Advertise on podcast: Deep Papers

We dive into the latest paper from Google and a team of academic researchers: "TUMIX: Multi-Agent Test-Time Scaling with Tool-Use Mixture." Hear from one of the paper's authors — Yongchao Chen, Research Scientist — walks through the research and its implications.  The paper proposes Tool-Use Mixture (TUMIX), an ensemble framework that runs multiple agents in parallel, each employing distinct tool-use strategies and answer paths. Agents in TUMIX iteratively share and refine responses based on the question and previous answers. In experiments, TUMIX achieves significant gains over state-of-the-art tool-augmented and test-time scaling methods. Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

In our latest paper reading, we had the pleasure of hosting Grégoire Mialon — Research Scientist at Meta Superintelligence Labs — to walk us through Meta AI’s groundbreaking paper titled “ARE: scaling up agent environments and evaluations" and the new ARE and Gaia2 frameworks. Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

Santosh Vempala, Frederick Storey II Chair of Computing and Distinguished Professor in the School of Computer Science at Georgia Tech, explains his paper co-authored by OpenAI's Adam Tauman Kalai, Ofir Nachum, and Edwin Zhang. Read the paper: Sign up for future AI research paper readings and author office hours. See LLM hallucination examples here for context. Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

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 designed to evaluate LLMs on summarizing real-world evidence from structured study output — an important but often under-tested scenario compared to the typical “summarize this PDF” task. Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

This episode dives into "Category-Theoretic Analysis of Inter-Agent Communication and Mutual Understanding Metric in Recursive Consciousness." The paper presents an extension of the Recursive Consciousness framework to analyze communication between agents and the inevitable loss of meaning in translation. We're thrilled to feature the paper's author, Stan Miasnikov, Distinguished Engineer, AI/ML Architecture, Consumer Experience at Verizon, to walk us through the research and its implications. Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

We had the privilege of hosting Peter Belcak – an AI Researcher working on the reliability and efficiency of agentic systems at NVIDIA – who walked us through his new paper making the rounds in AI circles titled “Small Language Models are the Future of Agentic AI.” The paper posits that small language models (SLMs) are sufficiently powerful, inherently more suitable, and necessarily more economical for many invocations in agentic systems, and are therefore the future of agentic AI. The authors’ argumentation is grounded in the current level of capabilities exhibited by SLMs, the common architectures of agentic systems, and the economy of LM deployment. The authors further argue that in situations where general-purpose conversational abilities are essential, heterogeneous agentic systems (i.e., agents invoking multiple different models) are the natural choice. They discuss the potential barriers for the adoption of SLMs in agentic systems and outline a general LLM-to-SLM agent conversion algorithm. Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

In this AI research paper reading, we dive into "A Watermark for Large Language Models" with the paper's author John Kirchenbauer.  This paper is a timely exploration of techniques for embedding invisible but detectable signals in AI-generated text. These watermarking strategies aim to help mitigate misuse of large language models by making machine-generated content distinguishable from human writing, without sacrificing text quality or requiring access to the model’s internals. Learn more about the A Watermark for Large Language Models paper.  Learn more about agent observability and LLM observability, join the Arize AI Slack community or get the latest on LinkedIn and X. Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

The authors of the new paper *Self-Adapting Language Models (SEAL)* shared a behind-the-scenes look at their work, motivations, results, and future directions. The paper introduces a novel method for enabling large language models (LLMs) to adapt their own weights using self-generated data and training directives — “self-edits.” Learn more about the Self-Adapting Language Models paper. Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

This week we discuss The Illusion of Thinking, a new paper from researchers at Apple that challenges today’s evaluation methods and introduces a new benchmark: synthetic puzzles with controllable complexity and clean logic.  Their findings? Large Reasoning Models (LRMs) show surprising failure modes, including a complete collapse on high-complexity tasks and a decline in reasoning effort as problems get harder. Dylan and Parth dive into the paper's findings as well as the debate around it, including a response paper aptly titled "The Illusion of the Illusion of Thinking." Read the paper: The Illusion of Thinking Read the response: The Illusion of the Illusion of Thinking Explore more AI research and sign up for future readings  Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

We discuss Accurate KV Cache Quantization with Outlier Tokens Tracing, a deep dive into improving the efficiency of LLM inference. The authors enhance KV Cache quantization, a technique for reducing memory and compute costs during inference, by introducing a method to identify and exclude outlier tokens that hurt quantization accuracy, striking a better balance between efficiency and performance. Read the paperAccess the slides Read the blogJoin us for Arize ObserveLearn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

In this week's episode, we talk about Elastic Reasoning, a novel framework designed to enhance the efficiency and scalability of large reasoning models 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 constraints. Our discussion explores how Elastic Reasoning contributes to more concise and efficient reasoning, even in unconstrained settings, and its implications for deploying LRMs in resource-limited environments. Read the paper  Join us live Read the blog Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

What if your LLM could think ahead—preparing answers before questions are even asked? In this week's paper read, we dive into a groundbreaking new paper from researchers at Letta, introducing sleep-time compute: a novel technique that lets models do their heavy lifting offline, well before the user query arrives. By predicting likely questions and precomputing key reasoning steps, sleep-time compute dramatically reduces test-time latency and cost—without sacrificing performance. ​We explore new benchmarks—Stateful GSM-Symbolic, Stateful AIME, and the multi-query extension of GSM—that show up to 5x lower compute at inference, 2.5x lower cost per query, and up to 18% higher accuracy when scaled. ​You’ll also see how this method applies to realistic agent use cases and what makes it most effective.If you care about LLM efficiency, scalability, or cutting-edge research. Explore more AI research, or sign up to hear the next session live.  Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

For this week's paper read, we dive into our own research. We wanted to create a replicable, evolving dataset that can keep pace with model training so that you always know you're testing with data your model has never seen before. We also saw the prohibitively high cost of running LLM evals at scale, and have used our data to fine-tune a series of SLMs that perform just as well as their base LLM counterparts, but at 1/10 the cost. So, over the past few weeks, the Arize team generated the largest public dataset of hallucinations, as well as a series of fine-tuned evaluation models. We talk about what we built, the process we took, and the bottom line results. You can read the recap of LibreEval here. Dive into the research, or sign up to join us next time.  Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

This week we talk about modern AI benchmarks, taking a close look at Google's recent Gemini 2.5 release and its performance on key evaluations, notably  Humanity's Last Exam (HLE). In the session we covered Gemini 2.5's architecture, its advancements in reasoning and multimodality, and its impressive context window. We also talked about how benchmarks like HLE and ARC AGI 2 help us understand the current state and future direction of AI. Join us for the next live recording, or check out the latest AI research.  Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.

We cover Anthropic’s groundbreaking Model Context Protocol (MCP). Though it was released in November 2024, we've been seeing a lot of hype around it lately, and thought it was well worth digging into.  Learn how this open standard is revolutionizing AI by enabling seamless integration between LLMs and external data sources, fundamentally transforming them into capable, context-aware agents. We explore the key benefits of MCP, including enhanced context retention across interactions, improved interoperability for agentic workflows, and the development of more capable AI agents that can execute complex tasks in real-world environments. Read our analysis of MCP on the blog, or dive into the latest AI research.  Learn more about AI observability and evaluation, join the Arize AI Slack community or get the latest on LinkedIn and X.