Tootfinder

MK2 at PBIG Competition: A Prompt Generation Solution
Yuzheng Xu, Tosho Hirasawa, Seiya Kawano, Shota Kato, Tadashi Kozuno
https://arxiv.org/abs/2507.08335

MK2 at PBIG Competition: A Prompt Generation Solution
The Patent-Based Idea Generation task asks systems to turn real patents into product ideas viable within three years. We propose MK2, a prompt-centric pipeline: Gemini 2.5 drafts and iteratively edits a prompt, grafting useful fragments from weaker outputs; GPT-4.1 then uses this prompt to create one idea per patent, and an Elo loop judged by Qwen3-8B selects the best prompt-all without extra training data. Across three domains, two evaluator types, and six criteria, MK2 topped the automatic le…

Low-latency neutrino follow-up combining diverse IceCube selections
Christoph Raab (for the IceCube Collaboration), Sam Hori (for the IceCube Collaboration), Steve Sclafani (for the IceCube Collaboration), Jessie Thwaites (for the IceCube Collaboration)
https://arxiv.org/abs/2507.08748

Low-latency neutrino follow-up combining diverse IceCube selections
Neutrino observations are crucial for multi-messenger astronomy, but currently limited by effective area and atmospheric background. However, while other telescopes must point their limited field of view, IceCube can perform full-sky realtime follow-up of astrophysical transient sources with high uptime. For this, IceCube uses its Fast Response Analysis (FRA), which can provide results from an unbinned maximum likelihood analysis within tens of minutes of an astrophysical transient. Besides man…

Enhancing Video-Based Robot Failure Detection Using Task Knowledge
Santosh Thoduka, Sebastian Houben, Juergen Gall, Paul G. Pl\"oger
https://arxiv.org/abs/2508.18705 https:…

Enhancing Video-Based Robot Failure Detection Using Task Knowledge
Robust robotic task execution hinges on the reliable detection of execution failures in order to trigger safe operation modes, recovery strategies, or task replanning. However, many failure detection methods struggle to provide meaningful performance when applied to a variety of real-world scenarios. In this paper, we propose a video-based failure detection approach that uses spatio-temporal knowledge in the form of the actions the robot performs and task-relevant objects within the field of vi…