SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.00534 · 2026

AIRE-Prune: Asymptotic Impulse-Response Energy for State Pruning in State Space Models

Saibal Mukhopadhyay, Apurba Padhy, Fernando Camacho

arXiv · PDF · Open in the Atlas

Code that ran

15 of 18 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; 3 honoured the contract we drafted, 1 violated it and 1 has no contract verdict; 10 failures of our instrument; 3 no recorded run), measured 2026-11-21; 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/2602.00534.json · badge: /api/ran/2602.00534.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
falcon-arrow/AIRE-Prune canonical (classifier label)
classifier basis: census
15 ran · 3 no recorded run
FunctionStatusWhere it lives
binary_operator Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
falcon-arrow/AIRE-Prune/s5/ssm.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from msgwak/last, is pointer only (licence: NONE) · get_code("b4fc72406d2ff673")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from msgwak/last, 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.

cosine_annealing Ran
ran, honoured the contract we drafted
falcon-arrow/AIRE-Prune/s5/train_helpers.py
code served (permissive licence) · get_code("faeebb869aee2124")
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 ArrayImpl, 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 jax.numpy as np

# The function takes scalar or array step values and scalar parameters
# step: can be an integer or array
# base_lr: base learning rate (scalar float)
# end_step: total number of steps (scalar int)
# lr_min: minimum learning rate (scalar float, optional)

ARGS = {
    "step": 0,
    "base_lr": 0.1,
    "end_step": 100,
    "lr_min": 1e-06,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, 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
create_lra_imdb_classification_dataset Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
falcon-arrow/AIRE-Prune/s5/dataloading.py
code served (permissive licence) · get_code("476b5f7f116d7c9b")
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, ndim 1, 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 torch
from pathlib import Path
from unittest.mock import Mock, MagicMock

torch.manual_seed(0)

class MockIMDB:
    """Mock IMDB dataset object matching the expected interface"""
    def __init__(self, name):
        self.name = name
        self.cache_dir = None
        self.d_output = 2  # binary classification
        self.l_max = 512  # max sequence length
        self.dataset_train = [(torch.randn(1, 512), torch.tensor(0)), (torch.randn(1, 512), torch.tensor(1))]
        self.dataset_test = [(torch.randn(1, 512), torch.tensor(0)), (torch.randn(1, 512), torch.tensor(1))]
        self._collate_fn = None
    
    def setup(self):
        """Setup method called by create_lra_imdb_classification_dataset"""
        pass

def mock_imdb_constructor(name):
    """Factory to create mock IMDB objects"""
    return MockIMDB(name)

# Patch the import in the function
import sys
mock_module = MagicMock()
mock_module.IMDB = mock_imdb_constructor
sys.modules['s5.dataloaders.lra'] = mock_module

ARGS = {
    'cache_dir': Path('./cache_dir/'),
    'bsz': 50,
    'seed': 42,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
s5, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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
deprecated Ran
ran; our contract draft was wrong, not the code
falcon-arrow/AIRE-Prune/s5/dataloaders/base.py
code served (permissive licence) · get_code("25edc240223b76c0")
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

torch.manual_seed(0)

# The function 'deprecated' takes a single parameter: cls_or_func
# It returns a decorator that wraps the input and prints a deprecation message
# The input can be either a class or a function

# Create a simple function to use as the input
def sample_function(x, y):
    """A simple sample function"""
    return x + y

# Create a simple class to use as the input
class SampleClass:
    """A simple sample class"""
    def __init__(self, value):
        self.value = value
    
    def method(self):
        return self.value * 2

# We can use either the function or the class as the argument
# Let's use the function as our test case
ARGS = {
    'cls_or_func': sample_function
}
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 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
dequantize Ran
ran; our contract draft was wrong, not the code
falcon-arrow/AIRE-Prune/s5/dataloaders/audio.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from xinruihe389-commits/LongSpike, is pointer only (licence: NOASSERTION) · get_code("ce310b19a858070f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from xinruihe389-commits/LongSpike, and GitHub could not classify the licence of that repository (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.

