SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.02387 · 2020

Few-shot Relation Extraction via Bayesian Meta-learning on Relation Graphs

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 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, 0 violated it and 2 have no contract verdict; 1 failure of our instrument; 0 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/2007.02387.json · badge: /api/ran/2007.02387.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
DeepGraphLearning/FewShotRE community (classifier label)
classifier basis: plan_row
4 ran · 0 no recorded run
FunctionStatusWhere it lives
collate_fn Ran
ran on a synthesized input
DeepGraphLearning/FewShotRE/fewshot_re_kit/data_loader.py
code served (permissive licence) · get_code("2f8315238ca03c2c")
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 torch.utils.data as data

torch.manual_seed(0)

# Create sample data items for the collate function
# Each item is a tuple of (support_set, query_set, query_labels)

# A support/query set has structure: {'word': [...], 'pos': [...], 'mask': [...], 'rel': [...]}
# where 'word', 'pos', 'mask' are tensors and 'rel' can be lists

# Create first sample
support_set_1 = {
    'word': [torch.randint(0, 100, (3, 4)), torch.randint(0, 100, (3, 4))],
    'pos': [torch.randint(0, 20, (3, 4)), torch.randint(0, 20, (3, 4))],
    'mask': [torch.ones(3, 4), torch.ones(3, 4)],
    'rel': [[1, 2], [3, 4]]
}

query_set_1 = {
    'word': [torch.randint(0, 100, (2, 4)), torch.randint(0, 100, (2, 4))],
    'pos': [torch.randint(0, 20, (2, 4)), torch.randint(0, 20, (2, 4))],
    'mask': [torch.ones(2, 4), torch.ones(2, 4)],
    'rel': [[5], [6]]
}

query_labels_1 = [0, 1]

sample_1 = (support_set_1, query_set_1, query_labels_1)

# Create second sample
support_set_2 = {
    'word': [torch.randint(0, 100, (3, 4))],
    'pos': [torch.randint(0, 20, (3, 4))],
    'mask': [torch.ones(3, 4)],
    'rel': [[7, 8, 9]]
}

query_set_2 = {
    'word': [torch.randint(0, 100, (2, 4))],
    'pos': [torch.randint(0, 20, (2, 4))],
    'mask': [torch.ones(2, 4)],
    'rel': [[10]]
}

query_labels_2 = [1]

sample_2 = (support_set_2, query_set_2, query_labels_2)

# Create a simple dataset-like list of samples
data_list = [sample_1, sample_2]

ARGS = {
    'data': data_list
}
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
collate_fn_pair Ran
ran on a synthesized input
DeepGraphLearning/FewShotRE/fewshot_re_kit/data_loader.py
code served (permissive licence) · get_code("04c039c4b9372fac")
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 torch.utils.data as data

torch.manual_seed(0)

# The function expects data to be an iterable of tuples: (fusion_sets, query_labels)
# fusion_sets is a dict with keys 'word', 'seg', 'mask' that map to lists of tensors
# query_labels is a list of labels

# Create sample data items
def create_sample_item():
    fusion_set = {
        'word': [torch.randn(4, 3), torch.randn(4, 3)],  # list of tensors
        'seg': [torch.randn(4, 3), torch.randn(4, 3)],
        'mask': [torch.randn(4, 3), torch.randn(4, 3)]
    }
    query_labels = [0, 1]  # list of labels
    return (fusion_set, query_labels)

# Create a dataset with 2 samples
sample_data = [create_sample_item() for _ in range(2)]

ARGS = {
    'data': sample_data
}
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
get_loader Ran
ran; our contract draft was wrong, not the code
DeepGraphLearning/FewShotRE/fewshot_re_kit/data_loader.py
code served (permissive licence) · get_code("5ce7216c1f8b98e4")
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 _SingleProcessDataLoaderIter. 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.utils.data as data
import os
import json
import tempfile

torch.manual_seed(0)

