m_o_m_e_n_t_e
@m_o_m_e_n_t_e
Followers
624
Following
10K
Media
2K
Statuses
52K
https://t.co/Ua2sbF4Ioa Mein neues Tool trainiert Theory-of-Mind-Modelle - OHNE Coding! 🧠 Ekman-Emotionen + ToM-Forschung 📊 12.000+ wissenschaftliche Datensätze 🚀 Live-Training mit Echtzeit-Charts #KI #TheoryOfMind #MachineLearning #Ekman #AITraining
2
1
2
Lebensaufgabe für Heute. 1) TOM-9D-Modelle trainieren 2) Das Lippenlesen-Modell komplett neu aufsetzen. (Shice - ich dreh durch, denn (2) funktioniert tatsächlich!)
0
0
0
Die 40k - Marke für die textbasierte TOM/9D-KI geknackt. Setze gleich einen neuen Trainingslauf an.
0
0
0
OPEN DRONE DETECTION - Open Acoustic CUAS Acoustic-Drones-Detection by orcohen9826 https://t.co/mRtWTCK0vp I see too little material & here too we've pushed enough I'd say... #Jammer #Jamming #Spoofer #Spoofing #EW #ElectronicWarfar
#Audio #Sound #Acoustic
#TDoA #AoA #DoA
DRONE DETECTION - Acoustic CUAS ARes https://t.co/mJba8ot1dG
#Jammer #Jamming #Spoofer #Spoofing #Audio #Sound #Acoustic #EW #ElectronicWarfar
13
168
826
World Model from Runway. Wow. THIS is what I’ve been waiting for. Realtime world exploration. You’ll be able to play and experience anything. World Models are the future. https://t.co/g2gheEwz36
15
40
312
Lebensaufgabe für heute: (1) ~ 6.7k Audioauswertungen / Chunks in die neue + geile KI prügeln (2) Bunte Bilder + komische Texte für die dusselige Werbung basteln. ^^
0
0
0
Boah ist das geil.
Karhade et al., "Any4D: Unified Feed-Forward Metric 4D Reconstruction" Map Anything-style architecture for estimating everything geometry + scene flow. Impressive dense estimation results!
0
0
1
First 2D - UFM, then 3D - MapAnything, now 4D!! Check out our latest work on taking a step towards flexible metric 4D reconstruction 🚀 Stronger MapAnything checkpoints coming next week followed by another Any4D release 😉
Introducing Any4D, a unified transformer for fully feed-forward, dense, metric-scale 4D reconstruction from flexible inputs! Any4D regresses per-pixel motion + geometry across frames in one pass — 15× faster, 2–3× more accurate reconstructions ⚡📈 Details + code below 👇
2
14
126
So. Wir besitzen jetzt eine KI, die auf Basis v. Stimmproben, Gespräche etc. TOM / 9D-Analysen durchführen kann. Erste Resultate sind extrem gut.
0
0
0
[NeurIPS 2025] SPIRAL: Semantic-Aware Progressive LiDAR Scene Generation and Understanding https://t.co/dCBcFw1MBW Existing LiDAR generative models are limited to producing unlabeled LiDAR scenes, lacking any semantic annotations. Performing post-hoc labeling on these generated
2
13
84
https://t.co/pUGLohpxp3 My last “Floor Plan Visualizer” post on X hit 800k views... so I turned it into a Scenario app. Just upload your floor plan, click "Generate", and you’re done! Prompts are optional. Included in the free plan. Enjoy ☺️ and stay tuned, more to come.
app.scenario.com
Transform 2D floor plans into photorealistic 3D rendered concepts. Upload any floor plan and describe the interior style, and the AI creates realistic visualizations with proper perspective, lighti...
5
19
166
🚀 Qwen3-Omni-Flash just got a massive upgrade (2025-12-01 version) ! What's improved: 🎙️ Enhanced multi-turn video/audio understanding - conversations flow naturally ✨ Customize your AI's personality through system prompts (think roleplay scenarios!) 🗣️ Smarter language
43
181
1K
A Python-based lightweight robot simulator designed for navigation, control, and reinforcement learning: [📍 Github Link] Most robotics simulators are powerful but heavy. Sometimes you just need a lightweight place to test navigation, control, or reinforcement learning without
7
148
864
data labeling is dead. long live distillation. from data to object detection endpoint in 90 seconds. link: https://t.co/ss4yC889KT
26
124
1K
We are releasing the Visual Geometry Benchmark (code + data) behind Depth Anything 3! Multi-view depth benchmarks often overweight overlapping regions due to higher point density. Ours treats the whole scene equally, providing a rigorous test for true geometry quality and
github.com
Depth Anything 3. Contribute to ByteDance-Seed/Depth-Anything-3 development by creating an account on GitHub.
5
49
314
MeshSplatting: Differentiable Rendering with Opaque Meshes Code： https://t.co/948RbEpz0I
#AI
github.com
Contribute to meshsplatting/mesh-splatting development by creating an account on GitHub.
0
1
12
Huawei just released EMMA: a unified multimodal AI for understanding, generation & editing This efficient architecture sets a new SOTA, outperforming larger models while handling complex image edits and diverse tasks. It's a leap for multimodal models!
2
20
183
Meier et al., "LeAD-M3D: Leveraging Asymmetric Distillation for Real-time Monocular 3D Detection" Blend two images together (mixup) and make the student detect 3D objects in both images, while the teacher sees the clean ones. Surprised that this sort of blending (+ more) works!
3
22
111
Qwen3 is Alibaba Cloud’s latest generation of Qwen large language models. And in this course, you'll learn how to train the LLM from scratch. You'll learn about configuring the model, training hyperparameters, RoPE positional embeddings, self-attention code, and lots more.
5
136
1K
Seltene Erden Schock für Deutschland. China vergibt erste Exportlizenzen und deutsche Firmen gehen komplett leer aus. Ein weiterer Beweis dafür, wie weit Deutschland in geopolitischen Fragen zurückgefallen ist. #China #Wadephul #Reuters #Deutschland Verlorene Einflussmacht:
reuters.com
German Foreign Minister Johann Wadephul appealed on Monday for an end to the uncertainty that has hit vital deliveries of Chinese semiconductors and rare earths to European manufacturers.
367
1K
4K
Lebensaufgabe für Heute. 15k Datensätze in das TOM/9D-Model klatschen u. einen (Halb)automatismus für die Generierung v. Ads, Kleinanzeigen etc. hochfahren. Erste manuelle Proben sind extrem gut.
0
0
0