🌟 Welcome to NICE Talk 124! 🥳 Dynamic Multimodal Latent Reasoning Framework --> #Training-Free" + "Self-#Adapting" 📌 Register: luma.com/19jp1nw4?tk=FP… 📌 YouTube livestream: youtube.com/live/g0b80rns5… 🧠 Join us as we dive into "#Reasoning Within the Mind": Dynamic

✨Welcome to NICE Talk 126! (Chinese Talk) 🎭 Specific #LLM has Specific #Reasoning ➡️ One general #RL method to fit them all 📌 Register: luma.com/na6v3hxc 📌 YouTube livestream and video summaries: youtube.com/watch?v=0vib-9… 🧠Join us as we dive into "Bottom-up Policy

🤩 NICE Talk 127 ⭐️#Al Agents Through 20+ Real-World Case Studies⭐️ 📌 Stream it live — no app needed, click register and watch: luma.com/cfezxymd 🧐 How to turn AI agents into real-world production-level systems? ⚠️6️⃣8️⃣% of agents fail after 10 steps without #human

🌟 Welcome to NICE Talk 128! 🚀 Dynamic Large Concept Models ➡️ Adaptive Semantic #Reasoning Beyond Token-Level Computation 📌 Register: luma.com/7y9z666u 📌 YouTube livestream and video summaries: youtube.com/live/UfZrHRL7K… 🧠 Join us as we dive into "Toward Adaptive

Why is the gap between Agent "Demo" and "Production" so wide? 🚨 68% of AI Agents fail after 10 steps without human intervention. Building Agents is easy. Making them reliable is hard. We analyzed 306 industry practitioners & 20+ case studies to find out why: • Over-reliance

VLA (#Vision Language Action Model) and World Model: The future or a bubble? Is it the brain (algorithm), the body (hardware), or just the data that robots lack? 🗓️ LIVE TODAY! ⏰Pacific Time: 26.01.24 (Sat) 18:00 Our podcast will discuss the main questions about #Embodied #AI,

🚀 Welcome to NICE Talk 130! (Chinese Talk) 🎯 Reverse-Engineered Reasoning for Open-Ended Generation 📝 How to build high-quality reasoning chains without verifiable rewards? 📌 Register: luma.com/atv9qtu8 📌 YouTube livestream and video summaries:

🌟 Welcome to NICE Talk 131 | Agent Memory Self-Evolution 🚀 This is Era of Experience for AI Agent. The core is not the simple replay of past episodes, but whether they can, through runtime learning, transform accumulated experience into a self-evolving drive for tackling

We invite Jiaqian Wang, a PhD student at Xidian University, to discuss Agent Memory Self-Evolution. 😊USA Eastern Standard Time: 2026.01.31 (Sat) 21:30 The core is not the simple replay of past episodes, but whether they can, through runtime learning, transform accumulated

🌟 Welcome to NICE Talk 132 (Chinese Talk) | LLM Optimizer: From AdamW to Muon 🚀 Decoupling and controllability are the essential demands of scaling. The evolution of optimizers is critical for the future of large language models. Join us as we explore the transition from AdamW

🌟 Welcome to NICE AI Talk 135 (Chinese Talk) | Can LLMs Truly Build a Complete Project Repository from Scratch?🌟 🚀 Recent advances in code generation have delivered impressive results on short-horizon tasks like function synthesis and local code completion. But a fundamental

🌟 NICE AI Talk 136 (Chinese Talk) | Where Does the Slash in Attention Matrices Come From? 🌟 🚀 What causes the “slash pattern” in LLM attention heatmaps? This work defines these patterns as Slash-Dominant Heads (SDHs), which focus attention along fixed positional offsets. This

NICE Talk 137🌟invites Dr. Zhenrui (Zhenrui Yue) to discuss agent self-evolution without specific training data. Time⏰PST 02.13 18:30-19:30 Watch through this link: youtube.com/watch?v=lStHXZ… Or register with a time reminder through this link: luma.com/hf09u5ty Feel free

NICE Talk 138 👏 invites Qizheng Zhang Qizheng Zhang to discuss how models achieve continuous learning without fine-tuning. Time ⏰ EST 02.27 21:30–22:30 PST 02.27 18:30–19:30 📌Watch through this link: youtube.com/watch?v=d5QMyO… 📌 Or register with a time reminder through this