SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2009.08115 · EMNLP · 2020

A Probabilistic End-To-End Task-Oriented Dialog Model with Latent Belief States towards Semi-Supervised Learning

Yichi Zhang, Min Hu, Zhijian Ou, Junlan Feng

arXiv · PDF · Open in the Atlas

Code that ran

4 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 3 have no contract verdict; 0 failures of our instrument; 6 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/2009.08115.json · badge: /api/ran/2009.08115.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
thu-spmi/LABES marked official in Papers with Code
Papers with Code archive
4 ran · 6 no recorded run
FunctionStatusWhere it lives
BeamState Ran
ran on a synthesized input · constructed an object; no result was computed
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("e928e03af1929603")
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, BeamState: 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
import copy

torch.manual_seed(0)

# Construct arguments for BeamState.__init__
score = -2.5  # sum of log-probabilities (float)

# last_hidden: a small tensor representing hidden state
last_hidden = torch.randn(1, 4)  # shape [batch=1, hidden_dim=4]

# decoded: list of Variable[1x1] of all decoded words
# Using small tensors to represent decoded tokens
decoded = [torch.tensor([[5]]), torch.tensor([[10]])]  # list of 1x1 tensors

# length: current decoded sentence length
length = 2

ARGS = {
    "score": score,
    "last_hidden": last_hidden,
    "decoded": decoded,
    "length": length
}
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 BeamState — 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 BeamState 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
Encoder Ran
ran on a synthesized input · constructed an object; no result was computed
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("8da4276bc72ed659")
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, Encoder: 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
import torch.nn as nn

torch.manual_seed(0)

# Create a small embedding layer
vocab_size = 10
embed_size = 4
embedding = nn.Embedding(vocab_size, embed_size)

# Define arguments for Encoder
ARGS = {
    'embedding': embedding,
    'input_size': 10,
    'embed_size': 4,
    'hidden_size': 3,
    'n_layers': 2,
    'dropout': 0.5
}
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 Encoder — 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 Encoder 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
MultinomialKLDivergenceLoss Ran
ran on a synthesized input
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("ef419c77fa0af064")
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 MultinomialKLDivergenceLoss(nn.Module):

    def __init__(self):
        super().__init__()

    def forward(self, p_proba, q_proba):
        loss = q_proba * (torch.log(q_proba) - torch.log(p_proba))
        return torch.sum(loss, dim=-1).mean()


INIT_ARGS = {}

ARGS = {
    'p_proba': torch.tensor([[0.2, 0.3, 0.5], [0.1, 0.4, 0.5]], dtype=torch.float32),
    'q_proba': torch.tensor([[0.25, 0.25, 0.5], [0.15, 0.35, 0.5]], dtype=torch.float32),
}

if __name__ == '__main__':
    model = MultinomialKLDivergenceLoss(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output: {output}")
    print(f"Output shape: {output.shape}")
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)
toss_ Ran
ran, violated the contract we drafted
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("3bcbe2a9cc88b480")
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 bool, 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 json
import random
import copy

random.seed(0)

# The function toss_(p) takes a single parameter p
# It uses p as the upper bound in random.randint(0, 99) <= p
# So p should be an integer representing a probability percentile (0-100)

ARGS = {
    "p": 50
}
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 bool, 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
BLEUScorer Not yet run
unverified (harvested, no recorded run)
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("e3e9fea7b52bb054")
How it ran

no invocation recorded.

BaseModel Not yet run
unverified (harvested, no recorded run)
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("4ba40c6c59d34ab3")
How it ran

no invocation recorded.

SemiCVAE Not yet run
unverified (harvested, no recorded run)
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("d08881e82724d6bc")
How it ran

no invocation recorded.

_Config Not yet run
unverified (harvested, no recorded run)
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("5183511da7aa2db6")
How it ran

no invocation recorded.

cuda_ Not yet run
unverified (harvested, no recorded run)
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("01b0659849fc1d12")
How it ran

no invocation recorded.

get_one_hot_input Not yet run
unverified (harvested, no recorded run)
thu-spmi/LABES/vae_model.py
code served (permissive licence) · get_code("1570b4198e9ba646")
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, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 0 later) and 8 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
toss_
brick of this paper's assembly: helper
2 other papers
2 dated earlier
1909.05528 (2019-09)
MOSS: End-to-End Dialog System Framework with Modular Supervision
2004.07462 (2020-04)
Paraphrase Augmented Task-Oriented Dialog Generation
cuda_
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2004.07462 (2020-04)
Paraphrase Augmented Task-Oriented Dialog Generation

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

Structured belief states are crucial for user goal tracking and database query in task-oriented dialog systems. However, training belief trackers often requires expensive turn-level annotations of every user utterance. In this paper we aim at alleviating the reliance on belief state labels in building end-to-end dialog systems, by leveraging unlabeled dialog data towards semi-supervised learning. We propose a probabilistic dialog model, called the LAtent BElief State (LABES) model, where belief states are represented as discrete latent variables and jointly modeled with system responses given user inputs. Such latent variable modeling enables us to develop semi-supervised learning under the principled variational learning framework. Furthermore, we introduce LABES-S2S, which is a copyaugmented Seq2Seq model instantiation of LABES 1 . In supervised experiments, LABES-S2S obtains strong results on three benchmark datasets of different scales. In utilizing unlabeled dialog data, semi-supervised LABES-S2S significantly outperforms both supervisedonly and semi-supervised baselines. Remarkably, we can reduce the annotation demands to 50% without performance loss on MultiWOZ.

For agents

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

get_harvested_code_for_paper("2009.08115")
get_code_for_paper("2009.08115")
have("2009.08115")

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

curl https://syntology.ai/api/ran/2009.08115.json

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

[![Syntology run record](https://syntology.ai/api/ran/2009.08115.svg)](https://syntology.ai/paper/2009.08115)

Connect an agent — have() is free.