Shows
LessWrong (30+ Karma)“Do LLMs Have Desires?” by Christopher Ackerman Work conducted with Yujun Zhou (yzhou25@nd.edu) and supported by SPAR TL;DR: In paired-choice paradigms, LLMs report consistent preferences over outcomes (e.g., types and number of lives saved, types of policies enacted)Some have suggested that this indicates that LLMs have human-like value systemsWe design an experimental framework where LLMs are able to modulate their output quality based on prompt contextWe find that LLMs modulate their output quality in response to effort exhortations, role-play instructions, and harmfulness cues, but NOT to opportunities to achieve the outcomes they report preferring in the paired-choice experimentsWe...
2026-06-2814 min
Un truco al día de Google+IA de Antonio Glez. +1 millón de visitas/mes👉trei.es/info. 19 años en SEOArchivo LLMS.txt ¿Google está a favor o en contra?Truco diario 429: Google y el archivo LLMS.txtComentario recibido por WhatsApp:Discrepancia de ideas entre Google Search y Google Chrome sobre el famoso llms.txtDocumentación oficial de Google, menciona literalmente esto en la sección de mitos:- Archivos LLMS.txt y otro etiquetado "especial": no es necesario crear archivos legibles por máquinas, archivos de texto de IA, etiquetado, ni Markdown para que aparezcan en la búsqueda con IA generativa. Ten en cuenta que Goog...
2026-06-2211 min
James Dooley PodcastGoogle AI Overviews - Get LLMs to Recommend You 24/7 (James Dooley x Chris Munch from Ampifire)In this episode, James Dooley interviews Chris Munch (Amplifier) about how brands can get cited in Google AI Overviews and LLM-driven answers (ChatGPT, Gemini, Claude, etc.). Chris explains why GEO/AI visibility overlaps with SEO, but requires stronger brand/entity signals, third-party mentions, and structured information that machines can trust. They break down why AI brings back demand for highly specific answers, how to build consensus across platforms using multi-format content (articles, video, podcasts, images, PDFs), and why e-commerce brands should prioritise product schema + Google Shopping feeds for AI shopping recommendations. The discussion also covers practical content strategy for...
2026-06-1954 min
James Dooley PodcastPersonalizing LLMs: How Your ChatGPT & Gemini Differ | James Dooley x Benjamin TannenbaumJames Dooley speaks with Benjamin Tannenbaum from AISO about the personalisation of large language models and why identical prompts in ChatGPT, Gemini and other AI search systems can produce different answers. Benjamin Tannenbaum explains the difference between true personalisation and simple probabilistic variability, showing how response volatility often comes from token selection rather than user history. The conversation breaks down location signals, memory settings, login state, and how query fan out becomes weighted by personal attributes. They also explore why personalisation can reduce randomness by narrowing candidate sets, how share of voice replaces fixed rankings in AI visibility tracking...
2026-06-1915 min
James Dooley PodcastDo LLMs Use Metadata or Page Content? James Dooley Interviews Sergey LucktinovJames Dooley is joined by Sergey Lucktinov to explain how large language models retrieve information during AI searches. They break down the full retrieval pipeline, from metadata-only filtering to light skimming and deep page parsing. The discussion clarifies when LLMs rely on meta titles and descriptions, when pages are never opened, how schema markup is interpreted, and how knowledge vault answers bypass search entirely. This episode gives SEOs and marketers a clear framework for optimising content to survive each LLM retrieval stage.Where to Listen to This EpisodeDo LLMs Use Metadata...
2026-06-1908 min
The AI BriefingWhen NOT to Use LLMs: Choosing the Right AI Tool for Your Data PipelineIn this episode of the AI Briefing, Tom challenges the LLM hype cycle and explains why traditional machine learning models and statistical approaches often outperform large language models for data processing tasks. Learn when to use LLMs appropriately versus more efficient, cost-effective alternatives.Episode Show NotesKey Topics CoveredThe LLM Hype Cycle Reality CheckWhy LLMs aren't always the answer for data processingThe hidden costs of using LLMs for inappropriate tasksUnderstanding when simpler solutions outperform complex AITraditional AI & ML Still MatterStatistical models and their advantages...
2026-06-1803 min
Our Lives With BotsThe dark side of personalization in LLMsWhy does your version of ChatGPT tell you lies, but others' ChatGPT tells them the truth - for the same prompt?In other words, what is personalization in LLMs, and why should you care about it? Hint: it's much more opaque than customizing your chatbot in your custom prompt settings, and potentially much more harmful. Also, any LLM you use (ChatGPT, Gemini, Claude) does automatic personalization behind the scenes.According to our expert, Dr. Angelina Wang (https://angelina-wang.github.io/), an assistant professor at Cornell Tech in computer science, personalization might mean that your chatbot...
