Papers › A Simple Language Model for Task-Oriented Dialogue

A Simple Language Model for Task-Oriented Dialogue

2 May 2020NeurIPS 2020 12arXiv:2005.00796archive 2025-07-28

Ehsan Hosseini-Asl, Bryan McCann, Chien-Sheng Wu, Semih Yavuz, Richard Socher

Task-oriented dialogue is often decomposed into three tasks: understanding user input, deciding actions, and generating a response. While such decomposition might suggest a dedicated model for each sub-task, we find a simple, unified approach leads to state-of-the-art performance on the MultiWOZ dataset. SimpleTOD is a simple approach to task-oriented dialogue that uses a single, causal language model trained on all sub-tasks recast as a single sequence prediction problem. This allows SimpleTOD to fully leverage transfer learning from pre-trained, open domain, causal language models such as GPT-2. SimpleTOD improves over the prior state-of-the-art in joint goal accuracy for dialogue state tracking, and our analysis reveals robustness to noisy annotations in this setting. SimpleTOD also improves the main metrics used to evaluate action decisions and response generation in an end-to-end setting: inform rate by 8.1 points, success rate by 9.7 points, and combined score by 7.2 points.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2005.00796")

Code

Syntology Ran 3 of 14 code samples harvested from 1 repository linked to this paper; 11 have no recorded run. Of those that ran: 1 ran · violated contract; 1 ran · our draft was wrong; 1 ran with no contract checked.

By repository: official repository: 14 samples from 1 repository, 3 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

salesforce/simpletod officialmentioned in papermentioned on GitHubpytorchBSD-3-Clause report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

14 samples harvested; 3 ran; 0 honoured the contract we drafted; 11 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · violated contract
1ran · our draft was wrong
1ran
11unverified

Licence: 0 of the 14 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from salesforce/simpletod. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

is_ascii salesforce/simpletod/preprocess_multiwoz.py official repository ran · violated contract BSD-3-Clause (permissive) · 1b465b9239cb631c · report
prune_linear_layer salesforce/simpletod/models/modeling_utils.py official repository ran · our draft was wrong BSD-3-Clause (permissive) · d30d6c3098df2c42 · report
url_to_filename salesforce/simpletod/models/file_utils.py official repository ran fingerprinted BSD-3-Clause (permissive) · e827d94caab33f59 · report
create_position_ids_from_input_ids salesforce/simpletod/models/modeling_utils.py official repository unverified BSD-3-Clause (permissive) · d4f3546f2ecff417 · report
fixDelex salesforce/simpletod/preprocess_multiwoz.py official repository unverified BSD-3-Clause (permissive) · 9db2246dbf2fe48c · report
get_belief salesforce/simpletod/utils/simpletod.py official repository unverified BSD-3-Clause (permissive) · 6593f647e65bca52 · report
get_belief_dbsearch salesforce/simpletod/utils/simpletod.py official repository unverified BSD-3-Clause (permissive) · d273f88c13f089a1 · report
get_belief_openaigpt salesforce/simpletod/utils/simpletod.py official repository unverified BSD-3-Clause (permissive) · 817ada1e305378b6 · report
hf_bucket_url salesforce/simpletod/models/file_utils.py official repository unverified BSD-3-Clause (permissive) · b2be6c932ce41084 · report
is_remote_url salesforce/simpletod/models/file_utils.py official repository unverified BSD-3-Clause (permissive) · 4381d1f588a7f42a · report
load_tf_weights_in_gpt2 salesforce/simpletod/models/modeling_gpt2.py official repository unverified BSD-3-Clause (permissive) · c4bac98236c1fc2c · report
postprocess_gpt2 salesforce/simpletod/evaluate_multiwoz.py official repository unverified BSD-3-Clause (permissive) · 9b5c8cad10804374 · report
remove_model_mismatch_and_db_data salesforce/simpletod/evaluate_multiwoz.py official repository unverified BSD-3-Clause (permissive) · 489b09c8fef69f12 · report
top_k_top_p_filtering salesforce/simpletod/models/modeling_utils.py official repository unverified BSD-3-Clause (permissive) · 8e873fc150477bd1 · report

Tasks

Dialogue State TrackingEnd-To-End Dialogue ModellingLanguage ModelingLanguage ModellingMulti-domain Dialogue State TrackingResponse GenerationTransfer Learningmodel

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
End-To-End Dialogue Modelling MULTIWOZ 2.0 SimpleTOD BLEU 15.0 #5 of 6 Archive leaderboard report
End-To-End Dialogue Modelling MULTIWOZ 2.0 SimpleTOD MultiWOZ (Inform) 84.4 #5 of 6 Archive leaderboard report
End-To-End Dialogue Modelling MULTIWOZ 2.0 SimpleTOD MultiWOZ (Success) 70.1 #5 of 6 Archive leaderboard report
End-To-End Dialogue Modelling MULTIWOZ 2.1 SimpleTOD BLEU 15.2 #4 of 4 Archive leaderboard report
End-To-End Dialogue Modelling MULTIWOZ 2.1 SimpleTOD MultiWOZ (Inform) 85.0 #4 of 4 Archive leaderboard report
End-To-End Dialogue Modelling MULTIWOZ 2.1 SimpleTOD MultiWOZ (Success) 70.5 #4 of 4 Archive leaderboard report
Response Generation MMConv SimpleTOD BLEU 20.3 #2 of 2 Archive leaderboard report
Response Generation MMConv SimpleTOD Comb. 32.2 #2 of 2 Archive leaderboard report
Response Generation MMConv SimpleTOD Inform 14.6 #2 of 2 Archive leaderboard report
Response Generation MMConv SimpleTOD Success 9.2 #2 of 2 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AdamAttentionAttention DropoutBPECosine AnnealingDense ConnectionsDiscriminative Fine-TuningDropoutGPT-2Layer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionResidual ConnectionSoftmaxWeight Decay

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections