CUTLASS Tutorial: Fast Matrix-Multiplication with WGMMA on NVIDIA® Hopper™ GPUs https://
bit.ly/4eKO530
#AI #MachineLearning #DeepLearning #LLMs #DataScience
LLMS
-
Fast Matrix Multiplication with WGMMA on Hopper GPUs
By
–
-
GPT-4 Outperforms Humans in Creativity and Clinical Reasoning
By
–
My bet is that in long run, AI will be beter at creativity than humans, including entrepreneurship. Related to this is this study that says: GPT-4 achieves 88% diagnostic accuracy, outperforming doctors by 15% in clinical reasoning test Another intepretation: GPT-4 alone was
-
ACL 2024 Presidential Address on NLP and AI now available
By
–
My #acl2024nlp Presidential Address is now publicly available. If you saw the slides & discussion of them in August, especially, please have a listen to the actual talk. It starts at 40'50" in this video: https://
underline.io/events/466/ses
sions/18203/lecture/104931-test-of-time-awards-lifetime-achievement-awards-presidential-address
… -
ML Experts Discuss Speculative Decoding for Fast Token Generation
By
–
🎥 Watch as two of our #ML experts discuss the nuances of #GenAI with @v_mohan_.
— SambaNova (@SambaNovaAI) 6 octobre 2024
In this episode, they talk about speculative decoding and its role in being able to achieve fast token generation. ⚡️
Accelerate your #dev journey ⤵️Watch as two of our #ML experts discuss the nuances of #GenAI with @v_mohan_
. In this episode, they talk about speculative decoding and its role in being able to achieve fast token generation. Accelerate your #dev journey -
Comprehensive Evaluation of OpenAI’s o1-preview LLM Model
By
–
9). Evaluation of o1 – provides a comprehensive evaluation of OpenAI's o1-preview LLM; shows strong performance across many tasks such as competitive programming, generating coherent and accurate radiology reports, high school-level mathematical reasoning tasks, chip design
-
LLM Reasoning Gaps in Grade-School Math Problem Solving
By
–
8). Not All LLM Reasoners Are Created Equal – investigates in depth the grade-school math problem-solving capabilities of LLMs; reports that LLMs show a significant gap in reasoning; finds that LLMs display a huge performance difference when solving compositional pairs and
-
o1-preview Analysis: Advanced Reasoning Models Show Similar LLM Trends
By
–
6). An Analysis of o1-preview – reports that large reasoning models like o1-preview, while improving on more difficult tasks, display similar qualitative trends as previous LLMs…
-
FRAMES: Framework for Evaluating LLM Factuality and Reasoning
By
–
7). FRAMES – a unified framework to evaluate an LLM’s ability to provide factual responses, assess retrieval capabilities, and the reasoning required to generate final responses…
-
Architecture Search Framework Optimizes LLM Inference Techniques
By
–
4). Architecture Search Framework for Inference-Time Techniques – introduces a modular framework for building and optimizing LLMs by combining multiple inference-time techniques…
-
RATIONALYST: Process Supervision Model for Reasoning Generalization
By
–
5). RATIONALYST – a model for process-supervision of reasoning that enables generalization across diverse reasoning tasks; this process is achieved with pre-training on a collection of 79k rationales from the Pile and a combination of reasoning datasets with minimal human