linear_warmup Ran
ran, honoured the contract we drafted
falcon-arrow/AIRE-Prune/s5/train_helpers.py
code served (permissive licence) · get_code("a2cd676d4d4c4910")
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 float, 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)

# Analyzing the function:
# def linear_warmup(step, base_lr, end_step, lr_min=None):
#     return base_lr * (step + 1) / end_step
#
# Parameters:
# - step: integer, used in arithmetic operations
# - base_lr: float/numeric, used in multiplication
# - end_step: integer, used in division
# - lr_min: optional parameter (default None), not used in function body

ARGS = {
    "step": 100,
    "base_lr": 0.001,
    "end_step": 1000,
    "lr_min": None,
}
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 float, 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
listops_tokenizer Ran
ran; our contract draft was wrong, not the code
falcon-arrow/AIRE-Prune/s5/dataloaders/lra.py
code served (permissive licence) · get_code("72d464924b5dee96")
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 list, ndim 1, 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 torch
import numpy as np

torch.manual_seed(0)

def listops_tokenizer(s):
    return s.translate({ord(']'): ord('X'), ord('('): None, ord(')'): None}).split()

# The function takes a single string parameter 's' and:
# 1. Calls translate() on it with a translation table that replaces ']' with 'X' and removes '(' and ')'
# 2. Calls split() on the result to return a list of tokens
# 
# We need to provide a valid string that demonstrates the tokenizer works

s = "[ nested [ list ] ( with ) parens ] here"

ARGS = {
    "s": s
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
make_DPLR_HiPPO Ran
ran; our contract draft was wrong, not the code
falcon-arrow/AIRE-Prune/s5/ssm_init.py
code served (permissive licence) · get_code("2b5694fa1220f8bd")
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 jax.numpy as np
from jax.numpy.linalg import eigh

ARGS = {
    "N": 4,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — 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
make_HiPPO Ran
ran, honoured the contract we drafted
falcon-arrow/AIRE-Prune/s5/ssm_init.py
code served (permissive licence) · get_code("3f26e3f7ea0487f3")
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 ArrayImpl, ndim 2, dtype family float, shape (4, 4) 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 jax.numpy as np

ARGS = {
    "N": 4,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family float, shape (4, 4) — shape (4, 4) 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
make_NPLR_HiPPO Ran
ran; our contract draft was wrong, not the code
falcon-arrow/AIRE-Prune/s5/ssm_init.py
code served (permissive licence) · get_code("0b867bda221aedfc")
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 jax.numpy as np

ARGS = {
    "N": 8,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — 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
make_data_loader Ran
ran on a synthesized input
falcon-arrow/AIRE-Prune/s5/dataloading.py
code served (permissive licence) · get_code("508a42e36707cd15")
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 DataLoader. 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.utils.data import Dataset

torch.manual_seed(0)

class SimpleDataset(Dataset):
    def __init__(self, size=10, input_dim=4, output_dim=3):
        self.size = size
        self.input_dim = input_dim
        self.output_dim = output_dim
    
    def __len__(self):
        return self.size
    
    def __getitem__(self, idx):
        x = torch.randn(self.input_dim)
        y = torch.randint(0, self.output_dim, (1,)).item()
        return x, y

class DataObject:
    def __init__(self):
        pass
    
    def _collate_fn(self, batch):
        xs = torch.stack([item[0] for item in batch])
        ys = torch.tensor([item[1] for item in batch])
        return xs, ys

dset = SimpleDataset(size=10, input_dim=4, output_dim=3)
dobj = DataObject()

ARGS = {
    "dset": dset,
    "dobj": dobj,
    "seed": 42,
    "batch_size": 2,
    "shuffle": True,
    "drop_last": True,
    "collate_fn": None,
}
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 DataLoader — a DataLoader output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataLoader 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
masked_meanpool Ran
ran, violated the contract we drafted
falcon-arrow/AIRE-Prune/s5/seq_model.py
code served (permissive licence) · get_code("261d3ddc455af2a0")
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 ArrayImpl, ndim 1, dtype family float, shape (4,) 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 jax.numpy as jnp
import numpy as np

