No the prompt was just “have a chat with ChatGPT”
LLMS
-
LangGraph: Eight New Videos for Building AI Agents
By
–
We've added eight videos to help get started building with LangGraph Intro video to discuss what and why LangGraph is
Two different common agent run times
Five different examples of modifications to those agent runtimes -
Speculative Decoding: Optimizing Large Language Model Inference Efficiency
By
–
Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding Xia et al.: https://
arxiv.org/abs/2401.07851 #ArtificialIntelligence #DeepLearning #LargeLanguageModel -
32k Models Discussion and Model Capability Limitations
By
–
don't think you should get to work on that because there are a few 32k models in the replies here
-
Set-Based Language Models Without Positional Embeddings
By
–
actually I don't need anything with positional embeddings since i'm doing something weird, i just realized this, my objective is 'set based' so there's no ordering. so really i guess i could use most LMs with positional encoding disabled?
-
LangChain v1.4.10 Release: New Features and Improvements
By
–
1st release in 2024 v1.4.10: Global variables LCEL AstraDB Vectara MMR Langfuse SessionId Improvement on Airtable, Qdrant, PGVector, Milvus, LocalAI
-
LLMLingua: Microsoft Tool on Hugging Face
By
–
HF:
https://huggingface.co/spaces/microsoft/LLMLingua -
Long Context Encoder Models and Token Processing Optimization Techniques
By
–
anyone know a model that’s out there that’s encoder-only (BERT-like) but supports a really long context length? also what's the most efficient way of processing many tokens like this? i know about enabling FlashAttention & BetterTransformer. what else is out there?
-
Stanford Study Reveals Legal AI Errors Deepen Inequalities
By
–
A new study from HAI & Stanford RegLab shows pervasive errors in legal AI with a concerning implication: “Our findings suggest that the current limitations of LLMs pose a risk of further deepening existing legal inequalities.” (via @thehill
) -
Synthetic Data Generation for Olympiad-Level Mathematical AI Training
By
–
The lack of training data for maths, especially those at the Olympiad level, motivated us to generate everything synthetically. Key to this process is the idea of "symbolic deduction and traceback". We start from random diagrams, try to find all properties through forward