrument; 1 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/2403.16182.json · badge: /api/ran/2403.16182.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
OpenGVLab/EgoExoLearn canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
FunctionStatusWhere it lives
attentive_entropy Ran
ran; our contract draft was wrong, not the code
OpenGVLab/EgoExoLearn/action_anticipation_planning_benchmark/loss.py
code served (permissive licence) · get_code("5dafc748f41191f5")
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 as nn

torch.manual_seed(0)

# The function expects two logit tensors
# pred: logits for the main task (used to compute entropy)
# pred_domain: logits for domain prediction (used to compute entropy weights)
# Both are passed through softmax and logsoftmax, so they should be 2D tensors
# with shape (batch_size, num_classes)

batch_size = 2
num_classes = 4

pred = torch.randn(batch_size, num_classes)
pred_domain = torch.randn(batch_size, num_classes)

ARGS = {
    "pred": pred,
    "pred_domain": pred_domain,
}
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
attentive_entropy Ran
ran; our contract draft was wrong, not the code
OpenGVLab/EgoExoLearn/temporal_action_segmentation_benchmark/loss.py
code served (permissive licence) · get_code("829b3a3e9d894fa8")
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
import torch.nn as nn

torch.manual_seed(0)

# The function takes two logit tensors as input
# pred: logits for the main task (used in softmax/logsoftmax)
# pred_domain: logits for domain prediction (used in softmax/logsoftmax for entropy weighting)
# Both should have shape (batch_size, num_classes)

batch_size = 2
num_classes = 4

# Create small logit tensors
pred = torch.randn(batch_size, num_classes)
pred_domain = torch.randn(batch_size, num_classes)

ARGS = {
    "pred": pred,
    "pred_domain": pred_domain
}
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
binary_precision_recall_curve Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
OpenGVLab/EgoExoLearn/action_anticipation_planning_benchmark/anticipation_main.py
code served (permissive licence) · get_code("b25ea1a523927d43")
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

# Create small numpy arrays for testing
# y_score: prediction scores, shape (num_classes,)
# y_true: ground truth many-hot vector, shape (num_classes,)

num_classes = 5

# y_score: random scores between 0 and 1
y_score = np.array([0.9, 0.1, 0.7, 0.3, 0.5], dtype=np.float32)

# y_true: binary labels (0 or 1)
y_true = np.array([1, 0, 1, 0, 1], dtype=np.int32)

ARGS = {
    "y_score": y_score,
    "y_true": y_true,
}
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 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
cross_entropy_soft Ran
ran; our contract draft was wrong, not the code
OpenGVLab/EgoExoLearn/action_anticipation_planning_benchmark/loss.py
code served (permissive licence) · get_code("b1e8dce608f3f658")
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 as nn

torch.manual_seed(0)

# The function takes a single parameter 'pred' which is used as:
# - Input to softmax and logsoftmax (both expect dim=1)
# - The result is summed along dim=1
# So pred should be a 2D tensor with shape (batch_size, num_classes)

pred = torch.randn(2, 4)

ARGS = {
    "pred": pred
}
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
dis_MCD Ran
ran; our contract draft was wrong, not the code
OpenGVLab/EgoExoLearn/action_anticipation_planning_benchmark/loss.py
code served (permissive licence) · get_code("0c3c2aa9cf2ac2f0")
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 two output tensors with shape (batch_size=2, num_classes=4)
# These represent logits from a model
out1 = torch.randn(2, 4)
out2 = torch.randn(2, 4)

ARGS = {
    "out1": out1,
    "out2": out2,
}
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
dis_mcd Ran
ran; our contract draft was wrong, not the code
OpenGVLab/EgoExoLearn/temporal_action_segmentation_benchmark/loss.py
code served (permissive licence) · get_code("e1dc22b6c3aff0f2")
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

torch.manual_seed(0)

# Create two small tensors for out1 and out2
# The function computes mean absolute difference between them
# Both should have the same shape
out1 = torch.randn(2, 4)
out2 = torch.randn(2, 4)

ARGS = {
    "out1": out1,
    "out2": out2,
}
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
mean_average_precision Ran
ran, honoured the contract we drafted
OpenGVLab/EgoExoLearn/action_anticipation_planning_benchmark/anticipation_main.py
code served (permissive licence) · get_code("2b38a33ad005fe8c")
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 float64, ndim 0, 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 numpy as np

np.random.seed(0)

