--- language: en license: apache-2.0 tags: - summarization datasets: arxiv-summarization model-index: - name: ArtifactAI/led_large_16384_arxiv_summarization results: - task: type: summarization name: Summarization dataset: name: ccdv/arxiv-summarization type: ccdv/arxiv-summarization config: section split: test metrics: - type: rouge value: 37.9472 name: ROUGE-1 verified: true verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZDFkMzY4YTk0NGUyNDJjYzc2MWFiMGJlNWUyYTM2YjlmNjlkY2VkYmVhMDk2YjIxMjE3MjE4M2ZkOTAwODE2ZSIsInZlcnNpb24iOjF9.t2x5mqi0xM9Q0K9MscHZ6v_5pc-MOw8KieFTvFMqh5K4UAvvvcVGOGfGQi_Qb57gQa2DkrW0cNrJADY0VA1tAQ - type: rouge value: 11.3138 name: ROUGE-2 verified: true verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiNjdlYmQ4ZmRkNzc3YzE0NGQ2MTRhNDE4YTExNDYwYmNjODFhYjdmYTJlZWE4OTRhYWRiZmNmODZkMDZjMWY3NSIsInZlcnNpb24iOjF9.RPWY5CZMjaFaQ1vRQPoHyZxPD67dQdbXYL0UlJ53b_q1dMczXb7HtE_UmDNPi6F7thciVt6xWIzsckVmp9ZJCw - type: rouge value: 20.5557 name: ROUGE-L verified: true verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiYWEwNTQ5MWViZTYwM2EyNzI0OWEyZDNlY2ExOTJiMjI3MmNjM2I4YmJjMzljYTQ3NjhkNjAzYzM5MDQzYjVkOCIsInZlcnNpb24iOjF9.ZgSkTbiUDaQRJGBIXjlTZKbtKmrIljEJ6btwhyfBsaz5oS0qmI76-b_vDRswnx96OcGTqdxICIjma6jgNbKiBA - type: rouge value: 33.8336 name: ROUGE-LSUM verified: true verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiY2EzNzNhMWVmYjM5ZWUwOTZkYjU0MGZjMWQ0YTQ1NzA1NWQ4MjBjNjNhM2FmMmE3MmM3NzQwMzVkN2QzMzQxZiIsInZlcnNpb24iOjF9.bhxtgWXjCEv5ZFY3F7Mp-r4EHrIU8BNZ8X2zhpjSoyVLmjbfdFB-lnJdoH3PfVZEa14T96SJqMSHa6yzlqGEAQ - type: loss value: 2.8064792156219482 name: loss verified: true verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiYzBhMTE0ZTdhOTRmYWE1Mjk5ZmViYjZiMjBmNzc2YzQ4YmNhYWM3NzRjYWUwYTEyZjU1NGI5MjVhODQwOTBlNCIsInZlcnNpb24iOjF9.l0nIJCcjoFyPF9M7MHiQxBQ3wtyk6jXURY0ZF6Xny3_DpkDh5YHs9kF494GJp5eYj6XG5HRGCgqhfmU7-fywAw - type: gen_len value: 157.4174 name: gen_len verified: true verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiNDY0ZmE4M2VmOTU1NWY5M2I4YTYxNjM3NTkxNWU4NDY3N2Y0MTM1YWNlNmNjMGQ4N2UzM2ZkZWJhZTVmMjQ2OCIsInZlcnNpb24iOjF9.sAp6g7nt1tKTdGfOlGm3fdxzH1jxjNOZO65BNnVJkxDhu86j8QP3ZvNPv7PpD2sK4p6yM_HlHPPeX4bgmDi2BQ --- ## Introduction A led-large-16384 model to summarize ArXiv papers. Inputs are the abstracts of papers and full documents, and outputs are the summaries of the papers. [Allenai's Longformer Encoder-Decoder (LED)](https://github.com/allenai/longformer#longformer). As described in [Longformer: The Long-Document Transformer](https://arxiv.org/pdf/2004.05150.pdf) by Iz Beltagy, Matthew E. Peters, Arman Cohan, *led-base-16384* was initialized from [*bart-base*](https://huggingface.co/facebook/bart-base) since both models share the exact same architecture. To be able to process 16K tokens, *bart-base*'s position embedding matrix was simply copied 16 times.