SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.07020 · ICLR · 2026

RESCHED: Rethinking Flexible Job Shop Scheduling from a Transformer-based Architecture with Simplified States

Zhiguang Cao, Wen Song, Cong Zhang, Xiangjie Xiao

arXiv · PDF

Code that ran

9 of 13 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 4 have no contract verdict; 4 failures of our instrument; 4 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/2603.07020.json · badge: /api/ran/2603.07020.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
XiangjieXiao/ReSched community
Papers with Code archive
9 ran · 4 no recorded run
FunctionStatusWhere it lives
Actor Ran
ran on a synthesized input · constructed an object; no result was computed
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("c43444a37a5a6755")
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 an instance of its own class, Actor: the run constructed the object and did not call it, so no result was computed. 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 = {
    "num_layers": 3,
    "input_dim": 4,
    "hidden_dim": 8,
    "output_dim": 2,
}
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 Actor — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Actor 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
Critic Ran
ran on a synthesized input · constructed an object; no result was computed
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("42d2bbe83c775a20")
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 an instance of its own class, Critic: the run constructed the object and did not call it, so no result was computed. 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 = {
    'num_layers': 3,
    'input_dim': 4,
    'hidden_dim': 8,
    'output_dim': 2
}
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 Critic — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Critic 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
FeedForwardModule Ran
ran on a synthesized input
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("e7771e74a3cf9d54")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 2, 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 torch.nn.functional as F

INIT_ARGS = {
    'embedding_dim': 4,
    'ff_hidden_dim': 6,
}

ARGS = {
    'input1': torch.randn(2, 4),
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
RMSNorm Ran
ran on a synthesized input · constructed an object; no result was computed
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("9465fd5c593cbef5")
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 an instance of its own class, RMSNorm: the run constructed the object and did not call it, so no result was computed. 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)

# RMSNorm.__init__ takes: self, dim, eps=1e-08
# RMSNorm.forward takes: self, x

# The forward method expects x to be a tensor with shape (..., dim)
# It performs operations on the last dimension

dim = 4
eps = 1e-8

# Create a small input tensor with batch size 2 and dimension 4
x = torch.randn(2, 4)

ARGS = {
    "dim": dim,
    "eps": eps,
}
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 RMSNorm — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a RMSNorm 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
apply_rope_mapping Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("8cad3cd44c72242b")
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 5, 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

torch.manual_seed(0)

# Parameters for apply_rope_mapping
# matrix_q: (bs, head, jo, dk)
# matrix_k: (bs, head, jo, dk)
# position: (bs, jo)

bs = 2  # batch size
head = 3  # number of heads
jo = 4  # sequence length
dk = 8  # key dimension

matrix_q = torch.randn(bs, head, jo, dk)
matrix_k = torch.randn(bs, head, jo, dk)
position = torch.arange(jo, dtype=torch.long).unsqueeze(0).expand(bs, jo)

ARGS = {
    "matrix_q": matrix_q,
    "matrix_k": matrix_k,
    "position": position,
}
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 tuple, ndim 5, 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
multi_head_attention Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("4b7e5c1a003881c3")
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 3, dtype family float, shape (2, 3, 32) 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)

# Construct tiny valid arguments for multi_head_attention
bs = 2  # batch size
h = 4   # number of heads
length_q = 3  # query sequence length
length_kv = 3  # key-value sequence length
key_dim = 8   # key dimension

# Create q, k, v tensors with shape (bs, h, length, key_dim)
q = torch.randn(bs, h, length_q, key_dim)
k = torch.randn(bs, h, length_kv, key_dim)
v = torch.randn(bs, h, length_kv, key_dim)

# Create mask tensor with shape (bs, length_q, length_kv)
# All True to allow all positions
mask = torch.ones(bs, length_q, length_kv, dtype=torch.bool)

# Create position tensor with shape (bs, length_q)
# Contains position indices for RoPE
position = torch.arange(length_q, dtype=torch.long).unsqueeze(0).expand(bs, -1)

# rope flag
rope = True

ARGS = {
    "q": q,
    "k": k,
    "v": v,
    "mask": mask,
    "position": position,
    "rope": rope,
}
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 3, dtype family float, shape (2, 3, 32) — shape (2, 3, 32) 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
nonzero_averaging Ran
ran, violated the contract we drafted
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("a3d2cc327e70d966")
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 2, dtype family float, shape (2, 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 torch

torch.manual_seed(0)

