jind11/TextFooler

Issues

• Which python version?

  #63 opened 9 months ago by jhuvehicle
  13

• Cannot reproduce results with FAKE

  #60 opened 2 years ago by AntoineGourru
  1

• Evaluation on AGnews

  #62 opened 2 years ago by MT010104
  0

• full adversarial example for FAKE using BERT

  #52 opened 3 years ago by maverickyu
  13

• Question about pretrained models w.r.t InferSent and ESIM

  #47 opened 3 years ago by adorable-mx
  5

• The usage of '<oov>' is not consistent with the paper

  #39 opened 4 years ago by plasmashen
  3

• Questions about the dataset

  #46 opened 3 years ago by Youoo1
  1

• Quality of adversaries and authenticity of results

  #24 opened 5 years ago by SachJbp
  4

• Question about the results of the attack

  #37 opened 4 years ago by milumilum
  3

• Problems in the process of reproducing the code

  #44 opened 4 years ago by ChenJing-825
  5

• Number of queries in attack_classification.py etc.

  #43 opened 4 years ago by superctj
  1

• Not able to run cos_sim.py

  #30 opened 4 years ago by ashwani-bhat
  1

• MemoryError for calculating cosine similarity scores

  #3 opened 5 years ago by sumuzhao
  5

• Questions about string cleaning

  #38 opened 4 years ago by Opdoop
  1

• question about sim_scores

  #36 opened 4 years ago by zedzx1uv
  1

• Is it possible to include support for any custom dataset in a given specified format?

  #35 opened 4 years ago by QMrpy
  1

• version

  #33 opened 4 years ago by yulu0724
  6

• POS filter - why 'NOUN' and 'VERB' can be replaced by each other

  #31 opened 4 years ago by sharon-gao
  1

• Implementation on other dataset

  #29 opened 4 years ago by sharon-gao
  2

• Are the models in drive already trained on Adversarial examples generated?

  #27 opened 5 years ago by SachJbp
  2

• How to install USE successfully?

  #28 opened 4 years ago by sharon-gao
  7

• Segmentation fault (core dumped) when running USE

  #5 opened 5 years ago by kiaia
  4

• About Data Size

  #26 opened 5 years ago by CuiMingyu
  2

• Adversarial examples provided in google drive are not able to fool your trained target model (BERT) itself

  #23 opened 5 years ago by SachJbp
  8

• The counter-fitting word embeddings

  #21 opened 5 years ago by Palipoor
  2

• Regarding Semantic Similarity

  #20 opened 5 years ago by RishabhMaheshwary
  5

• examples

  #19 opened 5 years ago by jxmorris12
  1

• Missing vocabularies for wordCNN and wordLSTM pretrained models

  #18 opened 5 years ago by mahossam
  0

• How do you measure the after-attack accuracy ?

  #17 opened 5 years ago by mahossam
  6

• Low Accuracy of BERT Pretrained Model on Fake News

  #15 opened 5 years ago by RishabhMaheshwary
  5

• Question on Experiment Details

  #16 opened 5 years ago by NoviScl
  2

• Invalid argument when calculating sim_predictor.senmantic_sim()

  #14 opened 5 years ago by tyz98
  1

• Formula for calculating USE cosine similarities: dividing by π

  #6 opened 5 years ago by jxmorris12
  4

• Results slightly different from paper

  #1 opened 5 years ago by uvafan
  5

• Regarding --max_sequence_length

  #13 opened 5 years ago by RishabhMaheshwary
  14

• Results differ from paper on AG news dataset

  #12 opened 5 years ago by RishabhMaheshwary
  3

• Regarding No. of classes in AG News

  #11 opened 5 years ago by RishabhMaheshwary
  0

• 特地前来感谢，个人认为是文本对抗领域的开山之作之一

  #10 opened 5 years ago by cwb96
  1

• Doubt regarding synonym replacement

  #9 opened 5 years ago by RishabhMaheshwary
  4

• Vocab.txt for running Imdb is not available

  #8 opened 5 years ago by RishabhMaheshwary
  3

• About pytorch_pretrained_bert

  #7 opened 5 years ago by lexmen318
  4

• Evaluation of models on SNLI and MNLI

  #2 opened 5 years ago by chauhanjatin10
  0