2026-06-0930 min
Hear every story in its fullest formLarge Language Models (LLMs)Visit audiobookzap.com to sign up to the Audible 30-day free trial to listen to full audiobooks for free!Title: Large Language Models (LLMs)Subtitle: A Comprehensive Guide to Understanding, Building, and Applying Large Language Models Across DomainsCategories: Artificial Intelligence, Computer Science, TechnologyAuthor: Sam MileyNarrator: Maha AmeerPublisher: May JuneRelease date: 02-24-2026Language: EnglishSummary:Large Language Models (LLMs) are revolutionizing the way we interact with technology—powering chatbots, automating content creation, summarizing information, writing code, and even generating art. Large Language Models (LL...
2026-02-2400 min
LLMs Research PodcastYour 70-Billion-Parameter Model Might Be 40% WastedYour 70-Billion-Parameter Model Might Be 40% WastedThree papers from February 1–6, 2026 converge on a question the field has been avoiding since 2016: what if most transformer layers aren't doing compositional reasoning at all, but just averaging noise?This video traces a decade of evidence, from Veit et al.'s original ensemble observation in ResNets through ShortGPT's layer pruning results and October 2025's formal proof, to three new papers that quantify the consequences. Inverse depth scaling shows loss improves as D to the negative 0.30, worse than one-over-n. TinyLoRA unlocks 91% GSM8K accuracy by training just 13 parameters with RL. An...
2026-02-1112 min
LLMs Research PodcastThe Evolution of Long-Context LLMs: From 512 to 10M TokensThe podcast discusses the technical shift in large language models from a standard 512-token context window to modern architectures capable of processing millions of tokens. Initial growth was constrained by the quadratic complexity of self-attention, which dictated that memory and computational needs increased by the square of the sequence length. To address this bottleneck, researchers developed sparse attention patterns and hardware-aware algorithms like FlashAttention to reduce memory and computational overhead. Current iterations, such as FlashAttention-3, leverage asynchronous operations on high-performance GPUs to facilitate context lengths exceeding 128,000 tokens.Changes to positional encodings also proved necessary. Methods such as...
2026-01-2415 min
ChatGPT: News on Open AI, MidJourney, NVIDIA, Anthropic, Open Source LLMs, Machine LearningGoogle's Gemini for Test PracticeIn this episode, we cover Google's latest move: leveraging Gemini to provide free SAT practice tests. We compare this against other AI advancements, including ChatGPT and various LLMs.Chapters00:00 Intro & AIbox.ai Plug02:00 Google's Free SAT Prep05:09 AI in Education Debate12:14 Impact on Tutors & Google's StrategyIn this episode, we discuss Google's new initiative offering free SAT practice exams powered by Gemini, exploring how this advancement will impact education and the broader implications of AI in learning. We also examine the potential disruption to traditional education companies and the...
2026-01-2210 minChatGPT: News on Open AI, MidJourney, NVIDIA, Anthropic, Open Source LLMs, Machine LearningChatGPT and Its Energy ThirstIn this episode, we discuss the energy consumption of platforms like ChatGPT and the Trump administration's $15 billion power plant proposal for tech companies. We highlight the implications for open-source LLMs and machine learning infrastructure.Chapters00:00 AI's Power Hunger01:48 Trump Admin's Plan10:07 Energy Source Debate13:30 Accountability for ConsumptionIn this episode, we explore the Trump administration's proposal for tech companies to invest $15 billion in power plants to meet the surging electricity demands of AI and data centers. We also discuss its potential impact on America's power grid, consumer costs, and...
2026-01-2111 minChatGPT: News on Open AI, MidJourney, NVIDIA, Anthropic, Open Source LLMs, Machine LearningMeta's Focus: ChatGPT, AIIn this episode, we discuss Meta's strategic move away from the Metaverse to concentrate on AI, influencing ChatGPT and other LLMs. We explore the implications for OpenAI, MidJourney, and open-source machine learning.Chapters00:00 Meta's Metaverse Investment02:05 VR's Limited Appeal06:21 Meta's Financial Losses11:24 AI: Meta's New FrontierIn this episode, we break down Meta's decision to discontinue its ambitious Metaverse project after a colossal $73 billion investment, leading to significant layoffs and the shutdown of several VR game studios. We also explore why the Metaverse failed to gain traction and how...
