SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.05751 · ACL Findings · 2022

Decomposed Meta-Learning for Few-Shot Named Entity Recognition

Huiqiang Jiang, Qianhui Wu, Chin-Yew Lin, Tiejun Zhao, Tingting Ma

arXiv · PDF · Open in the Atlas

Code that ran

3 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 1 failure of our instrument; 2 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/2204.05751.json · badge: /api/ran/2204.05751.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
microsoft/vert-papers community
Papers with Code archive
3 ran · 2 no recorded run
FunctionStatusWhere it lives
LinearCRF Ran
ran on a synthesized input · constructed an object; no result was computed
microsoft/vert-papers/papers/DecomposedMetaSL/model/joint_model.py
code served (permissive licence) · get_code("292da879de71f451")
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, LinearCRF: 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
from torch import nn

torch.manual_seed(0)

# Create a minimal label2idx dictionary for BIO schema
# We need at least a few labels to make the schema work
label2idx = {
    'O': 0,
    'B-PER': 1,
    'I-PER': 2,
    'B-LOC': 3,
    'I-LOC': 4,
}

tag_size = len(label2idx)
device = torch.device('cpu')
schema = 'BIO'
add_constraint = True

ARGS = {
    'tag_size': tag_size,
    'schema': schema,
    'add_constraint': add_constraint,
    'label2idx': label2idx,
    'device': device,
}
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 LinearCRF — 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 LinearCRF 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
MaxLoss Ran
ran on a synthesized input
microsoft/vert-papers/papers/DecomposedMetaSL/model/joint_model.py
code served (permissive licence) · get_code("76f3f1f4704f841d")
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 0, 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 torch import nn

class MaxLoss(nn.Module):

    def __init__(self, gamma=1.0):
        super(MaxLoss, self).__init__()
        self.gamma = gamma
        self.loss_fct = nn.CrossEntropyLoss(ignore_index=-1, reduction='none')
        return

    def forward(self, logits, targets):
        assert logits.dim() == 3
        batch_size, seq_len, class_num = logits.size()
        token_loss = self.loss_fct(logits.view(-1, class_num), targets.view(-1)).view(batch_size, seq_len)
        act_pos = targets.ne(-1)
        loss = token_loss[act_pos].mean()
        if self.gamma > 0:
            max_loss = torch.max(token_loss, dim=1)[0]
            loss += self.gamma * max_loss.mean()
        return loss


INIT_ARGS = {
    'gamma': 0.5
}

ARGS = {
    'logits': torch.randn(2, 4, 8),
    'targets': torch.tensor([[0, 1, 2, -1], [1, 0, -1, 3]])
}

if __name__ == '__main__':
    model = MaxLoss(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Loss output: {output}")
    print(f"Output shape: {output.shape}")
    print(f"Output is scalar: {output.dim() == 0}")
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 0, 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)
convert_bio2spans Ran
ran; our contract draft was wrong, not the code
microsoft/vert-papers/papers/DecomposedMetaSL/model/joint_model.py
code served (permissive licence) · get_code("2b9f711565f0a0f5")
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

# Example BIO tags
tags_bio = ['B-PER', 'I-PER', 'O', 'B-LOC', 'I-LOC']

# Example IO tags
tags_io = ['I-PER', 'I-PER', 'O', 'I-LOC', 'I-LOC']

# Example BIOES tags
tags_bioes = ['B-PER', 'I-PER', 'E-PER', 'O', 'S-LOC']

ARGS = {
    'tags': tags_bio,
    'schema': 'BIO'
}
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
FewShotSpanModel Not yet run
unverified (harvested, no recorded run)
microsoft/vert-papers/papers/DecomposedMetaSL/model/joint_model.py
code served (permissive licence) · get_code("539d381a77a39276")
How it ran

no invocation recorded.

SelectedJointModel Not yet run
unverified (harvested, no recorded run)
microsoft/vert-papers/papers/DecomposedMetaSL/model/joint_model.py
code served (permissive licence) · get_code("3ada8f03e3ed466b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 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 5 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

Few-shot named entity recognition (NER) systems aim at recognizing novel-class named entities based on only a few labeled examples. In this paper, we present a decomposed metalearning approach which addresses the problem of few-shot NER by sequentially tackling fewshot span detection and few-shot entity typing using meta-learning. In particular, we take the few-shot span detection as a sequence labeling problem and train the span detector by introducing the model-agnostic meta-learning (MAML) algorithm to find a good model parameter initialization that could fast adapt to new entity classes. For few-shot entity typing, we propose MAML-ProtoNet, i.e., MAML-enhanced prototypical networks to find a good embedding space that can better distinguish text span representations from different entity classes. Extensive experiments on various benchmarks show that our approach achieves superior performance over prior methods. 1

For agents

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

get_harvested_code_for_paper("2204.05751")
get_code_for_paper("2204.05751")
have("2204.05751")

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

curl https://syntology.ai/api/ran/2204.05751.json

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

[![Syntology run record](https://syntology.ai/api/ran/2204.05751.svg)](https://syntology.ai/paper/2204.05751)

Connect an agent — have() is free.