
Sign up to save your podcasts
Or


This paper presents a systematic approach for fusing transformer-based networks using Optimal Transport to align architectural components. The proposed method outperforms vanilla fusion and individual parent models, showcasing the potential of fusing multiple Transformers.
https://arxiv.org/abs//2310.05719
YouTube: https://www.youtube.com/@ArxivPapers
TikTok: https://www.tiktok.com/@arxiv_papers
Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016
Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers
By Igor Melnyk5
33 ratings
This paper presents a systematic approach for fusing transformer-based networks using Optimal Transport to align architectural components. The proposed method outperforms vanilla fusion and individual parent models, showcasing the potential of fusing multiple Transformers.
https://arxiv.org/abs//2310.05719
YouTube: https://www.youtube.com/@ArxivPapers
TikTok: https://www.tiktok.com/@arxiv_papers
Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016
Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers

953 Listeners

1,971 Listeners

438 Listeners

112,759 Listeners

10,063 Listeners

5,531 Listeners

214 Listeners

51 Listeners

99 Listeners

473 Listeners