# Create sample data for multi-label recognition
# scores: list of prediction scores for each sample
# labels: list of ground truth many-hot vectors for each sample

num_samples = 2
num_classes = 4

scores = [
    np.random.rand(num_classes).astype(np.float64),
    np.random.rand(num_classes).astype(np.float64),
]

labels = [
    np.array([1, 0, 1, 0], dtype=np.int64),
    np.array([0, 1, 1, 0], dtype=np.int64),
]

ARGS = {
    "scores": scores,
    "labels": labels,
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
train Not yet run
unverified (harvested, no recorded run)
OpenGVLab/EgoExoLearn/action_anticipation_planning_benchmark/anticipation_main.py
code served (permissive licence) · get_code("215d8bffab54d8ff")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 7 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 0 later) and 1 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
cross_entropy_soft 5 other papers
5 dated earlier
1907.12743 (2019-07)
Temporal Attentive Alignment for Large-Scale Video Domain Adaptation
2003.02824 (2020-03)
Action Segmentation with Joint Self-Supervised Temporal Domain Adaptation
2006.13256 (2020-06)
Rescaling Egocentric Vision
and 2 more not named here
attentive_entropy 4 other papers
4 dated earlier
1907.12743 (2019-07)
Temporal Attentive Alignment for Large-Scale Video Domain Adaptation
2006.13256 (2020-06)
Rescaling Egocentric Vision
2007.15829 (2020-07)
Adversarial Bipartite Graph Learning for Video Domain Adaptation
and 1 more not named here
dis_MCD 4 other papers
4 dated earlier
1907.12743 (2019-07)
Temporal Attentive Alignment for Large-Scale Video Domain Adaptation
2006.13256 (2020-06)
Rescaling Egocentric Vision
2007.15829 (2020-07)
Adversarial Bipartite Graph Learning for Video Domain Adaptation
and 1 more not named here
attentive_entropy 1 other paper
1 dated earlier
2003.02824 (2020-03)
Action Segmentation with Joint Self-Supervised Temporal Domain Adaptation
binary_precision_recall_curve 1 other paper
1 same year, order not determinable
Wang_CPR-Coach_Recognizing_Composite_Error_Actions_based_on_Single-class_Training_CVPR_2024_paper (2024 (year only))
CPR-Coach: Recognizing Composite Error Actions based on Single-class Training
dis_mcd 1 other paper
1 dated earlier
2003.02824 (2020-03)
Action Segmentation with Joint Self-Supervised Temporal Domain Adaptation
mean_average_precision 1 other paper
1 same year, order not determinable
Wang_CPR-Coach_Recognizing_Composite_Error_Actions_based_on_Single-class_Training_CVPR_2024_paper (2024 (year only))
CPR-Coach: Recognizing Composite Error Actions based on Single-class Training

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

Being able to map the activities of others into one's own point of view is a fundamental human skill even from a very early age. Taking a step toward understanding this human ability, we introduce EgoExoLearn, a large-scale dataset that emulates the human demonstration following process, in which individuals record egocentric videos as they execute tasks guided by exocentric-view demonstration videos. Focusing on the potential applications in daily assistance and professional support, EgoExoLearn contains egocentric and demonstration video data spanning 120 hours captured in daily life scenarios and specialized laboratories. Along with the videos we record high-quality gaze data and provide detailed multimodal annotations, formulating a playground for modeling the human ability to bridge asynchronous procedural actions from different viewpoints. To this end, we present benchmarks such as crossview association, cross-view action planning, and crossview referenced skill assessment, along with detailed analysis. We expect EgoExoLearn can serve as an important resource for bridging the actions across views, thus paving the way for creating AI agents capable of seamlessly learning by observing humans in the real world. The dataset and benchmark codes are available at https: //github.com/OpenGVLab/EgoExoLearn.

For agents

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

get_harvested_code_for_paper("2403.16182")
get_code_for_paper("2403.16182")
have("2403.16182")

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

curl https://syntology.ai/api/ran/2403.16182.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.16182.svg)](https://syntology.ai/paper/2403.16182)

Connect an agent — have() is free.