SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2110.06274 · 2021

LiST: Lite Prompted Self-training Makes Parameter-Efficient Few-shot Learners

arXiv · PDF · Open in the Atlas

Code that ran

9 of 12 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; 2 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 5 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/2110.06274.json · badge: /api/ran/2110.06274.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/list canonical (classifier label)
classifier basis not recorded on the link
9 ran · 3 no recorded run
FunctionStatusWhere it lives
default_dev_objective Ran
ran, honoured the contract we drafted
microsoft/list/src/trainer.py
pointer only (licence: MIT); the stored copy get_code serves, from uijeongjang/lora-ntk, is pointer only (licence: NONE) · get_code("e448786d5ff1875a")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from uijeongjang/lora-ntk, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

entropy Ran
ran on a synthesized input
microsoft/list/src/loss.py
code served (permissive licence) · get_code("3e6d40262bdef1d5")
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. 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.functional as F

torch.manual_seed(0)

# Create a small logit tensor (batch_size=2, num_classes=4)
logit = torch.randn(2, 4)

# Define ARGS dict with all parameters
ARGS = {
    "logit": logit,
    "epsilon": 1e-06,
    "reduce": 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 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
generate_noise Ran
ran, honoured the contract we drafted
microsoft/list/src/models.py
code served (permissive licence) · get_code("417ac95acb8c20b4")
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
import torch.nn as nn

torch.manual_seed(0)

# Create a small embedding tensor
# The function calls embed.data.new(embed.size()) and embed.data
# So embed needs to be a tensor with .data attribute
embed = torch.randn(2, 4, requires_grad=True)

# Create a mask tensor - used by the function indirectly if passed
# Keep it small, same shape as embed or compatible
mask = torch.ones(2, 4)

# epsilon is a float parameter with default value
epsilon = 1e-05

ARGS = {
    "embed": embed,
    "mask": mask,
    "epsilon": epsilon,
}
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
input_example_to_string Ran
ran; our contract draft was wrong, not the code
microsoft/list/src/dataset.py
code served (permissive licence) · get_code("e768e25ca95f70cd")
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 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.

class Example:
    def __init__(self, text_a, text_b=None):
        self.text_a = text_a
        self.text_b = text_b

example_with_text_b = Example("Hello world", "This is text b")
example_without_text_b = Example("Hello world", None)

ARGS = {
    "example": example_with_text_b,
    "sep_token": "[SEP]"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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
input_example_to_tuple Ran
ran; our contract draft was wrong, not the code
microsoft/list/src/dataset.py
code served (permissive licence) · get_code("fbef70dc7c700501")
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 logging
import pandas as pd

logger = logging.getLogger(__name__)

# Create a simple example class that mimics the structure needed
class Example:
    def __init__(self, text_a, text_b=None):
        self.text_a = text_a
        self.text_b = text_b

# Test case 1: example with both text_a and text_b
example_both = Example(text_a="Hello world", text_b="Goodbye world")

# Test case 2: example with only text_a
example_single = Example(text_a="Hello world", text_b=None)

# Test case 3: example with None text_a and None text_b
example_empty = Example(text_a=None, text_b=None)

# Use the most complete example for the ARGS dict
ARGS = {
    "example": example_both
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — 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
norm_grad Ran
ran on a synthesized input
microsoft/list/src/models.py
code served (permissive licence) · get_code("b9836599a17cd3c0")
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 4, 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)

# Create small tensors for grad and eff_grad
# grad shape: (batch=2, seq=3, dim=4) to test both sentence_level and token_level
grad = torch.randn(2, 3, 4)
eff_grad = torch.randn(2, 3, 4)

ARGS = {
    'grad': grad,
    'eff_grad': eff_grad,
    'sentence_level': False,
    'norm_p': 'max',
    'epsilon': 1e-05
}
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 4, 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
select_neighbors Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
microsoft/list/src/label_search.py
pointer only (licence: MIT); the stored copy get_code serves, from korawat-tanwisuth/pouf, is pointer only (licence: NONE) · get_code("b69b1ad29fd21db5")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from korawat-tanwisuth/pouf, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

stable_kl Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
microsoft/list/src/loss.py
code served (permissive licence) · get_code("7b0111630cee9629")
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. 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.functional as F

torch.manual_seed(0)

# Create small tensors for logit and target
# Both need to be 2D after view(-1, logit.size(-1))
# Let's use batch_size=2, num_classes=4
logit = torch.randn(2, 4)
target = torch.randn(2, 4)

ARGS = {
    'logit': logit,
    'target': target,
    'epsilon': 1e-06,
    'reduce': 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 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
text_classification_metrics Ran
ran; our contract draft was wrong, not the code
microsoft/list/src/processors.py
code served (permissive licence) · get_code("b62076c5483738dc")
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 dict. 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)

# Construct arguments for text_classification_metrics
task_name = "sentiment"

# Create predictions and labels as numpy arrays
# The function compares them with == and calls .mean()
preds = np.array([0, 1, 0, 1], dtype=np.int64)
labels = np.array([0, 1, 1, 1], dtype=np.int64)