# Create a small sequence with variable length
# x should be (L, d_model) where L is sequence length and d_model is model dimension
L = 5  # sequence length
d_model = 4  # model dimension

# Create input sequence as JAX array
x = jnp.array(np.random.randn(L, d_model), dtype=jnp.float32)

# Create lengths - this should be the original sequence length before padding
# Using a value less than L to simulate actual sequence being shorter than padded length
lengths = jnp.array(3, dtype=jnp.int32)

ARGS = {
    "x": x,
    "lengths": lengths
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 1, dtype family float, shape (4,) — shape (4,) 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
minmax_scale Ran
ran; our contract draft was wrong, not the code
falcon-arrow/AIRE-Prune/s5/dataloaders/audio.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from xinruihe389-commits/LongSpike, is pointer only (licence: NOASSERTION) · get_code("1b76e4737067023f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from xinruihe389-commits/LongSpike, and GitHub could not classify the licence of that repository (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.

quantize Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
falcon-arrow/AIRE-Prune/s5/dataloaders/audio.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from xinruihe389-commits/LongSpike, is pointer only (licence: NOASSERTION) · get_code("1631cd936e9b1a90")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from xinruihe389-commits/LongSpike, and GitHub could not classify the licence of that repository (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.

reduce_lr_on_plateau Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
falcon-arrow/AIRE-Prune/s5/train_helpers.py
code served (permissive licence) · get_code("2802de0d2519b69a")
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, ndim 1, 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 torch
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

# Construct the input tuple for the function
# The function expects: (lr, ssm_lr, count, new_acc, opt_acc)
lr = 0.001
ssm_lr = 0.0005
count = 5
new_acc = 0.95
opt_acc = 0.92

input_tuple = (lr, ssm_lr, count, new_acc, opt_acc)

ARGS = {
    'input': input_tuple,
    'factor': 0.2,
    'patience': 20,
    'lr_min': 1e-06
}
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, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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
create_lra_listops_classification_dataset Not yet run
unverified (harvested, no recorded run)
falcon-arrow/AIRE-Prune/s5/dataloading.py
code served (permissive licence) · get_code("1ee8fd3237e3786e")
How it ran

no invocation recorded.

discretize_bilinear Not yet run
unverified (harvested, no recorded run)
falcon-arrow/AIRE-Prune/s5/ssm.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from msgwak/last, is pointer only (licence: NONE) · get_code("c0135e087b049ad1")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from msgwak/last, 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. No invocation recorded.

