SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2109.13396 · 2021

Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasets

arXiv · PDF · Open in the Atlas

Code that ran

4 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 4 have no contract verdict; 0 failures of our instrument; 2 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/2109.13396.json · badge: /api/ran/2109.13396.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
yanlai00/bridge_data_imitation_learning community (classifier label)
classifier basis not recorded on the link
4 ran · 2 no recorded run
FunctionStatusWhere it lives
action2img Ran
ran on a synthesized input
yanlai00/bridge_data_imitation_learning/bridgedata/models/gcbc_images.py
code served (permissive licence) · get_code("90b46f884c68076f")
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 3, dtype family float, shape (64, 64, 3) 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
import cv2

# Construct arguments for action2img function
action = np.array([0.5, -0.3], dtype=np.float32)  # 2D action vector
res = 64  # resolution (image will be 64x64)
channels = 3  # RGB image
action_scale = 0.8  # scaling factor for action magnitude

ARGS = {
    "action": action,
    "res": res,
    "channels": channels,
    "action_scale": action_scale
}
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 ndarray, ndim 3, dtype family float, shape (64, 64, 3) — shape (64, 64, 3) 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
compute_alpha Ran
ran on a synthesized input
yanlai00/bridge_data_imitation_learning/bridgedata/models/utils/gradient_reversal_layer.py
code served (permissive licence) · get_code("4541f3b638b810b1")
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

class InputsConfig:
    def __init__(self, global_step, max_iterations):
        self.global_step = global_step
        self.max_iterations = max_iterations

inputs = InputsConfig(global_step=50, max_iterations=100)
alpha_delay = 10

ARGS = {
    "inputs": inputs,
    "alpha_delay": alpha_delay
}
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
compute_dataset_normalization Ran
ran on a synthesized input
yanlai00/bridge_data_imitation_learning/bridgedata/models/utils/compute_dataset_normalization.py
code served (permissive licence) · get_code("7e0c25eb35180e42")
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 dict. 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 a simple dataloader that yields batches with 'states' and 'actions'
class SimpleDataloader:
    def __init__(self, num_batches=5, state_dim=4, action_dim=3):
        self.num_batches = num_batches
        self.state_dim = state_dim
        self.action_dim = action_dim
    
    def __iter__(self):
        for batch_idx in range(self.num_batches):
            # Each batch has batch_size=10
            states = np.random.randn(10, self.state_dim).astype(np.float32)
            actions = np.random.randn(10, self.action_dim).astype(np.float32)
            yield {
                'states': states,
                'actions': actions
            }

# Create dataloader with enough samples to avoid the pdb.set_trace() warning
# We need more than 1000 samples total across batches
dataloader = SimpleDataloader(num_batches=150, state_dim=4, action_dim=3)

ARGS = {
    'dataloader': dataloader,
    'no_last_dim_norm': 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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
compute_gmm_loss Ran
ran on a synthesized input
yanlai00/bridge_data_imitation_learning/bridgedata/models/gcbc_images.py
code served (permissive licence) · get_code("5f02da9d73579fdd")
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)

# Define shapes based on function signature and usage:
# means: shape B, num_comp, adim
# covariances: shape B, num_comp, adim
# mixing_coefficients: shape B, num_comp (used with Softmax(dim=1))
# action_targets: shape B, adim

batch_size = 2
num_components = 3
action_dim = 4

means = torch.randn(batch_size, num_components, action_dim)
covariances = torch.randn(batch_size, num_components, action_dim)
mixing_coefficients = torch.randn(batch_size, num_components)
action_targets = torch.randn(batch_size, action_dim)

ARGS = {
    "means": means,
    "covariances": covariances,
    "mixing_coefficients": mixing_coefficients,
    "action_targets": action_targets,
}
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
get_random_crop Not yet run
unverified (harvested, no recorded run)
yanlai00/bridge_data_imitation_learning/bridgedata/data_sets/data_augmentation.py
code served (permissive licence) · get_code("8351f50d15778ac7")
How it ran

no invocation recorded.

get_tlen_from_padmask Not yet run
unverified (harvested, no recorded run)
yanlai00/bridge_data_imitation_learning/bridgedata/models/gcbc_images.py
code served (permissive licence) · get_code("99a85717759f4bce")
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, 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 6 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

Robot learning holds the promise of learning policies that generalize broadly. However, such generalization requires sufficiently diverse datasets of the task of interest, which can be prohibitively expensive to collect. In other fields, such as computer vision, it is common to utilize shared, reusable datasets, such as ImageNet, to overcome this challenge, but this has proven difficult in robotics. In this paper, we ask: what would it take to enable practical data reuse in robotics for end-to-end skill learning? We hypothesize that the key is to use datasets with multiple tasks and multiple domains, such that a new user that wants to train their robot to perform a new task in a new domain can include this dataset in their training process and benefit from cross-task and cross-domain generalization. To evaluate this hypothesis, we collect a large multi-domain and multi-task dataset, with 7,200 demonstrations constituting 71 tasks across 10 environments, and empirically study how this data can improve the learning of new tasks in new environments. We find that jointly training with the proposed dataset and 50 demonstrations of a never-before-seen task in a new domain on average leads to a 2x improvement in success rate compared to using target domain data alone. We also find that data for only a few tasks in a new domain can bridge the domain gap and make it possible for a robot to perform a variety of prior tasks that were only seen in other domains. These results suggest that reusing diverse multi-task and multi-domain datasets, including our open-source dataset, may pave the way for broader robot generalization, eliminating the need to re-collect data for each new robot learning project.

For agents

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

get_harvested_code_for_paper("2109.13396")
get_code_for_paper("2109.13396")
have("2109.13396")

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

curl https://syntology.ai/api/ran/2109.13396.json

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

[![Syntology run record](https://syntology.ai/api/ran/2109.13396.svg)](https://syntology.ai/paper/2109.13396)

Connect an agent — have() is free.