Papers › ERNIE-GEN: An Enhanced Multi-Flow Pre-training and Fine-tuning Framework for Natural...

ERNIE-GEN: An Enhanced Multi-Flow Pre-training and Fine-tuning Framework for Natural Language Generation

26 Jan 2020arXiv:2001.11314archive 2025-07-28

Dongling Xiao, Han Zhang, Yukun Li, Yu Sun, Hao Tian, Hua Wu, Haifeng Wang

Current pre-training works in natural language generation pay little attention to the problem of exposure bias on downstream tasks. To address this issue, we propose an enhanced multi-flow sequence to sequence pre-training and fine-tuning framework named ERNIE-GEN, which bridges the discrepancy between training and inference with an infilling generation mechanism and a noise-aware generation method. To make generation closer to human writing patterns, this framework introduces a span-by-span generation flow that trains the model to predict semantically-complete spans consecutively rather than predicting word by word. Unlike existing pre-training methods, ERNIE-GEN incorporates multi-granularity target sampling to construct pre-training data, which enhances the correlation between encoder and decoder. Experimental results demonstrate that ERNIE-GEN achieves state-of-the-art results with a much smaller amount of pre-training data and parameters on a range of language generation tasks, including abstractive summarization (Gigaword and CNN/DailyMail), question generation (SQuAD), dialogue generation (Persona-Chat) and generative question answering (CoQA).

PaperPDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Sharpiless/Versailles-text-generation-with-paddlepaddle mentioned on GitHubpaddleSyntology: not harvested report
https-github-com-GiangHoang9912/ernie-gen mentioned on GitHubpaddleSyntology: not harvested report
MindCode-4/code-10 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Fourier Contour Embedding for Arbitrary-Shaped Text Detection (arXiv:2104.10442; that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 1 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
MindSpore-scientific/code-12 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models (arXiv:2503.09573; that paper's own run record: 14 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 7 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
PaddlePaddle/PaddleNLP paddleSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ERNIE 3.0 Tiny: Frustratingly Simple Method to Improve Task-Agnostic Distillation... (arXiv:2301.03416; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Abstractive Text SummarizationDecoderDialogue GenerationGenerative Question AnsweringQuestion GenerationQuestion-GenerationText GenerationText Summarization

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Abstractive Text Summarization CNN / Daily Mail ERNIE-GENLARGE (large-scale text corpora) ROUGE-1 44.31 #15 of 53 Archive leaderboard report
Abstractive Text Summarization CNN / Daily Mail ERNIE-GENLARGE (large-scale text corpora) ROUGE-2 21.35 #15 of 53 Archive leaderboard report
Abstractive Text Summarization CNN / Daily Mail ERNIE-GENLARGE (large-scale text corpora) ROUGE-L 41.60 #15 of 53 Archive leaderboard report
Abstractive Text Summarization CNN / Daily Mail ERNIE-GENLARGE ROUGE-1 44.02 #20 of 53 Archive leaderboard report
Abstractive Text Summarization CNN / Daily Mail ERNIE-GENLARGE ROUGE-2 21.17 #20 of 53 Archive leaderboard report
Abstractive Text Summarization CNN / Daily Mail ERNIE-GENLARGE ROUGE-L 41.26 #20 of 53 Archive leaderboard report
Abstractive Text Summarization CNN / Daily Mail ERNIE-GENBASE ROUGE-1 42.30 #26 of 53 Archive leaderboard report
Abstractive Text Summarization CNN / Daily Mail ERNIE-GENBASE ROUGE-2 19.92 #26 of 53 Archive leaderboard report
Abstractive Text Summarization CNN / Daily Mail ERNIE-GENBASE ROUGE-L 39.68 #26 of 53 Archive leaderboard report
Generative Question Answering CoQA ERNIE-GEN F1-Score 84.5 #1 of 3 Archive leaderboard report
Question Generation SQuAD1.1 ERNIE-GENLARGE (beam size=5) BLEU-4 25.41 #1 of 13 Archive leaderboard report
Text Summarization GigaWord ERNIE-GENLARGE (large-scale text corpora) ROUGE-1 39.46 #10 of 41 Archive leaderboard report
Text Summarization GigaWord ERNIE-GENLARGE (large-scale text corpora) ROUGE-2 20.34 #10 of 41 Archive leaderboard report
Text Summarization GigaWord ERNIE-GENLARGE (large-scale text corpora) ROUGE-L 36.74 #10 of 41 Archive leaderboard report
Text Summarization GigaWord ERNIE-GENLARGE ROUGE-1 39.25 #13 of 41 Archive leaderboard report
Text Summarization GigaWord ERNIE-GENLARGE ROUGE-2 20.25 #13 of 41 Archive leaderboard report
Text Summarization GigaWord ERNIE-GENLARGE ROUGE-L 36.53 #13 of 41 Archive leaderboard report
Text Summarization GigaWord ERNIE-GENBASE ROUGE-1 38.83 #19 of 41 Archive leaderboard report
Text Summarization GigaWord ERNIE-GENBASE ROUGE-2 20.04 #19 of 41 Archive leaderboard report
Text Summarization GigaWord ERNIE-GENBASE ROUGE-L 36.20 #19 of 41 Archive leaderboard report
Text Summarization GigaWord-10k ERNIE-GENLARGE (large-scale text corpora) ROUGE-1 35.51 #1 of 3 Archive leaderboard report
Text Summarization GigaWord-10k ERNIE-GENLARGE (large-scale text corpora) ROUGE-2 16.79 #1 of 3 Archive leaderboard report
Text Summarization GigaWord-10k ERNIE-GENLARGE (large-scale text corpora) ROUGE-L 33.23 #1 of 3 Archive leaderboard report
Text Summarization GigaWord-10k ERNIE-GENLARGE ROUGE-1 35.05 #2 of 3 Archive leaderboard report
Text Summarization GigaWord-10k ERNIE-GENLARGE ROUGE-2 16.10 #2 of 3 Archive leaderboard report
Text Summarization GigaWord-10k ERNIE-GENLARGE ROUGE-L 32.50 #2 of 3 Archive leaderboard report
Text Summarization GigaWord-10k ERNIE-GENBASE ROUGE-1 33.75 #3 of 3 Archive leaderboard report
Text Summarization GigaWord-10k ERNIE-GENBASE ROUGE-2 15.23 #3 of 3 Archive leaderboard report
Text Summarization GigaWord-10k ERNIE-GENBASE ROUGE-L 31.35 #3 of 3 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

Introduced by this paper: ERNIE-GEN

ERNIE-GEN

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