qcri/LLMeBench

Issues

• Tests for new model

  #298 opened 6 months ago by MaramHasanain
  0

• Issue with OSACT4SubtaskB dataset loading

  #294 opened 6 months ago by AridHasan
  0

• Locally saved model

  #251 opened 10 months ago by Muhammad-Elmallah
  0

• Missing assets

  #201 opened a year ago by fdalvi
  1

• Extended experiment configuration backup

  #212 opened a year ago by MaramHasanain
  0

• Move common sections to config file

  #18 opened a year ago by hmubarak
  1

• openai api support

  #173 opened a year ago by firojalam
  1

• StanceUnifiedFC prompts don't fit into GPT3.5 prompt length limit

  #38 opened a year ago by fdalvi
  0

• Creating Parent Class for Tasks containing multiple datasets

  #91 opened a year ago by baselmousi
  0

• News categorization datasets and tasks need refactoring

  #37 opened a year ago by fdalvi
  1

• Environment variable for engine_name

  #20 opened a year ago by hmubarak
  1

• Defining a large dataset with several splits under it

  #199 opened a year ago by MaramHasanain
  1

• Allow framework to accept list of assets to run

  #192 opened a year ago by MaramHasanain
  0

• NewsCategorization BLOOM assets post-processing

  #160 opened a year ago by fdalvi
  0

• Location and NameInfo BLOOMz assets

  #159 opened a year ago by fdalvi
  0

• Keep track of inference time

  #53 opened a year ago by MaramHasanain
  0

• automate the collection of experiment results

  #145 opened a year ago by bsabri
  0

• Tests fail for QA tasks: Expecting Counter class to be exported in init file

  #85 opened a year ago by baselmousi
  1

• BLOOM assets that seem to output the same class

  #113 opened a year ago by fdalvi
  0

• `max_tries` should be a runtime argument

  #112 opened a year ago by fdalvi
  0

• STS task improvements

  #110 opened a year ago by fdalvi
  0

• FactualityCOVID19 Fewshot postprocessor always results in "no"

  #107 opened a year ago by fdalvi
  0

• Checkworthiness Few shot processing

  #105 opened a year ago by fdalvi
  0

• Parsing NER task

  #61 opened a year ago by fdalvi
  1

• ChatCompletion

  #6 opened a year ago by firojalam
  1

• FactualityKhouja20 post processing failures

  #28 opened a year ago by fdalvi
  1

• ensure_ascii=False on json dump/dumps

  #33 opened a year ago by firojalam
  1

• JSON files should be saved with `ensure_ascii=False`

  #15 opened a year ago by fdalvi
  2

• Better post processing parsing in QADI_ChatGPT_ZeroShot

  #59 opened a year ago by fdalvi
  0

• Some generic way to get Random baseline

  #49 opened a year ago by fdalvi
  0

• Extracting information from Content warning failures

  #48 opened a year ago by fdalvi
  0

• FactStance and FactClaim prompts

  #16 opened a year ago by fdalvi
  0

• Better summary files

  #7 opened a year ago by fdalvi
  0

• Cache is used across models

  #12 opened a year ago by fdalvi
  0