SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.02559 · 2020

Enhancing SAT solvers with glue variable predictions

arXiv · PDF · Open in the Atlas

Code that ran

8 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, 1 violated it and 0 have no contract verdict; 6 failures of our instrument; 0 no recorded run), measured 2026-11-21; 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/2007.02559.json · badge: /api/ran/2007.02559.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
jesse-michael-han/neuro-cadical canonical (classifier label)
classifier basis: classifier_2025
8 ran · 0 no recorded run
FunctionStatusWhere it lives
decode_activation Ran
ran; our contract draft was wrong, not the code
jesse-michael-han/neuro-cadical/python/gnn.py
code served (permissive licence) · get_code("d516335bf8ec4fd4")
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) and returned type. 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)

ARGS = {
    'activation': 'relu'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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
discount_cumsum Ran
ran; our contract draft was wrong, not the code
jesse-michael-han/neuro-cadical/python/rl_loop.py
code served (permissive licence) · get_code("658d370f8dacf5e3")
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) and returned ndarray, 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 numpy as np
import scipy.signal

# Create a simple numpy array as input
x = np.array([1.0, 2.0, 3.0, 4.0])

# Define the discount factor
discount = 0.99

# Construct the ARGS dictionary
ARGS = {
    "x": x,
    "discount": discount
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, 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
flop Ran
ran, honoured the contract we drafted
jesse-michael-han/neuro-cadical/python/gnn.py
code served (permissive licence) · get_code("2a1029d8aecc0f37")
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) and returned Tensor, ndim 2, dtype family float, shape (2, 6) 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)

# The function expects L_logits to be a tensor with at least 1 dimension
# It splits the tensor in half along dimension 0, then concatenates along dimension 1
# So we need a 2D tensor where we can split rows and concatenate columns

# Create a tensor with shape (4, 3) - 4 rows, 3 columns
# When split in half: first half is (2, 3), second half is (2, 3)
# When concatenated on dim=1: result is (2, 6)
L_logits = torch.randn(4, 3)

ARGS = {
    "L_logits": L_logits
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 2, dtype family float, shape (2, 6) — shape (2, 6) 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_name Ran
ran; our contract draft was wrong, not the code
jesse-michael-han/neuro-cadical/python/solver_eval.py
code served (permissive licence) · get_code("ff660a889a945b5b")
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) and returned str. 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 os
import tempfile

# Create a temporary directory structure to simulate a checkpoint path
temp_dir = tempfile.mkdtemp()
model_dir = os.path.join(temp_dir, "my_model")
os.makedirs(model_dir, exist_ok=True)
checkpoint_path = os.path.join(model_dir, "checkpoint_epoch_5.pt")

# Create the checkpoint file (empty file is fine for this function)
Path(checkpoint_path).touch()

ARGS = {
    "path": checkpoint_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
is_baseline Ran
ran, violated the contract we drafted
jesse-michael-han/neuro-cadical/python/solver_eval.py
code served (permissive licence) · get_code("5d3eed6eb2dfde23")
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) and returned bool, 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)

# The function is_baseline takes a single parameter 'mopts' which is a dict
# It calls .get() on it with various keys and checks their values
# We need to construct a dict that can be passed to this function

mopts = {
    'random_refocus': False,
    'branch_mode': 1,
    'refocus': True
}

ARGS = {
    'mopts': mopts
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, 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
mk_G Ran
ran; our contract draft was wrong, not the code
jesse-michael-han/neuro-cadical/python/rl_loop.py
code served (permissive licence) · get_code("be14495f3c766a0a")
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) and returned Tensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

class CL_idxs_obj:
    def __init__(self, C_idxs, L_idxs, n_clauses, n_vars):
        self.C_idxs = C_idxs
        self.L_idxs = L_idxs
        self.n_clauses = n_clauses
        self.n_vars = n_vars

# Create small test data
# 5 clause-literal pairs
C_idxs_list = [0, 1, 2, 0, 1]  # clause indices (5 values)
L_idxs_list = [0, 1, 2, 3, 4]  # literal indices (5 values)
n_clauses = 3
n_vars = 3

CL_idxs_instance = CL_idxs_obj(
    C_idxs=C_idxs_list,
    L_idxs=L_idxs_list,
    n_clauses=n_clauses,
    n_vars=n_vars
)

ARGS = {
    "CL_idxs": CL_idxs_instance
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, shape (3, 6) — shape (3, 6) as a later run recorded it
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
softmax_sample_from_logits Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
jesse-michael-han/neuro-cadical/python/rl_loop.py
code served (permissive licence) · get_code("4d3575b362679b8f")
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) and returned Tensor, ndim 0, dtype family int, 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 numpy as np

torch.manual_seed(0)
np.random.seed(0)

# Create a small logits tensor
# The function expects logits that can be passed to Categorical distribution
# Categorical expects logits of shape (..., num_classes)
logits = torch.randn(4, dtype=torch.float32)

ARGS = {
    "logits": logits,
    "exploration_eps": 0.01
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family int, 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
swap_polarity Ran
ran; our contract draft was wrong, not the code
jesse-michael-han/neuro-cadical/python/gnn.py
code served (permissive licence) · get_code("9139be05f74d45b8")
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) and returned Tensor. 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

torch.manual_seed(0)

# Create a sparse tensor G
# The function expects G to have:
# - .coalesce() method (sparse tensor method)
# - .indices() method (returns indices of non-zero elements)
# - .values() method (returns values of non-zero elements)
# - .size() method (returns tensor dimensions)

# Create a small sparse tensor with shape (4, 6)
# This gives us a pivot = 6/2 - 0.5 = 2.5
indices = torch.LongTensor([[0, 1, 2, 3], [0, 2, 4, 5]])
values = torch.FloatTensor([1.0, 2.0, 3.0, 4.0])
G = torch.sparse.FloatTensor(indices=indices, values=values, size=(4, 6))

ARGS = {
    "G": G
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, shape (4, 6) — shape (4, 6) as a later run recorded it
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

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, 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 8 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 2026-11-21.

Repositories linked to this paper

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

Abstract

Modern SAT solvers routinely operate at scales that make it impractical to query a neural network for every branching decision. NeuroCore, proposed by Selsam and Bjorner, offered a proof-of-concept that neural networks can still accelerate SAT solvers by only periodically refocusing a score-based branching heuristic. However, that work suffered from several limitations: their modified solvers require GPU acceleration, further ablations showed that they were no better than a random baseline on the SATCOMP 2018 benchmark, and their training target of unsat cores required an expensive data pipeline which only labels relatively easy unsatisfiable problems. We address all these limitations, using a simpler network architecture allowing CPU inference for even large industrial problems with millions of clauses, and training instead to predict {\em glue variables}---a target for which it is easier to generate labelled data, and which can also be formulated as a reinforcement learning task. We demonstrate the effectiveness of our approach by modifying the state-of-the-art SAT solver CaDiCaL, improving its performance on SATCOMP 2018 and SATRACE 2019 with supervised learning and its performance on a dataset of SHA-1 preimage attacks with reinforcement learning.

For agents

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

get_harvested_code_for_paper("2007.02559")
get_code_for_paper("2007.02559")
have("2007.02559")

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

curl https://syntology.ai/api/ran/2007.02559.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.02559.svg)](https://syntology.ai/paper/2007.02559)

Connect an agent — have() is free.