Tootfinder

Consolidating Reinforcement Learning for Multimodal Discrete Diffusion Models
Tianren Ma, Mu Zhang, Yibing Wang, Qixiang Ye
https://arxiv.org/abs/2510.02880 https://

Consolidating Reinforcement Learning for Multimodal Discrete Diffusion Models
Optimizing discrete diffusion model (DDM) with rewards remains a challenge: the non-autoregressive paradigm makes importance sampling intractable and rollout complex, puzzling reinforcement learning methods such as Group Relative Policy Optimization (GRPO). In this study, we introduce MaskGRPO, the first viable approach to enable scalable multimodal reinforcement learning in discrete diffusion with effective importance sampling and modality-specific adaptations. To this end, we first clarify th…

The Path of Self-Evolving Large Language Models: Achieving Data-Efficient Learning via Intrinsic Feedback
Hangfan Zhang, Siyuan Xu, Zhimeng Guo, Huaisheng Zhu, Shicheng Liu, Xinrun Wang, Qiaosheng Zhang, Yang Chen, Peng Ye, Lei Bai, Shuyue Hu
https://arxiv.org/abs/2510.02752

The Path of Self-Evolving Large Language Models: Achieving Data-Efficient Learning via Intrinsic Feedback
Reinforcement learning (RL) has demonstrated potential in enhancing the reasoning capabilities of large language models (LLMs), but such training typically demands substantial efforts in creating and annotating data. In this work, we explore improving LLMs through RL with minimal data. Our approach alternates between the LLM proposing a task and then attempting to solve it. To minimize data dependency, we introduce two novel mechanisms grounded in self-awareness: (1) self-aware difficulty predi…

In 2025 people posted TikToks where they try out an addictive kratom-based drink.
It's so weird that, for all of the rhetoric of the last few years, there are no real moves in the US to control it.
https://youtu.be/AjSTaKgByZg?si=prVZxXQo3k5AeMJM&t=745

🇺🇦 #NowPlaying on #BBC6Music's #SteveLamacq
HighSchool:
🎵 Sony Ericsson
#HighSchool
#newRelease 🆕 single
https://highschool8.bandcamp.com/track/sony-ericsson
https://open.spotify.com/track/4mw4weFq7yJMmS40MtWf40

Finally found some time to look into the QGISMCP plugin.
The original connects #QGIS :qgis: to #Claude through the Model Context Protocol #MCP. I rewired it to work with a local European alternative: the new m…

GitHub - anitagraser/qgis_mcp: Model Context Protocol (MCP) that allows LLMs to use QGIS Desktop
Model Context Protocol (MCP) that allows LLMs to use QGIS Desktop - anitagraser/qgis_mcp

🇺🇦 #NowPlaying on KEXP's #ElSonido
Kevin Kaarl:
🎵 no me culpes por sentir
#KevinKaarl
https://open.spotify.com/track/2XgCopGQ3HzMhxNPOYrW85

Heredity: A Special Issue of American Quarterly
https://ift.tt/5Qwxm3g
updated: Tuesday, October 21, 2025 - 4:13pmfull name / name of organization: American Quarterly…
via Input 4 RELCFP htt…

Heredity: A Special Issue of American Quarterly
https://ift.tt/5Qwxm3g
updated: Tuesday, October 21, 2025 - 4:13pmfull name / name of organization: American Quarterly…
via Input 4 RELCFP htt…

🇺🇦 #NowPlaying on KEXP's #JazzTheatre
Gšbor Szabó:
🎵 Autumn Leaves / Speak to Me of Love
#GáborSzabó
https://open.spotify.com/track/2q639KZYXn5Z4c4hLfVFCX

Journal of the Australian Catholic Historical Society
https://ift.tt/hq3XMO1
Mining the Logs: Sources on Blue Humor URL …
via Input 4 RELCFP https://ift.tt/qUQMEax