Tootfinder

Shield AI has emerged as a lead contender in the increasingly crowded race against competitors like Anduril and AeroVironment
to outfit the U.S. military with killer drones.
The $5 billion-valued startup, founded in 2015 by brothers Ryan and Brandon Tseng,
sells a suite of hardware and products, including autonomous piloting software that has been used to fly fighter jets.
In March, it raised $240 million in a funding round led by Korean conglomerate Hanwha and Amer…

How A Grisly Injury Threw $5 Billion Drone Startup Shield AI Off Course
Shield AI executives overlooked its V-BAT drone’s safety hazards. With sales falling short and the CEO stepping down, the company is trying to win back military customers.

ExAct: A Video-Language Benchmark for Expert Action Analysis
Han Yi, Yulu Pan, Feihong He, Xinyu Liu, Benjamin Zhang, Oluwatumininu Oguntola, Gedas Bertasius
https://arxiv.org/abs/2506.06277

ExAct: A Video-Language Benchmark for Expert Action Analysis
We present ExAct, a new video-language benchmark for expert-level understanding of skilled physical human activities. Our new benchmark contains 3521 expert-curated video question-answer pairs spanning 11 physical activities in 6 domains: Sports, Bike Repair, Cooking, Health, Music, and Dance. ExAct requires the correct answer to be selected from five carefully designed candidate options, thus necessitating a nuanced, fine-grained, expert-level understanding of physical human skills. Evaluating…

PerfTracker: Online Performance Troubleshooting for Large-scale Model Training in Production
Yu Guan, Zhiyu Yin, Haoyu Chen, Sheng Cheng, Chaojie Yang, Tianyin Xu, Yang Zhang, Hanyu Zhao, Yong Li, Dennis Cai, Ennan Zhai
https://arxiv.org/abs/2506.08528

PerfTracker: Online Performance Troubleshooting for Large-scale Model Training in Production
Troubleshooting performance problems of large model training (LMT) is immensely challenging, due to unprecedented scales of modern GPU clusters, the complexity of software-hardware interactions, and the data intensity of the training process. Existing troubleshooting approaches designed for traditional distributed systems or datacenter networks fall short and can hardly apply to real-world training systems. In this paper, we present PerfTracker, the first online troubleshooting system utilizing…