🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

We’ve just launched our official Twitter (X) account! We’ll be sharing exciting AI x Music projects from UCSD (PIs: McAuley Lab UCSD Taylor Berg-Kirkpatrick and Shlomo Dubnov) Stay tuned for updates on our ISMIR participation!

Tomorrow afternoon, our PhD student, Jingyue Huang (Jingyue Huang), will be presenting her collaborative work with Ke Chen and Yi-Hsuan Yang on Emotion-Driven Piano Music Generation at the ISMIR Conference conference. Join Session 2 to catch her presentation! #ismir2024

#ismir2024 Join us tomorrow afternoon at Session 6! Our PhD student Zachary Novack presents: 'DITTO-2: Distilled Diffusion Inference-Time T-Optimization for Music Generation' Don't miss this exciting work on AI music generation! 🎵⚡️

#ismir2024 Interested in lyrics datasets? Tomorrow Haven Kim presents: 'LyCon: Lyrics Reconstruction From the Bag-Of-Words Using Large Language Models' Featuring a new copyright-free lyrics dataset! Join the virtual LBD session at 12:15PM to learn more! 📝🎵

#nlp4musa #ismir2024 Attending NLP4MusA? Don't miss Junda Wu (Junda Wu)'s presentation: 'Futga: Fine-grained Music Understanding through Temporally-enhanced Generative Augmentation' A novel music captioning model that captures fine-grained musical nuances! 🎵🤖

Proud of our team's strong showing at #ismir2024! Despite being a relatively small group, we delivered: • 1 tutorial • 4 main track papers • 1 LBD paper • 1 workshop paper Growing fast and driving forward! Follow us for more exciting research ahead! 🎵

🦋 We're on Bluesky now! bsky.app/profile/ucsd-m…

Super excited for PDMX to be accepted at the NeurIPS Creativity and Generative AI Workshop Workshop on Creativity & Generative AI! Sad to miss it, but our wonderful lead on the project Phillip Long will be there to chat all things open-source symbolic music!

Exciting invited talk today by Megan Wei Megan Wei from Brown CS on her #ISMIR2024 work, 'Do Music Generation Models Encode Music Theory?'! Their novel SynTheory dataset & probing framework reveals how models like Jukebox & MusicGen encode fundamental music concepts. 🎵

Great AI Music showing at the UCSD CSE GenAI Summit! Huge thanks to our invited guests Nao Tokui and Anna Huang for coming to hang with the MUSAIC group past and present!

We had Danbinaerin Han DanbinaerinHan as our guest speaker! She presented her ISMIR 2024 Best Paper Award-winning work, Six Dragons Fly Again. We were also privileged to have the paper’s last author, Professor Dasaem Jeong Dasaem Jeong, join us. Thanks for coming, Danbi and Dasaem!

🚨New Audio Benchmark 🚨We find standard LLMs can solve Music-QA benchmarks by just guessing from text only, + LALMs can still answer well when given noise instead of music! Presenting RUListening: A fully automated pipeline for making Audio-QA benchmarks *actually* assess

Presenting RUListening! we edit Music-QA benchmarks to *actually* assess audio perception, using text-only LLMs to generate unimodally-hard distractors. Been super excited about this one (led by the beast Yongyi Zang), check out the full thread below! And at ISMIR 2025!🇰🇷