bart-large-xsum / README.md
autoevaluator's picture
Add evaluation results on the default config and test split of xsum
3c50149
|
raw
history blame
4.56 kB
metadata
language:
  - en
license: mit
tags:
  - summarization
model-index:
  - name: facebook/bart-large-xsum
    results:
      - task:
          type: summarization
          name: Summarization
        dataset:
          name: cnn_dailymail
          type: cnn_dailymail
          config: 3.0.0
          split: test
        metrics:
          - type: rouge
            value: 25.2697
            name: ROUGE-1
            verified: true
          - type: rouge
            value: 7.6638
            name: ROUGE-2
            verified: true
          - type: rouge
            value: 17.1808
            name: ROUGE-L
            verified: true
          - type: rouge
            value: 21.7933
            name: ROUGE-LSUM
            verified: true
          - type: loss
            value: 3.5042972564697266
            name: loss
            verified: true
          - type: gen_len
            value: 27.4462
            name: gen_len
            verified: true
      - task:
          type: summarization
          name: Summarization
        dataset:
          name: xsum
          type: xsum
          config: default
          split: test
        metrics:
          - type: rouge
            value: 45.4419
            name: ROUGE-1
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiNzkzZTM0ZTQ3OTJhODgxOGJhNWE0Y2QxMjcwMjBiODdlMzY3Yjk1MjQ1MDJlODQwZjNlZWUyMTEwNmYyYjUzYiIsInZlcnNpb24iOjF9.3_JITjNVx36poltYC02qpeuMiAyYu2AOrfMpCACYdX2_FTtSxxWeYkUJHEbBnuJQKgERHmJncLcQxbh4IlvXBA
          - type: rouge
            value: 22.3723
            name: ROUGE-2
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZDQzZTkxMWM4ZDgyOTcyYmI1MDgyNTk1YmExNTJmNDAxYjViNjlmZjUzNjQ3ZDQwNzQ5ZWQ0ZDU1YTFjYTdlYiIsInZlcnNpb24iOjF9.sDIZfKrHyHDcuYxKNYcvrl-1eMrnwMtm8cA-xDxNP4hX7eEhNoQSAo_CLiPibibcHNMOjZX9fPCMULiGb0qnBw
          - type: rouge
            value: 37.2229
            name: ROUGE-L
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiOWM2MThmY2Y1YmY4OTEwYTZjZjRiNTYxMTY4NDVmMzBhOGM3YTlkZmExMDZmOGU0ZmM1MjMwM2RjOWU1YWQxNyIsInZlcnNpb24iOjF9.TNMvdtMB-5DHUth3HeMc9IilhlciZgPI8AW8RLWl5fWTDko8X0JRk-gTMW6b6cNcRUe2lmfZ9I_ZSd-ZvnjEBA
          - type: rouge
            value: 37.2239
            name: ROUGE-LSUM
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiYzZlMGI2NmRmMTEwMWFhNzAxOTRlMWQyNzA0YjI1ODE3ZjVjYmE1ODMzZTE1M2Y3MTk2YTcxZDYwN2U2NGI4NiIsInZlcnNpb24iOjF9.tcsUnGTDhbrOi1ZNusrI8Do4kt8BuNLD91fhbJwOsr9EvP6NlAAWnfoG1iBSNYKByMcC9Y31lwZlUOUBvnUdDQ
          - type: loss
            value: 2.3128323554992676
            name: loss
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiNzQ3ODBjNWY2ODdhZTAwODUxY2ViMjBmNDY3N2EwMTUzYzdjOGJlZjZlMTI0ZjhkM2I0MGRjNjM5OWNhZDU3NSIsInZlcnNpb24iOjF9.IleOf5Dq60z64kqp6w5dyc6azb1egIARnnKch-x-hpKdQUdTMyPmUO34SpWzuhMt9bJQXRG5qNxb0mpr2-ZMCg
          - type: gen_len
            value: 25.5435
            name: gen_len
            verified: true
            verifyToken: >-
              eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZTE3ODlkZDhhMTEwNTlhNzVjMWMxMGQyZDc0OTc0NWY0MDBlMzUzNGI3MGQwNmJmNzQ3NTQ5MjhhNDhiYTM5YSIsInZlcnNpb24iOjF9.e7nHzg3OH3zkWiCj3iZVAAQG6Zy0E16_MJzBEEyGTlSVuPGMziNfcjRvLD6WeY_6lXUonEwc9lur0X-qUvB7Aw
      - task:
          type: summarization
          name: Summarization
        dataset:
          name: samsum
          type: samsum
          config: samsum
          split: train
        metrics:
          - type: rouge
            value: 24.7852
            name: ROUGE-1
            verified: true
          - type: rouge
            value: 5.2533
            name: ROUGE-2
            verified: true
          - type: rouge
            value: 18.6792
            name: ROUGE-L
            verified: true
          - type: rouge
            value: 20.629
            name: ROUGE-LSUM
            verified: true
          - type: loss
            value: 3.746837854385376
            name: loss
            verified: true
          - type: gen_len
            value: 23.1206
            name: gen_len
            verified: true
      - task:
          type: summarization
          name: Summarization
        dataset:
          name: samsum
          type: samsum
          config: samsum
          split: test
        metrics:
          - type: rouge
            value: 24.9158
            name: ROUGE-1
            verified: true
          - type: rouge
            value: 5.5837
            name: ROUGE-2
            verified: true
          - type: rouge
            value: 18.8935
            name: ROUGE-L
            verified: true
          - type: rouge
            value: 20.76
            name: ROUGE-LSUM
            verified: true
          - type: loss
            value: 3.775235891342163
            name: loss
            verified: true
          - type: gen_len
            value: 23.0928
            name: gen_len
            verified: true

Bart model finetuned on xsum

docs: https://huggingface.co/transformers/model_doc/bart.html

finetuning: examples/seq2seq/ (as of Aug 20, 2020)

Metrics: ROUGE > 22 on xsum.

variants: search for distilbart

paper: https://arxiv.org/abs/1910.13461