DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads https://
arxiv.org/abs/2410.10819 https://
github.com/mit-han-lab/du
o-attention
… #MIT @songhan_mit
MACHINE LEARNING
-
DuoAttention: Efficient Long-Context LLM Inference with Retrieval
By
–
-
MIT KV Cache Framework Reduces LLM Decoding Latency
By
–
MIT HAN Lab introduce a framework that only applies a full KV cache to retrieval heads while using a light-weight, constant-length KV cache for streaming heads, which reduces both LLM's decoding and pre-filling memory and latency without compromising its long-context abilities.
-
Neural and Non-Neural AI: Reasoning, Transformers, and LSTMs
By
–
Neural and Non-Neural AI, Reasoning, Transformers, and LSTMs https://
bit.ly/3XPO9aB
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
10000 SLMs Fine-tuned on Predibase Platform
By
–
Over 10,000 #SLMs have been fine-tuned on Predibase! 🎉
— Predibase by Rubrik (@predibase) 23 octobre 2024
Why do leading AI companies like #Checkr, #Nubank and #Upstage fine-tune and #serve models on Predibase?
🎯 Better Accuracy: Fine-tuned models on Predibase beat hashtag#GPT4 by 5-20% (see our leaderboard:… pic.twitter.com/KR6blwhg9QOver 10,000 #SLMs have been fine-tuned on Predibase! Why do leading AI companies like #Checkr, #Nubank and #Upstage fine-tune and #serve models on Predibase? Better Accuracy: Fine-tuned models on Predibase beat hashtag#GPT4 by 5-20% (see our leaderboard:
-
Extending Fineweb filtering to 1000+ languages for pretraining
By
–
We want to extend up to 1000+ languages the data-driven filtering approach we used to create the *Fineweb* and *Fineweb-edu* large scale pretraining datasets The first step –which proved surprisingly difficult– was to find reliable high-early-signal evaluations in many languages
-
A/B Testing Rigorously in Machine Learning and AI
By
–
A/B Testing Rigorously (without losing your job) https://
bit.ly/3Yhg8Sl #AI #MachineLearning #DeepLearning #LLMs #DataScience -
Corporate Governance Embraces Artificial Intelligence Discussion
By
–
Corporate Governance Jumps Into The Artificial Intelligence Discussion
#AI #AIio #BigData #ML #NLU #Futureofwork @oluskayacan @richardkimphd @MariaFariello @johnpearcenews5 @ResurgentAV @rohanmuralee @globaliqx http://
ow.ly/r0RC30sH90z -
OpenAI Launches MLE-Bench for ML Engineering AI Agent Evaluation
By
–
Introducing MLE-bench, a new benchmark from @OpenAI to evaluate the performance of AI agents across 75 ML engineering tasks. Excited to have the authors discussing their work here! @junshernchan @thelokasiffers @JaffeOliver @jjamesaung @danesherbs @evanon0ping @ChowdhuryNeil
-
Iterative BC-Max Reduces Binary File Size via ML
By
–
Introducing Iterative BC-Max, a new technique that reduces the size of compiled binary files by optimizing inlining decisions. This approach offers several advantages over standard reinforcement learning algorithms. Read all about BC-Max at: https://
goo.gle/3AcnLjG -
Gartner’s Top 10 Strategic Technology Trends for 2025
By
–
Gartner Top 10 Strategic Technology Trends for 2025 Agentic AI
Post-quantum Cryptography
Spatial Computing
AI Governance Platforms
Ambient Invisible Intelligence
Polyfunctional Robots
Disinformation Security
Energy-Efficient Computing