Skip to content
OpenTrain AIFor AI Companies

ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training

Published Jan 1, 2020
arXiv PDF
Researcher verdict
Starting point
Use as implementation starting point
Benchmark evidence
Missing
Not verified yet
Time to first repro
A few days
Plan setup time
Risk flags
1
Review before use

Results and benchmarks

Freshness tier: cold
ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training is the primary contribution described in this paper.

Implementation

Historical official implementation (not recommended for new builds)

Why this implementation
Confidence: low

huggingface/transformers is the closest maintained adjacent implementation (Partial overlap with paper title keywords). It is not paper-verified; validate algorithm and evaluation setup against the paper before trusting reported metrics. Community adoption signal: 164399 GitHub stars.

Open microsoft/ProphetNet
Reproduction risks
  • Adjacent implementations are not paper-verified
  • Recommended repository is adjacent and not paper-verified.
  • Adjacent implementation match confidence is low.
  • No direct maintained implementation is currently verified.
  • Only historical official repository was found: microsoft/ProphetNet.
  • No maintained paper-verified implementation met reliability thresholds.

Compare implementation paths

Compare maintenance quality, reproducibility coverage, and evidence confidence before choosing a reproduction baseline.

microsoft/ProphetNet
historical official
Maintenance
Stale
Confidence
High
Reproducibility
Limited
Stars
746
Last push
Jul 25, 2024 (762d)

Official implementation from Papers with Code · Repository link is mentioned in the paper metadata

  • No push in 12+ months
  • No CI pipeline detected
  • No tagged releases
wasiahmad/NeuralKpGen
alternative
Maintenance
Stale
Confidence
Low
Reproducibility
Moderate
Stars
8
Last push
Jun 12, 2023 (1170d)

Matched via arXiv identifier search · Partial overlap with paper title keywords

  • No push in 12+ months
  • No CI pipeline detected
  • No tagged releases
anna-kay/Reddit-summarization
alternative
Maintenance
Stale risk
Confidence
Low
Reproducibility
Limited
Stars
5
Last push
Oct 12, 2025 (317d)

Matched via arXiv identifier search

  • No CI pipeline detected
  • No tagged releases
  • No Docker setup

Reproduction readiness

Time to first repro: days
Last checked: Aug 25, 2026

Major work

No dependency manifest, manual reconstruction required

  • microsoft/ProphetNet has no requirements.txt, environment.yml, pyproject.toml, or Dockerfile.
  • You will need to reverse-engineer dependencies from import statements in the source code.
  • Last push was 762 days ago.
Open microsoft/ProphetNet

Hardware requirements

  • Expect multi-day setup/compute for meaningful reproduction based on current guidance.

Repositories and ecosystem

Closest related implementations

These are not paper-verified. Use them as reference points when no direct implementation is available.

No additional verified repositories beyond the primary recommendation.

Hugging Face artifacts

No trustworthy direct or curated related Hugging Face artifacts were found yet. Use targeted searches to quickly locate candidate models, datasets, and demos.

Search models Search datasets Search spaces

Tip: start with models, then check datasets and spaces if you need evaluation data or demos.

Research context

Evaluation and human feedback data

Open this paper in HFEPX to review benchmark signals, evaluation modes, and human-feedback protocol context.

Open in HFEPX

Data includes links from Papers with Code ( CC-BY-SA-4.0 ).