SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2009.12005 · EMNLP · 2020

MinTL: Minimalist Transfer Learning for Task-Oriented Dialogue Systems

Andrea Madotto, Zhaojiang Lin, Indra Genta, Fung, Pascale Winata

arXiv · PDF · Open in the Atlas

Code that ran

7 of 9 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; 4 failures of our instrument; 2 no recorded run), measured 2026-11-21; 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/2009.12005.json · badge: /api/ran/2009.12005.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
zlinao/MinTL canonical (classifier label)
classifier basis: census
7 ran · 2 no recorded run
FunctionStatusWhere it lives
average_distributed_scalar Ran
ran, honoured the contract we drafted
zlinao/MinTL/pretraining.py
pointer only (licence: MIT); the stored copy get_code serves, from dladustn95/ENMODEL, is pointer only (licence: NONE) · get_code("ebb24f4dc2a81286")
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 dladustn95/ENMODEL, 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.

clean_time Ran
ran; our contract draft was wrong, not the code
zlinao/MinTL/damd_multiwoz/clean_dataset.py
code served (permissive licence) · get_code("4c62f23ce8135bda")
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.

import re

def clean_time(utter):
    utter = re.sub('(\\d+) ([ap]\\.?m)', lambda x: x.group(1) + x.group(2), utter)
    utter = re.sub('((?<!\\d)\\d:\\d+)(am)?', '0\\1', utter)
    utter = re.sub('((?<!\\d)\\d)am', '0\\1:00', utter)
    utter = re.sub('((?<!\\d)\\d)pm', lambda x: str(int(x.group(1)) + 12) + ':00', utter)
    utter = re.sub('(\\d+)(:\\d+)pm', lambda x: str(int(x.group(1)) + 12) + x.group(2), utter)
    utter = re.sub('(\\d+)a\\.?m', '\\1', utter)
    return utter

ARGS = {
    'utter': 'The meeting is at 3 p.m. tomorrow at 2:30am'
}
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
invert_mask Ran
ran; our contract draft was wrong, not the code
zlinao/MinTL/BART.py
code served (permissive licence) · get_code("c9f6adbecd206460")
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)

# The function expects a 2D attention_mask tensor
# It calls .dim() to check it's 2D, then .eq(0) to create a boolean mask
attention_mask = torch.tensor([[1, 1, 0, 0],
                               [1, 0, 1, 1]], dtype=torch.long)

ARGS = {
    "attention_mask": attention_mask
}
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
make_padding_mask Ran
ran; our contract draft was wrong, not the code
zlinao/MinTL/BART.py
pointer only (licence: MIT); the stored copy get_code serves, from yeliu918/KG-BART, is pointer only (licence: NONE) · get_code("f20c9b1aead27fea")
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 yeliu918/KG-BART, 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.

padOutput Ran
ran on a synthesized input
zlinao/MinTL/pretraining.py
code served (permissive licence) · get_code("82241a05bdff196d")
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 3, dtype family float; no output shape is stated. It ran in syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1) at 2026-09-27T00:55:33Z. 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 numpy as np

# Create sequences as lists of arrays/values
# The function expects sequences to be indexable and have len()
seq1 = np.array([1, 2, 3, 4, 5, 6])
seq2 = np.array([10, 20, 30])
seq3 = np.array([100, 200, 300, 400, 500])

sequences = [seq1, seq2, seq3]

ARGS = {
    "sequences": sequences,
    "max_len": 400,
    "pad_token": 0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 3, 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 syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1); 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
puntuation_handler Ran
ran on a synthesized input
zlinao/MinTL/utils.py
code served (permissive licence) · get_code("a9b8c7a10994073f")
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.

import torch

torch.manual_seed(0)

# The function takes a single parameter: text (a string)
# It performs various string replacements on punctuation
# No complex objects needed, just a simple string

text_input = "Hello, world! What's happening? This is a test: yes."

