SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2512.00239 · ICML · 2025

Self-Supervised Dynamical System Representations for Physiological Time-Series

Christopher Rozell, James Rehg, Maxwell Xu, Yenho Chen

arXiv · PDF · Open in the Atlas

Code that ran

18 of 27 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, 2 violated it and 1 has no contract verdict; 14 failures of our instrument; 9 no recorded run), measured 2026-11-22; 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/2512.00239.json · badge: /api/ran/2512.00239.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
arsedler9/lfads-torch canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
yuqinie98/PatchTST canonical (classifier label)
classifier basis: census
6 ran · 6 no recorded run
Mingyue-Cheng/TimeMAE canonical (classifier label)
classifier basis: census
3 ran · 2 no recorded run
yenhochen/PULSE canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_l2_penalty Ran
ran, honoured the contract we drafted
arsedler9/lfads-torch/lfads_torch/modules/l2.py
pointer only (licence: NOASSERTION) · get_code("286f46042ec53f91")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, arsedler9/lfads-torch (NOASSERTION). 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.

default_device Ran
ran; our contract draft was wrong, not the code
yuqinie98/PatchTST/PatchTST_self_supervised/src/basics.py
code served (permissive licence) · get_code("e540fde478fcfbaa")
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 device. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

ARGS = {
    "use_cuda": False,
}
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 device — a device output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a device 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
has_image_loggers Ran
ran, violated the contract we drafted
arsedler9/lfads-torch/lfads_torch/callbacks.py
pointer only (licence: NOASSERTION) · get_code("df9f400e259f725e")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, arsedler9/lfads-torch (NOASSERTION). 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.

join_path_file Ran
ran on a synthesized input
yuqinie98/PatchTST/PatchTST_self_supervised/src/learner.py
code served (permissive licence) · get_code("400d8718dacb52b2")
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 PosixPath. 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.

from pathlib import Path
import tempfile

# Create a temporary directory for testing
temp_dir = tempfile.mkdtemp()

ARGS = {
    'file': 'test_file',
    'path': temp_dir,
    'ext': '.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 PosixPath — a PosixPath output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PosixPath 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
load_HAR Ran
ran; our contract draft was wrong, not the code
Mingyue-Cheng/TimeMAE/datautils.py
pointer only (licence: NONE) · get_code("360457508fde988c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Mingyue-Cheng/TimeMAE 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.

mae Ran
ran; our contract draft was wrong, not the code
yuqinie98/PatchTST/PatchTST_self_supervised/src/metrics.py
code served (permissive licence) · get_code("4d345bf5bfa84ace")
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 tensors for y_true and y_pred
# Both should be the same shape for l1_loss
y_true = torch.randn(2, 4)
y_pred = torch.randn(2, 4)

ARGS = {
    'y_true': y_true,
    'y_pred': y_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
mse Ran
ran; our contract draft was wrong, not the code
yuqinie98/PatchTST/PatchTST_self_supervised/src/metrics.py
code served (permissive licence) · get_code("5ef21ce9effc9300")
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 tensors for y_true and y_pred
# Both should have the same shape for MSE loss
y_true = torch.randn(2, 4)  # batch_size=2, features=4
y_pred = torch.randn(2, 4)  # batch_size=2, features=4

ARGS = {
    'y_true': y_true,
    'y_pred': y_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
nested_children Ran
ran; our contract draft was wrong, not the code
yuqinie98/PatchTST/PatchTST_self_supervised/src/utils.py
code served (permissive licence) · get_code("b7c5899bb16035d3")
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 torch
from torch import nn

torch.manual_seed(0)

# Create a simple nested module structure
class SimpleLeaf(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)

class SimpleParent(nn.Module):
    def __init__(self):
        super().__init__()
        self.leaf1 = SimpleLeaf()
        self.leaf2 = SimpleLeaf()

class SimpleGrandparent(nn.Module):
    def __init__(self):
        super().__init__()
        self.parent1 = SimpleParent()
        self.parent2 = SimpleParent()

# Create a nested module
nested_module = SimpleGrandparent()

ARGS = {
    "m": nested_module
}
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 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
pad_mask Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
arsedler9/lfads-torch/lfads_torch/modules/augmentations.py
pointer only (licence: NOASSERTION) · get_code("746e0e6bbeb89c5b")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, arsedler9/lfads-torch (NOASSERTION). 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.