ARGS = {
    "task_name": task_name,
    "preds": preds,
    "labels": labels,
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
eval_pairing_acc Not yet run
unverified (harvested, no recorded run)
microsoft/list/src/label_search.py
pointer only (licence: MIT); the stored copy get_code serves, from korawat-tanwisuth/pouf, is pointer only (licence: NONE) · get_code("47923f36537f5f79")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from korawat-tanwisuth/pouf, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

select_likely_words Not yet run
unverified (harvested, no recorded run)
microsoft/list/src/label_search.py
pointer only (licence: MIT); the stored copy get_code serves, from korawat-tanwisuth/pouf, is pointer only (licence: NONE) · get_code("be6b57168b0424f3")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from korawat-tanwisuth/pouf, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

tokenize_multipart_input Not yet run
unverified (harvested, no recorded run)
microsoft/list/src/dataset.py
code served (permissive licence) · get_code("71af7b8adce2cf48")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 11 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 4 the same month, 11 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
input_example_to_string 9 other papers
1 the same month, 8 later
2110.01256 (2021-10)
Revisiting Self-Training for Few-Shot Learning of Language Model
2202.11451 (2022-02)
Zero-shot Cross-lingual Transfer of Prompt-based Tuning with a Unified Multilingual Prompt
2205.01308 (2022-05)
Contrastive Learning for Prompt-Based Few-Shot Language Learners
and 6 more not named here
input_example_to_tuple 9 other papers
1 the same month, 8 later
2110.01256 (2021-10)
Revisiting Self-Training for Few-Shot Learning of Language Model
2202.11451 (2022-02)
Zero-shot Cross-lingual Transfer of Prompt-based Tuning with a Unified Multilingual Prompt
2205.01308 (2022-05)
Contrastive Learning for Prompt-Based Few-Shot Language Learners
and 6 more not named here
default_dev_objective 7 other papers
1 the same month, 6 later
2110.01256 (2021-10)
Revisiting Self-Training for Few-Shot Learning of Language Model
2202.11451 (2022-02)
Zero-shot Cross-lingual Transfer of Prompt-based Tuning with a Unified Multilingual Prompt
2205.01308 (2022-05)
Contrastive Learning for Prompt-Based Few-Shot Language Learners
and 4 more not named here
text_classification_metrics 6 other papers
1 the same month, 5 later
2110.01256 (2021-10)
Revisiting Self-Training for Few-Shot Learning of Language Model
2205.01308 (2022-05)
Contrastive Learning for Prompt-Based Few-Shot Language Learners
2210.05643 (2022-10)
A Kernel-Based View of Language Model Fine-Tuning
and 3 more not named here
generate_noise 3 other papers
3 later
2204.06625 (2022-04)
CAMERO: Consistency Regularized Ensemble of Perturbed Language Models with Weight Sharing
2212.09994 (2022-12)
Towards Robustness of Text-to-SQL Models Against Natural and Realistic Adversarial Table Perturbation
2023.findings-emnlp.528 (2023 (year only))
Uncertainty-aware Parameter-Efficient Self-training for Semi-supervised Language Understanding
stable_kl 3 other papers
3 later · 1 only as an assembly brick
2204.06625 (2022-04)
CAMERO: Consistency Regularized Ensemble of Perturbed Language Models with Weight Sharing
2212.09994 (2022-12)
Towards Robustness of Text-to-SQL Models Against Natural and Realistic Adversarial Table Perturbation
2023.findings-emnlp.528 (2023 (year only))
Uncertainty-aware Parameter-Efficient Self-training for Semi-supervised Language Understanding
eval_pairing_acc 2 other papers
2 later
2205.01308 (2022-05)
Contrastive Learning for Prompt-Based Few-Shot Language Learners
2305.00350 (2023-05)
POUF: Prompt-oriented unsupervised fine-tuning for large pre-trained models
select_likely_words 2 other papers
2 later
2205.01308 (2022-05)
Contrastive Learning for Prompt-Based Few-Shot Language Learners
2305.00350 (2023-05)
POUF: Prompt-oriented unsupervised fine-tuning for large pre-trained models
select_neighbors 2 other papers
2 later
2205.01308 (2022-05)
Contrastive Learning for Prompt-Based Few-Shot Language Learners
2305.00350 (2023-05)
POUF: Prompt-oriented unsupervised fine-tuning for large pre-trained models
entropy 1 other paper
1 later
2023.findings-emnlp.528 (2023 (year only))
Uncertainty-aware Parameter-Efficient Self-training for Semi-supervised Language Understanding
norm_grad 1 other paper
1 later
2023.findings-emnlp.528 (2023 (year only))
Uncertainty-aware Parameter-Efficient Self-training for Semi-supervised Language Understanding

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

We present a new method LiST is short for Lite Prompted Self-Training for parameter-efficient fine-tuning of large pre-trained language models (PLMs) for few-shot learning. LiST improves over recent methods that adopt prompt-based fine-tuning (FN) using two key techniques. The first is the use of self-training to leverage large amounts of unlabeled data for prompt-based FN in few-shot settings. We use self-training in conjunction with meta-learning for re-weighting noisy pseudo-prompt labels. Self-training is expensive as it requires updating all the model parameters repetitively. Therefore, we use a second technique for light-weight fine-tuning where we introduce a small number of task-specific parameters that are fine-tuned during self-training while keeping the PLM encoder frozen. Our experiments show that LiST can effectively leverage unlabeled data to improve the model performance for few-shot learning. Additionally, the fine-tuning is efficient as it only updates a small percentage of parameters and the overall model footprint is reduced since several tasks can share a common PLM encoder as backbone. A comprehensive study on six NLU tasks demonstrate LiST to improve by 35% over classic fine-tuning and 6% over prompt-based FN with 96% reduction in number of trainable parameters when fine-tuned with no more than 30 labeled examples from each task. With only 14M tunable parameters, LiST outperforms GPT-3 in-context learning by 33% on few-shot NLU tasks.

For agents

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

get_harvested_code_for_paper("2110.06274")
get_code_for_paper("2110.06274")
have("2110.06274")

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

curl https://syntology.ai/api/ran/2110.06274.json

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

[![Syntology run record](https://syntology.ai/api/ran/2110.06274.svg)](https://syntology.ai/paper/2110.06274)

Connect an agent — have() is free.