# Create a tensor with shape [sz_b, node_num, d]
# sz_b=2 (batch size), node_num=3 (number of nodes), d=4 (feature dimension)
# Include some zero vectors to test the nonzero_averaging logic
x = torch.tensor([
    [[1.0, 2.0, 3.0, 4.0],
     [0.0, 0.0, 0.0, 0.0],
     [5.0, 6.0, 7.0, 8.0]],
    [[2.0, 3.0, 4.0, 5.0],
     [1.0, 1.0, 1.0, 1.0],
     [0.0, 0.0, 0.0, 0.0]]
])

ARGS = {
    'x': x
}
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 2, dtype family float, shape (2, 4) — shape (2, 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
reshape_by_heads Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("36380487b1505ac6")
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 4, dtype family float, shape (2, 4, 3, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create a valid input tensor with at least 3 dimensions
# Shape: (batch=2, seq_len=3, feature_dim=8)
# The last dimension (8) must be divisible by head_num
input_tensor = torch.randn(2, 3, 8)

# head_num must divide evenly into the last dimension of input_tensor
head_num = 4

ARGS = {
    'input_tensor': input_tensor,
    'head_num': head_num
}
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 4, dtype family float, shape (2, 4, 3, 2) — shape (2, 4, 3, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
sinusoidal_position_embedding Ran
ran; our contract draft was wrong, not the code
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("04b13a6c7034789f")
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 4, dtype family float, shape (2, 8, 10, 64) 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)

# Analyze the function:
# - head: used in unsqueeze/expand, appears to be an integer for number of heads
# - dk: used in arange and division, appears to be an integer for embedding dimension
# - position: a tensor with shape (bs, jo) where bs is batch size and jo is sequence length
#   - must have a .device attribute
#   - must support .float(), .unsqueeze(), multiplication with theta

ARGS = {
    "head": 8,
    "dk": 64,
    "position": torch.randint(0, 100, (2, 10))  # batch_size=2, seq_len=10
}
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 4, dtype family float, shape (2, 8, 10, 64) — shape (2, 8, 10, 64) 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
AddAndNormalizationModule Not yet run
unverified (harvested, no recorded run)
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("1410adc07627c72d")
How it ran

no invocation recorded.

CrossAttentionBlock Not yet run
unverified (harvested, no recorded run)
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("c0a3b5b5264d59b5")
How it ran

no invocation recorded.

Model Not yet run
unverified (harvested, no recorded run)
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("ed62592b48438066")
How it ran

no invocation recorded.

multi_head_attention_with_edge Not yet run
unverified (harvested, no recorded run)
XiangjieXiao/ReSched/PPO/SchedulingModel.py
code served (permissive licence) · get_code("08223073fb865e2e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 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 13 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

Neural approaches to the Flexible Job Shop Scheduling Problem (FJSP), particularly those based on deep reinforcement learning (DRL), have gained growing attention in recent years. However, existing methods rely on complex featureengineered state representations (i.e., often requiring more than 20 handcrafted features) and graph-biased neural architectures. To reduce modeling complexity and advance a more generalizable framework for FJSP, we introduce RESCHED, a minimalist DRL framework that rethinks both the scheduling formulation and model design. First, by revisiting the Markov Decision Process (MDP) formulation of FJSP, we condense the state space to just four essential features, eliminating historical dependencies through a subproblem-based perspective. Second, we employ Transformer blocks with dot-product attention, augmented by three lightweight but effective architectural modifications tailored to scheduling tasks. Extensive experiments show that RESCHED outperforms classical dispatching rules and state-of-the-art DRL methods on FJSP. Moreover, RESCHED also generalizes well to the Job Shop Scheduling Problem (JSSP) and the Flexible Flow Shop Scheduling Problem (FFSP), achieving competitive performance against neural baselines specifically designed for these variants. Our code is available at https://github.com/XiangjieXiao/ReSched. * Corresponding author. 1 For instance, we demonstrate in Appendix B.3 using DANIEL (Wang et al., 2024b) that removing half of the input features does not compromise performance.

For agents

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

get_harvested_code_for_paper("2603.07020")
get_code_for_paper("2603.07020")
have("2603.07020")

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

curl https://syntology.ai/api/ran/2603.07020.json

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

[![Syntology run record](https://syntology.ai/api/ran/2603.07020.svg)](https://syntology.ai/paper/2603.07020)

Connect an agent — have() is free.