1. Arize AX
  2. Phoenix OSS
  3. Docs
  4. Learn
  5. Company
  6. Pricing
Book a demo Sign in Sign up

Blog

Filter
close Clear all

How do you make an LLM, anyway? Microsoft just published a textbook.

Microsoft published a 109-page technical report on MAI-Thinking-1. Here’s the abbreviated version of how a modern lab actually trains a frontier reasoning model — from scraping the web to reinforcement learning, judges, and anti-cheating.

10 minutes read
By Laurie Voss | 10 minutes read
Generative AI Large Language Models LLM Evals Paper Readings

CUGA Agent: From Benchmarks to Business Impact of IBM’s Generalist Agent

This paper reading features several of the researchers — including Segev Shlomov‏ (PhD), Ido Levy, Asaf Adi, and Avi Yaeli — behind the widely acclaimed paper “From Benchmarks to Business Impact: Deploying IBM Generalist Agent in Enterprise Production.” The paper reports IBM’s experience developing and piloting the Computer Using Generalist Agent (CUGA), which has been…

34 seconds read
By David Burch | 34 seconds read
Paper Readings

ServiceNow’s Tara Bogavelli on AgentArch: Benchmarking AI Agents for Enterprise Workflows

In our latest AI research paper reading, we hosted Tara Bogavelli, Machine Learning Engineer at ServiceNow, to discuss her team’s recent work on AgentArch, a new benchmark designed to evaluate and compare AI agent architectures across real-world enterprise workflows.  As Tara described, the motivation behind AgentArch is to “move benchmarking closer to reality.” Instead of…

3 minutes read
By Julian Reeves | 3 minutes read
LLM Evals Paper Readings Research

OpenAI’s Santosh Vempala Explains Why Language Models Hallucinate

In our latest AI research paper reading, we hosted Santosh Vempala, Professor at Georgia Tech and co-author of OpenAI’s paper, “Why Language Models Hallucinate.” This paper offers one of the clearest theoretical explanations to date of why AI models make up facts — connecting hallucinations to classical misclassification theory and statistical calibration. The conversation explored…

4 minutes read
By Julian Reeves | 4 minutes read
Large Language Models Paper Readings Research

Atropos Health’s Arjun Mukerji, PhD, Explains RWESummary: A Framework and Test for Choosing LLMs to Summarize Real-World Evidence (RWE) Studies

Large language models are increasingly used to turn complex study output into plain-English summaries. But how do we know which models are safest and most reliable for healthcare?  In this most recent community AI research paper reading, Arjun Mukerji, PhD – Staff Data Scientist at Atropos Health – walks us through RWESummary, a new benchmark…

1 minute read
By Jason Lopatecki | 1 minute read
Paper Readings
Text reads: The Illusion of Thinking Understanding the Strengths and lImitations of Reasoning Models via the Lens of Problem Complexity

The Illusion of Thinking: What the Apple AI Paper Says About LLM Reasoning

A recent paper from Apple researchers—The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity—has stirred up significant discussion in the AI community. The authors argue that Large Reasoning Models (LRMs), which generate detailed “thinking traces” before answering questions, may not truly reason in the way we…

4 minutes read
By Jason Lopatecki | 4 minutes read
Paper Readings
Reads: Accurate KV Cache Quantization with Outlier Tokens Tracing

Accurate KV Cache Quantization with Outlier Tokens Tracing

Deploying large language models (LLMs) at scale is expensive—especially during inference. One of the biggest memory and performance bottlenecks? The KV Cache. In a new research paper, Accurate KV Cache Quantization with Outlier Tokens Tracing (OTT), researchers propose a smarter way to compress the KV Cache while preserving model quality. The result: up to 6.4×…

4 minutes read
By Jason Lopatecki | 4 minutes read
Paper Readings

Scalable Chain of Thoughts via Elastic Reasoning

This paper introduces Elastic Reasoning, a novel framework designed to enhance the efficiency and scalability of large reasoning models (LRMs) by explicitly separating the reasoning process into two distinct phases: thinking and solution. This separation allows for independent allocation of computational budgets, addressing challenges related to uncontrolled output lengths in real-world deployments with strict resource…

4 minutes read
By Sarah Welsh | 4 minutes read
Paper Readings
Load More

Subscribe to our resources and blogs

Subscribe