I'm doing a fireside chat with a Character AI tech lead on Tuesday in SF. This is what I'm planning to ask: – War stories from scaling to 30k QPS
– Why scaling LLM apps != scaling regular apps
– Internal tools built If this is interesting, come on by: https://
lu.ma/rujejluk
LLMS
-
Scaling LLM Apps to 30k QPS: Character AI Tech Lead Insights
By
–
-
DataLab: LLM-Powered Business Intelligence Platform with Agent Automation
By
–
9). DataLab – a unified business intelligence platform powered by LLM-based agents that integrates task planning, reasoning, and computational notebooks to streamline the entire BI workflow.
-
Procedural Knowledge in Pretraining Drives LLM Reasoning
By
–
10). Procedural Knowledge in Pretraining Drives Reasoning in LLMs – studies what documents in the pertaining influence model outputs; by looking at the pertaining data, it tries to understand better what kind of generalization strategies LLMs use to perform reasoning tasks.
-
Retrieval-Augmented Reasoning Enhances LLM Accuracy
By
–
8). Retrieval-Augmented Reasoning for LLMs – extends the rStar reasoning framework to enhance reasoning accuracy and factual reliability of LLMs; it leverages a Monte Carlos Tree Search (MCTS) framework with explicit retrieval-augmented reasoning to produce multiple candidate
-
ALAMA: Framework for Language Agents Learning Mechanisms
By
–
4). ALAMA – a new framework that helps language agents automatically learn when to use different mechanisms (ReAct, CoT, Reflection, etc.) for automatically completing tasks, improving on current approaches that use fixed or predefined mechanisms.
-
Auto-RAG: Autonomous Iterative Retrieval Model with Superior Performance
By
–
5). Auto-RAG – an autonomous iterative retrieval model with superior performance across many datasets; Auto-RAG is a fine-tuned LLM that leverages the decision-making capabilities of an LLM.
-
Reverse Thinking Improves LLM Reasoning Performance Efficiency
By
–
3). Reverse Thinking – shows that training LLMs to learn "reverse thinking" helps to improve performance in commonsense, math, and logical reasoning tasks. It claims to outperform a standard fine-tuning method trained on 10x more forward reasoning.
-
Top ML Papers Week: Genie 2, GenCast, OpenAI o1
By
–
It was a huge week of AI and LLM papers. Here are the top ML Papers of the Week (Dec 2-8): – Genie 2
– GenCast
– OpenAI o1
– Auto-RAG
– Reverse Thinking
– Retrieval-Augmented Reasoning for LLMs Read on for more: -
OpenAI o1 Model Series Demonstrates 50% Faster Reasoning Capabilities
By
–
1). OpenAI o1 – a model series trained with large-scale reinforcement learning to reason using chain of thought; o1 shows significant improvements across benchmarks related to math, code, and science; o1 is claimed to be 50% faster in generating thinking steps than o1-preview.
-
Groq API and Langflow: Build High-Performance Apps Visually
By
–
Go from idea to implementation with minimal code while maintaining high performance by harnessing the Groq API with an easy-to-build visual interface thanks to @langflow_ai
. Don't miss our upcoming webinar with @ozenhati & @MisbahSy
.
Register here: