SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.11298 · 2022

Tele-Knowledge Pre-training for Fault Analysis

arXiv · PDF · Open in the Atlas

Code that ran

7 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 6 have no contract verdict; 0 failures of our instrument; 3 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2210.11298.json · badge: /api/ran/2210.11298.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
hackerchenzhuo/ktelebert canonical (classifier label)
classifier basis not recorded on the link
7 ran · 3 no recorded run
FunctionStatusWhere it lives
debug Ran
ran on a synthesized input
hackerchenzhuo/ktelebert/KTeleBERT/model/HWBert.py
code served (permissive licence) · get_code("82954c7f58e9d215")
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 MockBatch. 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 copy import deepcopy

torch.manual_seed(0)

class MockBatch:
    def __init__(self):
        self.input_ids = torch.randint(0, 100, (5, 4))
        self.attention_mask = torch.ones(5, 4, dtype=torch.long)
        self.chinese_ref = torch.randint(0, 50, (5, 3))
        self.kpi_ref = torch.randint(0, 50, (5, 3))
        self.labels = torch.randint(0, 10, (5, 2))

batch = MockBatch()
input_list = [batch]

ARGS = {
    "input": input_list,
    "kk": 3,
    "begin": 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 MockBatch — a MockBatch output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MockBatch 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_data Ran
ran on a synthesized input
hackerchenzhuo/ktelebert/KTeleBERT/src/data.py
code served (permissive licence) · get_code("f5c6ca1407832288")
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 torch
import random
import json
import os.path as osp
from unittest.mock import Mock, MagicMock
import tempfile
import os

torch.manual_seed(0)

# Create a temporary directory for mock data files
temp_dir = tempfile.mkdtemp()

# Create mock data files
sample_data = [
    {"text": "sample1", "id": 1},
    {"text": "sample2", "id": 2},
    {"text": "sample3", "id": 3},
    {"text": "sample4", "id": 4},
]

sample_chinese_ref = [
    {"ref": "中文1"},
    {"ref": "中文2"},
    {"ref": "中文3"},
    {"ref": "中文4"},
]

sample_kpi_ref = [
    {"kpi": "kpi1"},
    {"kpi": "kpi2"},
    {"kpi": "kpi3"},
    {"kpi": "kpi4"},
]

data_name = "test_data"

with open(osp.join(temp_dir, f'{data_name}_cws.json'), 'w') as fp:
    json.dump(sample_data, fp)

with open(osp.join(temp_dir, f'{data_name}_chinese_ref.json'), 'w') as fp:
    json.dump(sample_chinese_ref, fp)

with open(osp.join(temp_dir, f'{data_name}_kpi_ref.json'), 'w') as fp:
    json.dump(sample_kpi_ref, fp)

# Create a mock logger
logger = Mock()
logger.info = Mock()

# Create mock args object
args = Mock()
args.data_path = temp_dir
args.seq_data_name = data_name
args.rank = 0
args.train_ratio = 0.75
args.use_mlm_task = True
args.use_NumEmb = True

ARGS = {
    "logger": logger,
    "args": args,
}
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 — 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
load_order_data Ran
ran on a synthesized input
hackerchenzhuo/ktelebert/KTeleBERT/src/data.py
code served (permissive licence) · get_code("c6ee981fa5d5d2dc")
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 torch
import json
import tempfile
import os.path as osp
from unittest.mock import Mock

torch.manual_seed(0)

# Create a temporary directory and JSON data file
temp_dir = tempfile.mkdtemp()

# Create sample order data
sample_data = [
    {"order_id": i, "value": float(i * 10)}
    for i in range(100)
]

# Write the JSON file
data_file = osp.join(temp_dir, 'test_orders.json')
with open(data_file, 'w') as f:
    json.dump(sample_data, f)

# Create a mock logger
mock_logger = Mock()
mock_logger.info = Mock()

# Create mock args
mock_args = Mock()
mock_args.rank = 0
mock_args.data_path = temp_dir
mock_args.order_test_name = ''
mock_args.order_data_name = 'test_orders'
mock_args.train_ratio = 0.8

ARGS = {
    'logger': mock_logger,
    'args': mock_args
}
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 — 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
ltp_debug Ran
ran on a synthesized input
hackerchenzhuo/ktelebert/KTeleBERT/get_chinese_ref.py
code served (permissive licence) · get_code("4b31b866a812e8ed")
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 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

class MockLTP(nn.Module):
    def __init__(self):
        super().__init__()
    
    def pipeline(self, data, tasks=None):
        # Return a mock output object with a cws attribute
        return MockOutput()

class MockOutput:
    def __init__(self):
        # cws should be a list of lists where each inner list contains strings
        self.cws = [["hello"], ["world"], ["test"]]

class MockOP:
    def __init__(self):
        self.data = ["sample text one", "sample text two"]
    
    def __iter__(self):
        return iter(self.data)
    
    def __len__(self):
        return len(self.data)

ltp = MockLTP()
op = MockOP()

ARGS = {
    "ltp": ltp,
    "op": op
}
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 list, ndim 2, 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
ortho_penalty Ran
ran on a synthesized input
hackerchenzhuo/ktelebert/KTeleBERT/model/Numeric.py
code served (permissive licence) · get_code("43c5f80cd8738ea3")
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. 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, 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 a tensor t that can be matrix-multiplied with its transpose
# and compared to an identity matrix. The function uses:
# - t @ t.T (matrix multiplication)
# - torch.eye(t.shape[0]) (identity matrix based on first dimension)
# - .cuda() (moves to CUDA, which will fail on CPU-only, but we pass CPU tensor)
# - ** 2 (element-wise squaring)
# - .sum() (sum reduction)

# Create a small rectangular tensor (2x4) to test the function
t = torch.randn(2, 4, dtype=torch.float32)

ARGS = {
    "t": t
}
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
reduce_value Ran
ran, violated the contract we drafted
hackerchenzhuo/ktelebert/KTeleBERT/src/distributed_utils.py
code served (permissive licence) · get_code("4fe6de9a59664ee4")
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 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 torch
import torch.distributed as dist

torch.manual_seed(0)

# Since dist functions require actual distributed setup which we can't do in this sandbox,
# we'll create a minimal test case where distributed is not initialized.
# The function gracefully handles the non-distributed case.

# Create a small tensor as the value to reduce
value = torch.tensor([1.5, 2.5, 3.5, 4.5])

# The function signature is: reduce_value(value, average=True)
ARGS = {
    "value": value,
    "average": True
}
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 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
refresh_data Ran
ran on a synthesized input
hackerchenzhuo/ktelebert/KTeleBERT/get_chinese_ref.py
code served (permissive licence) · get_code("015da2ce3f106d89")
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
from collections import Counter

# Construct arguments for refresh_data function

# ref: list of strings, each string contains space-separated tokens
ref = [
    "apple banana cherry apple",
    "banana date apple fig banana",
    "cherry fig grape apple cherry",
    "date grape apple banana date",
    "elephant apple banana cherry elephant elephant"
]

# freq: minimum frequency threshold for words to be included
freq = 2

# special_token: list of special tokens to add
special_token = ["<PAD>", "<UNK>", "<CLS>", "<SEP>"]

ARGS = {
    "ref": ref,
    "freq": freq,
    "special_token": special_token
}
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 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
LLRD Not yet run
unverified (harvested, no recorded run)
hackerchenzhuo/ktelebert/KTeleBERT/src/utils.py
code served (permissive licence) · get_code("b945e1b921c61e2c")
How it ran

no invocation recorded.

add_special_token Not yet run
unverified (harvested, no recorded run)
hackerchenzhuo/ktelebert/KTeleBERT/src/utils.py
code served (permissive licence) · get_code("c316d7769642f082")
How it ran

no invocation recorded.

load_data_kg Not yet run
unverified (harvested, no recorded run)
hackerchenzhuo/ktelebert/KTeleBERT/src/data.py
code served (permissive licence) · get_code("9b97761b751593d2")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 9 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
reduce_value 4 other papers
1 dated earlier, 3 later
2104.11493 (2021-04)
Stroke-Based Scene Text Erasing Using Synthetic Data for Training
2212.14454 (2022-12)
MEAformer: Multi-modal Entity Alignment Transformer for Meta Modality Hybrid
2301.11259 (2023-01)
Domain-Agnostic Molecular Generation with Chemical Feedback
and 1 more not named here

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

In this work, we share our experience on tele-knowledge pre-training for fault analysis, a crucial task in telecommunication applications that requires a wide range of knowledge normally found in both machine log data and product documents. To organize this knowledge from experts uniformly, we propose to create a Tele-KG (tele-knowledge graph). Using this valuable data, we further propose a tele-domain language pre-training model TeleBERT and its knowledge-enhanced version, a tele-knowledge re-training model KTeleBERT. which includes effective prompt hints, adaptive numerical data encoding, and two knowledge injection paradigms. Concretely, our proposal includes two stages: first, pre-training TeleBERT on 20 million tele-related corpora, and then re-training it on 1 million causal and machine-related corpora to obtain KTeleBERT. Our evaluation on multiple tasks related to fault analysis in tele-applications, including root-cause analysis, event association prediction, and fault chain tracing, shows that pre-training a language model with tele-domain data is beneficial for downstream tasks. Moreover, the KTeleBERT re-training further improves the performance of task models, highlighting the effectiveness of incorporating diverse tele-knowledge into the model.

For agents

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

get_harvested_code_for_paper("2210.11298")
get_code_for_paper("2210.11298")
have("2210.11298")

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

curl https://syntology.ai/api/ran/2210.11298.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.11298.svg)](https://syntology.ai/paper/2210.11298)

Connect an agent — have() is free.