neulab/BARTScore

Issues

• package versions have conflicting dependencies.

  #46 opened 7 months ago by bigbrother001
  1

• Installation Issue

  #45 opened 8 months ago by DurgaSandeep25
  0

• Which prompt was used in the paper's result?

  #43 opened 9 months ago by Hyfred
  4

• Hi, can we use BARTScorer to evaluate any open source LLM model?

  #40 opened 9 months ago by SnehaDeepak
  1

• Demo is not working

  #42 opened 9 months ago by dzf0023
  1

• Installation issue

  #44 opened 9 months ago by ndenicoo
  1

• there is no code of bart_score_cnn_src_hypo?

  #41 opened a year ago by lunar333
  1

• PyPi?

  #3 opened 3 years ago by BramVanroy
  5

• Using BARTScore to Compare 2 summaries without Human Evaluation

  #39 opened a year ago by pranamyapatil
  2

• Baseline result in SummEval is different from original paper?

  #38 opened a year ago by SefaZeng
  0

• hello, is this metric the lower the better? and I wonder Why the value is negative?

  #35 opened a year ago by meiling-fdu
  1

• Question about unreferenced usage

  #37 opened 2 years ago by susisheng
  2

• question about usage of different correlation coefficients for various datasets

  #36 opened 2 years ago by xuyifan-0731
  0

• Unusual method for population re-sampling in bootstrap

  #34 opened 2 years ago by dennlinger
  0

• Link to demo in readme is no longer applicable

  #33 opened 2 years ago by mslucke99
  0

• How to choose scoring model?

  #32 opened 2 years ago by xjw-star
  0

• how can I download the checkpoints and run with two summaries ?

  #31 opened 2 years ago by shamanez
  0

• Python details

  #30 opened 2 years ago by dzf0023
  1

• Some questions in Shift_Right_Tokens

  #29 opened 2 years ago by Ricardokevins
  0

• Bartscore calculation question

  #28 opened 2 years ago by gmichalo
  1

• How do I map the results in the interval 0 to 1?

  #16 opened 2 years ago by HaoBytes
  11

• WMT 2019 - Can you provide more details about the split and dataset you used?

  #27 opened 2 years ago by PastelBelem8
  0

• Meaning of the value range?

  #26 opened 2 years ago by GabrielLin
  4

• human scores in QAGS_XSUM are all 0/1

  #25 opened 2 years ago by cyr19
  2

• Question about D2T datasets

  #24 opened 2 years ago by taku-ito
  4

• "EOFError: Ran out of input" when trying to load QAGS_CNN/data.pkl

  #23 opened 2 years ago by taku-ito
  1

• Spearman Corrleations for Table-4

  #22 opened 2 years ago by Atharva-Phatak
  3

• <mask> token disappear?

  #21 opened 2 years ago by KpKqwq
  2

• Use BARTScore as loss function

  #18 opened 2 years ago by hoangftran
  1

• Trouble with ROUGE score

  #17 opened 2 years ago by JohnGiorgi
  1

• question before using it

  #15 opened 3 years ago by xinli2008
  1

• colab

  #14 opened 3 years ago by HaoBytes
  1

• Some problem while fine-tuing on Paraphrase Dataset

  #6 opened 3 years ago by Ricardokevins
  9

• Pearson Correlations results in Table 4

  #7 opened 3 years ago by tanay2001
  2

• A slight problem about Different Baseline result

  #13 opened 3 years ago by Ricardokevins
  2

• License

  #12 opened 3 years ago by Sophylax
  1

• Multilingual?

  #9 opened 3 years ago by puraminy
  1

• Scope of application of the BARTScore

  #10 opened 3 years ago by ShiYaya
  1

• Score if hyp==ref

  #11 opened 3 years ago by jheinecke
  1

• A question about prompt-based BARTScore

  #8 opened 3 years ago by StevenTang1998
  2

• Mutlireference support

  #5 opened 3 years ago by tanay2001
  2

• A problem while using

  #4 opened 3 years ago by Ricardokevins
  4

• How the score is different from perplexity?

  #2 opened 3 years ago by fabrahman
  1

• Can't install `BLEURT==0.0.2`

  #1 opened 3 years ago by danieldeutsch
  1