ARGS = {
    "text": text_input
}
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 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
shift_tokens_right Ran
ran; our contract draft was wrong, not the code
zlinao/MinTL/BART.py
pointer only (licence: MIT); the stored copy get_code serves, from yeliu918/KG-BART, is pointer only (licence: NONE) · get_code("e00ccf7847d12b59")
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 yeliu918/KG-BART, 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.

clean_text Not yet run
unverified (harvested, no recorded run)
zlinao/MinTL/damd_multiwoz/clean_dataset.py
code served (permissive licence) · get_code("5b9d615f09edd9c8")
How it ran

no invocation recorded.

random_spans_noise_mask Not yet run
unverified (harvested, no recorded run)
zlinao/MinTL/pretraining.py
code served (permissive licence) · get_code("01cc69c1c0585dba")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 6 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 3 the same month, 6 later) and 3 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
shift_tokens_right 11 other papers
1 dated earlier, 1 the same month, 9 later
1910.13461 (2019-10)
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
2009.12677 (2020-09)
KG-BART: Knowledge Graph-Augmented BART for Generative Commonsense Reasoning
2021.acl-long.390 (2021 (year only))
Few-Shot Text Ranking with Meta Adapted Synthetic Weak Supervision
and 8 more not named here
invert_mask 10 other papers
1 dated earlier, 1 the same month, 8 later
1910.13461 (2019-10)
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
2009.12677 (2020-09)
KG-BART: Knowledge Graph-Augmented BART for Generative Commonsense Reasoning
2021.acl-long.390 (2021 (year only))
Few-Shot Text Ranking with Meta Adapted Synthetic Weak Supervision
and 7 more not named here
make_padding_mask 9 other papers
1 dated earlier, 1 the same month, 7 later
1910.13461 (2019-10)
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
2009.12677 (2020-09)
KG-BART: Knowledge Graph-Augmented BART for Generative Commonsense Reasoning
2021.acl-long.390 (2021 (year only))
Few-Shot Text Ranking with Meta Adapted Synthetic Weak Supervision
and 6 more not named here
clean_time 7 other papers
7 later
2021.findings-emnlp.112 (2021 (year only))
Improving End-to-End Task-Oriented Dialogue System with A Simple Auxiliary Task
2109.14739 (2021-09)
Multi-Task Pre-Training for Plug-and-Play Task-Oriented Dialogue System
2204.00763 (2022-04)
Metaphorical User Simulators for Evaluating Task-oriented Dialogue Systems
and 4 more not named here
average_distributed_scalar 3 other papers
2 dated earlier, 1 later
1901.08149 (2019-01)
TransferTransfo: A Transfer Learning Approach for Neural Network Based Conversational Agents
2008.09075 (2020-08)
Controlling Dialogue Generation with Semantic Exemplars
2310.15896 (2023-10)
BianQue: Balancing the Questioning and Suggestion Ability of Health LLMs with Multi-turn Health Conversations Polished by ChatGPT
puntuation_handler 1 other paper
1 later
2302.10342 (2023-02)
Fantastic Rewards and How to Tame Them: A Case Study on Reward Learning for Task-oriented Dialogue Systems

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 2026-11-21.

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 paper, we propose Minimalist Transfer Learning (MinTL) to simplify the system design process of task-oriented dialogue systems and alleviate the over-dependency on annotated data. MinTL is a simple yet effective transfer learning framework, which allows us to plug-and-play pre-trained seq2seq models, and jointly learn dialogue state tracking and dialogue response generation. Unlike previous approaches, which use a copy mechanism to "carryover" the old dialogue states to the new one, we introduce Levenshtein belief spans (Lev), that allows efficient dialogue state tracking with a minimal generation length. We instantiate our learning framework with two pretrained backbones: T5 (Raffel et al., 2019) and BART (Lewis et al., 2019), and evaluate them on MultiWOZ. Extensive experiments demonstrate that: 1) our systems establish new state-of-the-art results on end-to-end response generation, 2) MinTL-based systems are more robust than baseline methods in the low resource setting, and they achieve competitive results with only 20% training data, and 3) Lev greatly improves the inference efficiency 1 .

For agents

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

get_harvested_code_for_paper("2009.12005")
get_code_for_paper("2009.12005")
have("2009.12005")

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

curl https://syntology.ai/api/ran/2009.12005.json

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

[![Syntology run record](https://syntology.ai/api/ran/2009.12005.svg)](https://syntology.ai/paper/2009.12005)

Connect an agent — have() is free.