O->LED_para document simplification system
This is a pretrained version of the document simplification model presented in the Findings of ACL 2023 paper "Context-Aware Document Simplification".
It is a system based on the Longformer encoder-decoder that operates at the paragraph-level and is intended to be guided by a planner.
Target reading levels (1-4) should be indicated via a control token prepended to each input sequence ("<RL_1>", "<RL_2>", "<RL_3>", "<RL_4>"). If using the terminal interface, this will be handled automatically.
How to use
It is recommended to use the plan_simp library to interface with the model.
Here is how to load this model in PyTorch:
# loading
from plan_simp.models.bart import load_simplifier
simplifier, tokenizer, hparams = load_simplifier("liamcripwell/o-ledpara")
# generation
from plan_simp.scripts.generate import Launcher
launcher = Launcher()
launcher.dynamic(model_ckpt="liamcripwell/o-ledpara", clf_model_ckpt="liamcripwell/pgdyn-plan", **params)
Plan-guided generation and evaluation can be run from the terminal (see the repo for more details).
python doc_simp/scripts/generate.py dynamic
--clf_model_ckpt=liamcripwell/pgdyn-plan
--model_ckpt=liamcripwell/o-ledpara
--test_file=<test_data>
--doc_id_col=pair_id
--context_dir=<context_dir>
--out_file=<output_csv>
--reading_lvl=s_level
--context_doc_id=pair_id
--para_lvl=True
python plan_simp/scripts/eval_simp.py
--input_data=newselaauto_docs_test.csv
--output_data=test_out_oledpara.csv
--x_col=complex_str
--r_col=simple_str
--y_col=pred
--doc_id_col=pair_id
--prepro=True
--sent_level=True
- Downloads last month
- 13
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.