discretize_zoh Not yet run
unverified (harvested, no recorded run)
falcon-arrow/AIRE-Prune/s5/ssm.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from msgwak/last, is pointer only (licence: NONE) · get_code("f9485e8e0d1e9aee")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from msgwak/last, 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. No invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 17 are also present in other papers whose code we hold (17 with at least one paper dated earlier, 0 the same month, 3 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
dequantize 6 other papers
5 dated earlier, 1 later
2208.04933 (2022-08)
Simplified State Space Layers for Sequence Modeling
2310.02980 (2023-10)
Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors
2402.15584 (2024-02)
State Space Models for Event Cameras
and 3 more not named here
listops_tokenizer 6 other papers
6 dated earlier
2206.05852 (2022-06)
ChordMixer: A Scalable Neural Attention Model for Sequences with Different Lengths
2208.04933 (2022-08)
Simplified State Space Layers for Sequence Modeling
2301.10540 (2023-01)
Modelling Long Range Dependencies in $N$D: From Task-Specific to a General Purpose CNN
and 3 more not named here
minmax_scale 6 other papers
5 dated earlier, 1 later
2208.04933 (2022-08)
Simplified State Space Layers for Sequence Modeling
2310.02980 (2023-10)
Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors
2402.15584 (2024-02)
State Space Models for Event Cameras
and 3 more not named here
quantize 6 other papers
5 dated earlier, 1 later
2208.04933 (2022-08)
Simplified State Space Layers for Sequence Modeling
2310.02980 (2023-10)
Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors
2402.15584 (2024-02)
State Space Models for Event Cameras
and 3 more not named here
deprecated 5 other papers
5 dated earlier
2310.02980 (2023-10)
Never Train from Scratch: Fair Comparison of Long-Sequence Models Requires Data-Driven Priors
2402.15584 (2024-02)
State Space Models for Event Cameras
2403.03234 (2024-03)
Caduceus: Bi-Directional Equivariant Long-Range DNA Sequence Modeling
and 2 more not named here
binary_operator 4 other papers
4 dated earlier
2208.04933 (2022-08)
Simplified State Space Layers for Sequence Modeling
2402.15584 (2024-02)
State Space Models for Event Cameras
2407.19115 (2024-07)
Towards Scalable and Stable Parallelization of Nonlinear RNNs
and 1 more not named here
discretize_bilinear 3 other papers
3 dated earlier
2208.04933 (2022-08)
Simplified State Space Layers for Sequence Modeling
2402.15584 (2024-02)
State Space Models for Event Cameras
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
discretize_zoh 3 other papers
3 dated earlier
2208.04933 (2022-08)
Simplified State Space Layers for Sequence Modeling
2402.15584 (2024-02)
State Space Models for Event Cameras
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
make_DPLR_HiPPO 3 other papers
3 dated earlier
2402.15584 (2024-02)
State Space Models for Event Cameras
2404.18508 (2024-04)
Scalable Event-by-event Processing of Neuromorphic Sensory Signals With Deep State-Space Models
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
make_HiPPO 3 other papers
3 dated earlier
2402.15584 (2024-02)
State Space Models for Event Cameras
2404.18508 (2024-04)
Scalable Event-by-event Processing of Neuromorphic Sensory Signals With Deep State-Space Models
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
make_NPLR_HiPPO 3 other papers
3 dated earlier
2402.15584 (2024-02)
State Space Models for Event Cameras
2404.18508 (2024-04)
Scalable Event-by-event Processing of Neuromorphic Sensory Signals With Deep State-Space Models
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
masked_meanpool 3 other papers
3 dated earlier
2402.15584 (2024-02)
State Space Models for Event Cameras
2404.18508 (2024-04)
Scalable Event-by-event Processing of Neuromorphic Sensory Signals With Deep State-Space Models
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
cosine_annealing 2 other papers
2 dated earlier
2402.15584 (2024-02)
State Space Models for Event Cameras
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
create_lra_imdb_classification_dataset 2 other papers
2 dated earlier
2402.15584 (2024-02)
State Space Models for Event Cameras
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
linear_warmup 2 other papers
2 dated earlier
2402.15584 (2024-02)
State Space Models for Event Cameras
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
make_data_loader 2 other papers
2 dated earlier
2402.15584 (2024-02)
State Space Models for Event Cameras
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models
reduce_lr_on_plateau 2 other papers
2 dated earlier
2402.15584 (2024-02)
State Space Models for Event Cameras
2411.02824 (2024-11)
Layer-Adaptive State Pruning for Deep State Space Models

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

State space models (SSMs) often sacrifice capacity, search space, or stability to offset the memory and compute costs of large state dimensions. We introduce a structured post-training pruning method for SSMs -AIRE-Prune (Asymptotic Impulse-Response Energy for State PRUN(E)ing ) -that reduces each layer's state dimension by directly minimizing long-run output-energy distortion. AIRE-Prune assigns every state a closed-form asymptotic impulse-response energy based score, i.e., the total impulse-response energy it contributes over an infinite horizon (time), and normalizes these scores layer-wise to enable global cross-layer comparison and selection. This extends modal truncation from single systems to deep stacks and aligns pruning with asymptotic response energy rather than worstcase gain. Across diverse sequence benchmarks, AIRE-Prune reveals substantial redundancy in SISO and MIMO SSMs with average pruning of 60.8% , with average accuracy drop of 0.29% without retraining while significantly lowering compute. Code will be released: https://github.com/falcon-arrow/AIRE-Prune.

For agents

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

get_harvested_code_for_paper("2602.00534")
get_code_for_paper("2602.00534")
have("2602.00534")

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

curl https://syntology.ai/api/ran/2602.00534.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.00534.svg)](https://syntology.ai/paper/2602.00534)

Connect an agent — have() is free.