Goto

Collaborating Authors

 foo fighter


DeFine: A Decomposed and Fine-Grained Annotated Dataset for Long-form Article Generation

Wang, Ming, Wang, Fang, Hu, Minghao, He, Li, Wang, Haiyang, Zhang, Jun, Yan, Tianwei, Li, Li, Luo, Zhunchen, Luo, Wei, Bai, Xiaoying, Geng, Guotong

arXiv.org Artificial Intelligence

Long-form article generation (LFAG) presents challenges such as maintaining logical consistency, comprehensive topic coverage, and narrative coherence across extended articles. Existing datasets often lack both the hierarchical structure and fine-grained annotation needed to effectively decompose tasks, resulting in shallow, disorganized article generation. To address these limitations, we introduce DeFine, a Decomposed and Fine-grained annotated dataset for long-form article generation. DeFine is characterized by its hierarchical decomposition strategy and the integration of domain-specific knowledge with multi-level annotations, ensuring granular control and enhanced depth in article generation. To construct the dataset, a multi-agent collaborative pipeline is proposed, which systematically segments the generation process into four parts: Data Miner, Cite Retreiver, Q&A Annotator and Data Cleaner. To validate the effectiveness of DeFine, we designed and tested three LFAG baselines: the web retrieval, the local retrieval, and the grounded reference. We fine-tuned the Qwen2-7b-Instruct model using the DeFine training dataset. The experimental results showed significant improvements in text quality, specifically in topic coverage, depth of information, and content fidelity. Our dataset publicly available to facilitate future research.


Assisting in Writing Wikipedia-like Articles From Scratch with Large Language Models

Shao, Yijia, Jiang, Yucheng, Kanell, Theodore A., Xu, Peter, Khattab, Omar, Lam, Monica S.

arXiv.org Artificial Intelligence

We study how to apply large language models to write grounded and organized long-form articles from scratch, with comparable breadth and depth to Wikipedia pages. This underexplored problem poses new challenges at the pre-writing stage, including how to research the topic and prepare an outline prior to writing. We propose STORM, a writing system for the Synthesis of Topic Outlines through Retrieval and Multi-perspective Question Asking. STORM models the pre-writing stage by (1) discovering diverse perspectives in researching the given topic, (2) simulating conversations where writers carrying different perspectives pose questions to a topic expert grounded on trusted Internet sources, (3) curating the collected information to create an outline. For evaluation, we curate FreshWiki, a dataset of recent high-quality Wikipedia articles, and formulate outline assessments to evaluate the pre-writing stage. We further gather feedback from experienced Wikipedia editors. Compared to articles generated by an outline-driven retrieval-augmented baseline, more of STORM's articles are deemed to be organized (by a 25% absolute increase) and broad in coverage (by 10%). The expert feedback also helps identify new challenges for generating grounded long articles, such as source bias transfer and over-association of unrelated facts.


Rage Against the Machine, Foo Fighters headlining 2022 return of Boston Calling Music Festival

Boston Herald

Boston Calling is bringing in some of the biggest names in rock and roll -- including Rage Against the Machine and the Foo Fighters -- to headline the music festival when it returns in 2022 after being canceled twice due to the coronavirus pandemic. The acclaimed three-day music festival will return to Allston next Memorial Day, giving top billing to the iconic rock band and a lineup that includes more than 60 performers. An additional headliner will be revealed in the coming months. Eager fans can purchase a limited number of early bird three-day tickets on sale now by visiting bostoncalling.com. Foo Fighters and Rage Against the Machine were both set to headline the 2020 edition of Boston Calling before the festival was canceled due to the pandemic.


Mike Watt to release album of 1995 live gig featuring a backing band of Eddie Vedder, Dave Grohl, Pat Smear and others

Los Angeles Times

Columbia-Legacy has announced the release of a crucial Southland document that captures a snapshot moment when bassist Mike Watt, co-founder of influential San Pedro band Minutemen, went on the road with members of Foo Fighters, Pearl Jam, the Germs and others in support of Watt's first solo album, "Ball Hog or Tugboat?" Called "ring spiel tour '95," the album captures a gig at Chicago's Cabaret Metro that featured Watt and a backing band consisting of musicians from opening acts Foo Fighters and Hovercraft. Specifically, playing with Watt were Eddie Vedder (Pearl Jam, Hovercraft), Dave Grohl (Foo Fighters, Nirvana), Pat Smear (Foo Fighters, Germs) and William Goldsmith (Sunny Day Real Estate, Foo Fighters). For Grohl, the tour was his first since the suicide of Nirvana bandmate Kurt Cobain. The forthcoming archival release will arrive on Nov. 11 as a double LP, CD or digital download. Among its 50-plus collaborators were future or current members of Bikini Kill, Beastie Boys, Nirvana, the Pixies, Swell Maps, Pearl Jam, Meat Puppets, Black Flag, Queens of the Stone Age, Wilco (guitarist Nels Cline) and Sonic Youth.