Jiahao Zhao

Question about generated summary

5

Thanks sharing the awesome implementation. In batcher.py the padding function seems to pad decoder input to max_len. Is your model able to generate variable lenghth summary eventually? > def pad_decoder_inp_targ(self,...

Questions about string cleaning

1

Thanks for this solid work. In the `clean_str`, it seems that `Every dataset is lower cased except for TREC` but in the example, in Table 6 the sentence is cased....

Sentence-level or paraphrase-based attack required

3

**Is your feature request related to a problem? Please describe.** Current textattack is substitute-based attack in char/word-level. Add some paraphrase-based attack could increase the diversity of attack recipe. **Describe alternatives...

After updating, `attack_evals.DefaultAttackEval` no longer exist. The test below is broken. https://github.com/thunlp/OpenAttack/blob/bfedfa74f37c69db6d7092d9cc61822ee324919d/slow_tests/test_multi_process.py#L32 Because we don't have `DefaultAttackEval` anymore, it seems like we don't have a function to generate adversarial examples....

Build custom non-English dataset with ARPABET

2

Hi, thanks for opening this project. I'm a newbie in VC and I try to add a new speaker to assem-vc. In Prepare Metadata section, @wookladin uses python datasets/g2p.py to...

Code release date

1

Great work! When will the code be released?

Question about the tradeoff between robustness with generalization in ABSA task

2

Great work and solid experiment 🎉 In the paper table 10b, column 'Adv→N' shows that adversarial training could improve the model performance in the ARTS test dataset. My question is...

Classifier parameter setting

1

感谢@hankcs。请问如何使用这段grid search？ https://github.com/hankcs/text-classification-svm/blob/be8c6541720363faeccf21a05021dd9e30047229/src/main/java/com/hankcs/hanlp/classification/classifiers/LinearSVMClassifier.java#L96

According to [HISTORY.md](https://github.com/DAI-Lab/fibber/blob/master/HISTORY.md#version-030---2021-06-05), `BertSamplingStrategy` has been renamed to `ASRSStrategy`, but it remain `BertSamplingStrategy` in `BENCHMARK.md`. Thus update of `BENCHMARK.md` is needed. By the way, I think descriptions in `Advanced Usage`...

Error caused by this line https://github.com/DAI-Lab/fibber/blob/ac278a06ae3204ba54d2ba2cbdd4dbe11f0bd745/fibber/benchmark/benchmark.py#L135 I think it should change to ``` self._metric_bundle.add_classifier(customized_clf, set_target_clf=True) ```