padding_varying_length Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Mingyue-Cheng/TimeMAE/datautils.py
pointer only (licence: NONE) · get_code("de18456a7e019a9d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Mingyue-Cheng/TimeMAE 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.

r2_score Ran
ran; our contract draft was wrong, not the code
arsedler9/lfads-torch/lfads_torch/metrics.py
pointer only (licence: NOASSERTION) · get_code("a27593d38113b6a5")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, arsedler9/lfads-torch (NOASSERTION). 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.

recalls_and_ndcgs_for_ks Ran
ran; our contract draft was wrong, not the code
Mingyue-Cheng/TimeMAE/loss.py
pointer only (licence: NONE) · get_code("c59c3677a8ada751")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Mingyue-Cheng/TimeMAE 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.

regional_bits_per_spike Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
arsedler9/lfads-torch/lfads_torch/metrics.py
pointer only (licence: NOASSERTION) · get_code("b2719609b2d567a8")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, arsedler9/lfads-torch (NOASSERTION). 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.

reshuffle_train_valid Ran
ran; our contract draft was wrong, not the code
arsedler9/lfads-torch/lfads_torch/datamodules.py
pointer only (licence: NOASSERTION) · get_code("f399660b7aaaf996")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, arsedler9/lfads-torch (NOASSERTION). 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.

resolve_save_dir Ran
ran; our contract draft was wrong, not the code
yenhochen/PULSE/scripts/run/transfer.py
pointer only (licence: NONE) · get_code("2501bcbdbdc51f74")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yenhochen/PULSE 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.

rmse Ran
ran; our contract draft was wrong, not the code
yuqinie98/PatchTST/PatchTST_self_supervised/src/metrics.py
code served (permissive licence) · get_code("1799dcc70885ac60")
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 tensor arguments for the rmse function
# y_true and y_pred should be tensors of the same shape
y_true = torch.randn(2, 4)
y_pred = torch.randn(2, 4)

ARGS = {
    'y_true': y_true,
    'y_pred': y_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
supervised_classification Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yenhochen/PULSE/scripts/run/pretrain.py
pointer only (licence: NONE) · get_code("cd59ac6456cef2c5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yenhochen/PULSE 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.

to_tensor Ran
ran, violated the contract we drafted
arsedler9/lfads-torch/lfads_torch/datamodules.py
pointer only (licence: NOASSERTION) · get_code("99ba6555f6cb1bc6")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, arsedler9/lfads-torch (NOASSERTION). 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.

bits_per_spike Not yet run
unverified (harvested, no recorded run)
arsedler9/lfads-torch/lfads_torch/metrics.py
pointer only (licence: NOASSERTION) · get_code("e44e57b951a765b2")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, arsedler9/lfads-torch (NOASSERTION). 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. No invocation recorded.

fit_lr Not yet run
unverified (harvested, no recorded run)
Mingyue-Cheng/TimeMAE/classification.py
pointer only (licence: NONE) · get_code("d354e00aa94a14de")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Mingyue-Cheng/TimeMAE 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. No invocation recorded.

flatten_dict Not yet run
unverified (harvested, no recorded run)
yuqinie98/PatchTST/PatchTST_self_supervised/src/utils.py
code served (permissive licence) · get_code("4ac8e022e2aa0b43")
How it ran

no invocation recorded.

get_available_cuda Not yet run
unverified (harvested, no recorded run)
yuqinie98/PatchTST/PatchTST_self_supervised/src/basics.py
code served (permissive licence) · get_code("129e9af6b7bc7405")
How it ran

no invocation recorded.

get_device Not yet run
unverified (harvested, no recorded run)
yuqinie98/PatchTST/PatchTST_self_supervised/src/basics.py
code served (permissive licence) · get_code("e9773590d2d18ec2")
How it ran

no invocation recorded.

get_model Not yet run
unverified (harvested, no recorded run)
yuqinie98/PatchTST/PatchTST_self_supervised/src/learner.py
code served (permissive licence) · get_code("53d79ebc1fd8ae72")
How it ran

no invocation recorded.