# Create a mock encoder that implements the tokenize method
class MockEncoder:
    def tokenize(self, tokens, h_pos, t_pos):
        # Return (word_ids, pos_ids, mask)
        word = [1, 2, 3, 4]  # token IDs
        pos = [0, 1, 2, 3]   # position IDs
        mask = [1, 1, 1, 1]  # attention mask
        return word, pos, mask

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

# Create mock FewRel data
mock_data = {
    "relation_1": [
        {
            "tokens": ["word1", "word2", "word3", "word4"],
            "h": [0, 1, [0]],  # head entity info
            "t": [2, 3, [1]]   # tail entity info
        } for _ in range(10)
    ],
    "relation_2": [
        {
            "tokens": ["word1", "word2", "word3", "word4"],
            "h": [0, 1, [0]],
            "t": [2, 3, [1]]
        } for _ in range(10)
    ],
    "relation_3": [
        {
            "tokens": ["word1", "word2", "word3", "word4"],
            "h": [0, 1, [0]],
            "t": [2, 3, [1]]
        } for _ in range(10)
    ],
}

# Write to JSON file
data_file = os.path.join(temp_dir, "test_data.json")
with open(data_file, 'w') as f:
    json.dump(mock_data, f)

# Define ARGS dictionary with all parameters for get_loader
ARGS = {
    "name": "test_data",
    "encoder": MockEncoder(),
    "N": 2,  # number of classes
    "K": 2,  # number of support examples per class
    "Q": 2,  # number of query examples per class
    "batch_size": 2,
    "num_workers": 0,  # Use 0 to avoid multiprocessing in tests
    "collate_fn": None,  # Will use default from function signature
    "na_rate": 0.0,
    "root": temp_dir
}
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 _SingleProcessDataLoaderIter — a _SingleProcessDataLoaderIter output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _SingleProcessDataLoaderIter 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
warmup_linear Ran
ran, honoured the contract we drafted
DeepGraphLearning/FewShotRE/fewshot_re_kit/framework.py
code served (permissive licence) · get_code("8570f28ccf82469e")
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

torch.manual_seed(0)

# The function warmup_linear takes two parameters:
# - global_step: an integer representing current step number
# - warmup_step: an integer representing the warmup threshold step
# 
# The function uses only basic integer comparison and division,
# so we just need integer values.

ARGS = {
    "global_step": 5,
    "warmup_step": 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 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

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 3 are also present in other papers whose code we hold (1 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
warmup_linear 2 other papers
1 dated earlier, 1 later
1910.07124 (2019-10)
FewRel 2.0: Towards More Challenging Few-Shot Relation Classification
2105.07464 (2021-05)
FEW-NERD: A Few-shot Named Entity Recognition Dataset
collate_fn_pair 1 other paper
1 later
2023.findings-emnlp.153 (2023 (year only))
Chain of Thought with Explicit Evidence Reasoning for Few-shot Relation Extraction
get_loader 1 other paper
1 later
2023.findings-emnlp.153 (2023 (year only))
Chain of Thought with Explicit Evidence Reasoning for Few-shot Relation Extraction

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

This paper studies few-shot relation extraction, which aims at predicting the relation for a pair of entities in a sentence by training with a few labeled examples in each relation. To more effectively generalize to new relations, in this paper we study the relationships between different relations and propose to leverage a global relation graph. We propose a novel Bayesian meta-learning approach to effectively learn the posterior distribution of the prototype vectors of relations, where the initial prior of the prototype vectors is parameterized with a graph neural network on the global relation graph. Moreover, to effectively optimize the posterior distribution of the prototype vectors, we propose to use the stochastic gradient Langevin dynamics, which is related to the MAML algorithm but is able to handle the uncertainty of the prototype vectors. The whole framework can be effectively and efficiently optimized in an end-to-end fashion. Experiments on two benchmark datasets prove the effectiveness of our proposed approach against competitive baselines in both the few-shot and zero-shot settings.

For agents

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

get_harvested_code_for_paper("2007.02387")
get_code_for_paper("2007.02387")
have("2007.02387")

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

curl https://syntology.ai/api/ran/2007.02387.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.02387.svg)](https://syntology.ai/paper/2007.02387)

Connect an agent — have() is free.