OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Iterater

Iterater is a public Python research repository containing datasets and code to preprocess, train and test models for iterative text revision, with released Hugging Face datasets and model checkpoints.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Iterater — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Iterater when a research or development workflow needs iterative-revision datasets, model code or released checkpoints, rather than a hosted general-purpose writing editor.

Who it is for

  • NLP researchers studying how people revise written text
  • Developers building or evaluating models for multi-round text editing
  • Teams exploring human-in-the-loop revision workflows

Documented uses

  • Prepare sentence- or document-level revision data for model work
  • Train or test BART and PEGASUS text-revision models
  • Load published revision datasets and pretrained checkpoints for research
  • Study model-assisted editing alongside human writers
Before you choose

Requirements & limits

The GitHub repository is public and identifies an Apache-2.0 license.

  • The README describes Python 3.7 as the tested version and installs dependencies from requirements.txt.
  • The code is mainly based on Hugging Face Transformers; data and checkpoint examples use datasets and transformers.
  • The repository includes several dataset families and task granularities, so select the variant matching the intended sentence- or document-level task.
  • Its documentation covers research code and model artifacts; the linked Colab demo is presented as a simple example for iterative text revision.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Load a published revision dataset and checkpoint

Load a named IteraTeR dataset and PEGASUS revision model using the README’s Hugging Face examples.

Before you start

  • Python environment with the project requirements installed using pip install -r requirements.txt
  • Hugging Face datasets and transformers packages, used in the README examples

Steps

  1. Import load_dataset from datasets and load a listed IteraTeR dataset, such as wanyu/IteraTeR_human_sent.

  2. Import AutoTokenizer and AutoModelForSeq2SeqLM from transformers.

  3. Load the tokenizer and model checkpoint wanyu/IteraTeR-PEGASUS-Revision-Generator using from_pretrained.

Expected outcome

The example provides a loaded dataset, tokenizer and PEGASUS revision model for a downstream research workflow.

Guide limitations

  • The README example documents loading artifacts, not a complete revision-generation or evaluation procedure.
  • The README says the project was tested using Python 3.7; it does not establish compatibility with other environments.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
vipulraheja/iterater · Official homepage title · retained in its source language
How we review sourcesSuggest a correction