load_UCR Not yet run
unverified (harvested, no recorded run)
Mingyue-Cheng/TimeMAE/datautils.py
pointer only (licence: NONE) · get_code("d3bc4a1920a996ab")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Mingyue-Cheng/TimeMAE 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. No invocation recorded.

transfer_weights Not yet run
unverified (harvested, no recorded run)
yuqinie98/PatchTST/PatchTST_self_supervised/src/learner.py
code served (permissive licence) · get_code("81ff9ec6d6ca9477")
How it ran

no invocation recorded.

unwrap_model Not yet run
unverified (harvested, no recorded run)
yuqinie98/PatchTST/PatchTST_self_supervised/src/utils.py
code served (permissive licence) · get_code("3cc94cf719642c1f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 27 samples lifted from this paper's linked repositories, 27 carry a function identity and 0 do not. Of the 27 distinct functions with an identity, 20 are also present in other papers whose code we hold (18 with at least one paper dated earlier, 0 the same month, 2 later) and 7 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
bits_per_spike 2 other papers
2 dated earlier
2109.04463 (2021-09)
Neural Latents Benchmark '21: Evaluating latent variable models of neural population activity
2403.01371 (2024-03)
eXponential FAmily Dynamical Systems (XFADS): Large-scale nonlinear Gaussian state-space modeling
get_model 2 other papers
2 dated earlier
2402.02475 (2024-02)
TimeSiam: A Pre-Training Framework for Siamese Time-Series Modeling
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
default_device 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
fit_lr 1 other paper
1 later
2602.20459 (2026-02)
PRESCIENCE: A Dataset and Benchmark for Scientific Forecasting
flatten_dict 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
get_available_cuda 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
get_device 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
has_image_loggers 1 other paper
1 dated earlier
2212.03771 (2022-12)
Expressive architectures enhance interpretability of dynamics-based neural population models
join_path_file 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
mae 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
mse 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
nested_children 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
padding_varying_length 1 other paper
1 dated earlier
2403.01493 (2024-03)
ConvTimeNet: A Deep Hierarchical Fully Convolutional Model for Multivariate Time Series Analysis
r2_score 1 other paper
1 dated earlier
2109.04463 (2021-09)
Neural Latents Benchmark '21: Evaluating latent variable models of neural population activity
recalls_and_ndcgs_for_ks 1 other paper
1 later
2602.20459 (2026-02)
PRESCIENCE: A Dataset and Benchmark for Scientific Forecasting
regional_bits_per_spike 1 other paper
1 dated earlier
2109.04463 (2021-09)
Neural Latents Benchmark '21: Evaluating latent variable models of neural population activity
rmse 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
to_tensor 1 other paper
1 dated earlier
2212.03771 (2022-12)
Expressive architectures enhance interpretability of dynamics-based neural population models
transfer_weights 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM
unwrap_model 1 other paper
1 dated earlier
2407.10240 (2024-07)
xLSTMTime : Long-term Time Series Forecasting With xLSTM

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

Self-supervised learning for physiological timeseries aims to captures the identity of the underlying dynamical process while filtering irrelevant noise. However, existing approaches may obscure the clinical semantics important for downstream transferability. Weakly constrained pretext tasks (i.e. contrastive learning, MAE) may incorrectly ignore the underlying dynamical structure, while structurally constrained models (i.e. SVAEs) are unable to selectively filter sample-specific noise. To bridge this gap, we propose PULSE, a novel pretraining objective that simultaneously preserves dynamical relationships important to physiological time-series while selectively removing irrelevant noise. We achieve this by formulating a dynamical systems model to identify transferable and non-transferable information between time-series windows, and target the former through a novel cross-reconstruction objective. We establish theory that provides conditions for when transferrable information is recovered, and empirically validate it through synthetic experiments. On several real-world datasets, PULSE effectively distinguishes clinical semantic classes, increases label efficiency, and improves transfer learning performance.

For agents

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

get_harvested_code_for_paper("2512.00239")
get_code_for_paper("2512.00239")
have("2512.00239")

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

curl https://syntology.ai/api/ran/2512.00239.json

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

[![Syntology run record](https://syntology.ai/api/ran/2512.00239.svg)](https://syntology.ai/paper/2512.00239)

Connect an agent — have() is free.