Top Tweets for #efficientAI
効率的なAIとは、小さなAIのことではない。無駄の少ないAIのことだ。
何を守り、何を削るのか。それを決めた者から、AIは軽くなる。
📄 効率的なAIの時代 - 学習から推論へのパラダイムシフト
https://t.co/bULshNqlkH
#EfficientAI #AI推論
Every AI answer has a physical cost.
Moving data takes energy, memory has limits, and computation takes time.
Designing intelligence means deciding what to remember, when to compute and when more work is worth its cost.
#AI #ComputerScience #EfficientAI
🎉 Our paper has been accepted at NeurIPS 2026 GlobalSouthAI Workshop!
📄 Scaling Vision Transformers Without Training: Prune-and-Merge Token Compression
Excited to share this work! 🚀
#NeurIPS2026 #ComputerVision #DeepLearning #EfficientAI
We just shipped Plivo integration for @AiEfficient 🎉
Real phone calls → your open-source voice AI eval stack.
- Pull call recordings straight from Plivo
- Run synthetic evals using your Plivo numbers (inbound + outbound)
#VoiceAI #OpenSource #Plivo #EfficientAI #AIAgents
Dense LLMs activate all parameters for every token.
MoE activates only relevant experts.
This selective activation delivers strong performance at lower cost.
#open #MoE #LLM #AIArchitecture #EfficientAI #hiring #lagos #osint
More companies are shifting toward smaller, specialized models that run locally for better privacy and lower costs.
#ArtificialIntelligence #EdgeAI #PrivacyFirst #EfficientAI
😂 Early AI recipe generators invented horrors like "Beothurtreed Tuna Pie." AI creativity is wild... and sometimes terrifying for your taste buds. 🍔🤢
#AI #WeirdFood #FunnyAI #ArtificialIntelligence
Excited to share that our paper EGM: Efficient Visual Grounding Language Models has been accepted to ECCV 2026！
arXiv: https://t.co/AVO8vAs06J
GitHub: https://t.co/jLBlV4odZy
Project: https://t.co/q5f5vd2jQU
#ECCV2026 #VLM #VisualGrounding #EfficientAI
𝗝𝗼𝗶𝗻 𝗡𝗼𝘁𝗮 𝗔𝗜’s 𝗞𝗼𝗿𝗲𝗮 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁 𝗗𝗮𝘆𝘀 𝗗𝘂𝗿𝗶𝗻𝗴 𝗜𝗖𝗠𝗟 𝟮𝟬𝟮𝟲
Held alongside #ICML2026, Korea Efficient Days brings together researchers, engineers, and industry professionals interested in the future of #EfficientAI.
👉 https://t.co/71VSSFaI5L
⚡️ ZAYA1-8B proves small can dominate—8B total, <1B active, outperforms larger models in math, coding & reasoning. 🚀 Like, comment, and join Newline AI Accelerator today!
#ai #efficientAI #smallLLM #reasoningAI #MoE
On-device AI is exploding — faster, private, and running locally without cloud dependency. Smaller, efficient models are outperforming giants in real-world use. Privacy + speed = the new standard. Edge computing just got intelligent.
#EdgeAI #OnDeviceAI #EfficientAI #AITrends
🎉 Congrats to Sudo (https://t.co/rsZlbDYPoG) on getting Tempora accepted at #ICML2026!
📄 Paper: https://t.co/ifCcH8H7h6
Proud to have been part of the journey as a collaborator/mentor alongside @cecim! 🙌
#TestTimeAdaptation #TTA #EfficientAI #ICML2026
Grateful to my mentors @nrsuri and @soumyabrata_pal, and @AdobeResearch for supporting this work.
Paper link coming soon. Any queries? Drop a comment 👇
#ACL2026 #LLM #ReasoningAI #EfficientAI #SpeculativeDecoding #InferenceOptimization #AdobeResearch
#NVIDIA’s #Nemotron 3 stands out for smart efficiency: Nano activates only 3.2B of 31B parameters, offers 3x throughput and 1M context, challenging bigger-is-better assumptions in AI.
https://t.co/1F9MaSOHI9
#NVIDIA #Nemotron3 #AIModels #EfficientAI #GenerativeAI
More details, experiments, and implementation are available here:
https://t.co/L3TK7ajrnj
https://t.co/GbjLVhLOPw
Joint work with Yehonathan Refael, Jonathan Svirsky, Boris Shustin, and Wasim Huleihel
#MachineLearning #DeepLearning #LLMs #EfficientAI
🚀 Call for Participation: @iwslt Model Compression 2026
Make large multilingual foundation models small ⚡ without losing power in EN→DE/ZH speech-to-text translation.
📅 Evaluation: Apr 1–15
https://t.co/LabFdqa4n2
#IWSLT2026 #SpeechAI #Qwen2 #EfficientAI
Very thrilled to announce our paper “Towards Sparse Video Understanding and Reasoning” has been accepted to CVPR 2026!
Please feel free to check our paper and contact us if you have any comments or questions!
Paper: https://t.co/aom8tdLFp8
#CVPR2026 #VLM #EfficientAI
🔥Two AAAI’26 papers on making large models adapt faster & deploy cheaper:
🔹 TR-DQ — Time-Rotation Diffusion Quantization
🔹 ICM-Fusion — In-Context Meta-Optimized LoRA Fusion for multi-task adaptation
📍 In Singapore this week — happy to meet & discuss.
#AAAI26 #EfficientAI
Noeum-1-Nano is live
Competitive AI model trained on just 18B tokens (vs 2-12T standard)
- 20-667× more data-efficient
- Matches major lab nano models
- Built entirely from scratch
- Solo research from Vienna
Model: https://t.co/Oju2QYU7tg
#AI #EfficientAI #SovereignAI
Last Seen Hashtags on Sotwe
福利姬
Seen from Japan
incestoconlahija
Seen from Ecuador
แอบถ่ายสาวอาบน้ำ
Seen from Thailand
minichat(***)************************+filter:native_video
Seen from Canada
แอบมีใจ
Seen from Thailand
Ahmak
Seen from United States
minichat #snapchat #candid ()*****************
Seen from Netherlands
GhanimAlMuftah
膨体化
Seen from Peru
inktober2026smack
Seen from France
Trends for you
Most Popular Users
Elon Musk 
@elonmusk
241.7M followers
Barack Obama
@barackobama
119M followers
Cristiano Ronaldo
@cristiano
114.6M followers
Donald J. Trump
@realdonaldtrump
111.9M followers
Narendra Modi
@narendramodi
107.2M followers
Rihanna
@rihanna
98.7M followers
NASA
@nasa
92.4M followers
Justin Bieber
@justinbieber
91.8M followers
KATY PERRY
@katyperry
90.1M followers
Taylor Swift
@taylorswift13
84M followers
Lady Gaga
@ladygaga
75.5M followers
Virat Kohli
@imvkohli
73.5M followers
Kim Kardashian
@kimkardashian
70.9M followers
YouTube
@youtube
68.8M followers
Neymar Jr
@neymarjr
66.5M followers
Bill Gates
@billgates
65.2M followers
Selena Gomez
@selenagomez
63.1M followers
The Ellen Show
@theellenshow
62.3M followers
CNN
@cnn
61.8M followers
X
@x
60.7M followers