2026-01-2013 min
The FP&A Guy NetworkHow Excel AI Agents Actually Work for Financial Modelers to Understand LLMs & Tools with Tim JacksIn this episode of The ModSquad, hosts Paul Barnhurst, Ian Schnoor, and Giles Male welcome Tim Jacks, founder of Taglo, for an insightful discussion on the integration of AI in financial modeling. Tim’s expertise bridges the worlds of financial modeling and AI, and in this episode, he shares his journey and discusses how AI is reshaping the financial modeling landscape.Tim Jacks is the founder of Taglo, a company dedicated to improving financial modeling with AI technology. His career journey spans financial consulting and software development, including building financial modeling tools. Over time, Tim's interest in ar...
2026-01-2043 minFinancial Modeler's CornerHow Excel AI Agents Actually Work for Financial Modelers to Understand LLMs & Tools with Tim JacksIn this episode of The ModSquad, hosts Paul Barnhurst, Ian Schnoor, and Giles Male welcome Tim Jacks, founder of Taglo, for an insightful discussion on the integration of AI in financial modeling. Tim’s expertise bridges the worlds of financial modeling and AI, and in this episode, he shares his journey and discusses how AI is reshaping the financial modeling landscape.Tim Jacks is the founder of Taglo, a company dedicated to improving financial modeling with AI technology. His career journey spans financial consulting and software development, including building financial modeling tools. Over time, Tim's interest in ar...
2026-01-2043 minChatGPT: News on Open AI, MidJourney, NVIDIA, Anthropic, Open Source LLMs, Machine LearningChatGPT Cracks Unsolved MathIn this episode, we report on ChatGPT's remarkable feat of solving a previously unsolved math problem, detailing its unique reasoning process. We discuss how this showcases the evolving capabilities of OpenAI's flagship model and other LLMs.Resources Mentioned12:48 AIbox.ai
See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
2026-01-1513 minChatGPT: News on Open AI, MidJourney, NVIDIA, Anthropic, Open Source LLMs, Machine LearningYann LeCun: LLMs Dead End Architecture for MetaDead end architecture LLMs Meta Yann LeCun declares incapable autonomous reasoning planning potently fundamentally. Pattern-matching obsession chains transformers lacking biological world models critically radically. Meta scientist ignites JEPA revolution dismantling trillion-parameter illusion disruptively comprehensively.Get the top 40+ AI Models for $20 at AI Box: https://aibox.aiAI Chat YouTube Channel: https://www.youtube.com/@JaedenSchaferJoin my AI Hustle Community: https://www.skool.com/aihustleSee Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
2026-01-0708 minChatGPT: News on Open AI, MidJourney, NVIDIA, Anthropic, Open Source LLMs, Machine LearningNvidia's Aggressive $1B+ AI Ecosystem InvestmentsAggressive investments $1B+ AI ecosystem Nvidia strategically dominates end-to-end compute potently disruptively. Strategic portfolio covers LLMs, edge computing, creative generation amplifying revenue streams globally comprehensively. Trillion-dollar empire accelerates leadership aggressively.Get the top 40+ AI Models for $20 at AI Box: https://aibox.aiAI Chat YouTube Channel: https://www.youtube.com/@JaedenSchaferJoin my AI Hustle Community: https://www.skool.com/aihustleSee Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.
2026-01-0713 min
Leading DetectionLeveraging LLMs in fighting fraudIn this episode of the Leading Detection podcast, host Matt speaks with Chen Zamir about the role of large language models (LLMs) in fraud detection. They discuss the current state of LLMs, their practical applications in automating fraud investigations, and the importance of human analysts in the process. Chen emphasises the need for trust in technology, the potential for LLMs to enhance existing fraud detection methods, and the challenges posed by biases in data. The conversation also touches on the evolving landscape of fraud detection tools and the necessity of safeguards when implementing new technologies.Key Takeaways• LLMs are au...
2025-12-0537 min
Deep Dive with GeminiAre LLMs glorified compression Algos ? A new take from ancient perspective !Research The debate over Large Language Models (LLMs) often uses the term "glorified compression algorithm" as a modern litmus test, separating reductionists who view LLMs as "blurry JPEGs" of the internet from proponents who see compression as the very proof of emergent intelligence. By synthesizing information theory with ancient philosophy, we find that LLMs are indeed powerful compression systems, but the nature of this compression elevates the process to a cognitive function.LLMs are mathematically optimized to minimize Cross-Entropy Loss, which is synonymous with minimizing the bits required to represent their training data. This process...
2025-11-2635 min
Lion's Share: The Research CastMKTG 556 | Session 9 | AI–Human Hybrids for Marketing Research: Leveraging Large Language Models (LLMs) as CollaboratorsMKTG 556 | Session 9 | AI–Human Hybrids for Marketing Research: Leveraging Large Language Models (LLMs) as Collaborators - 2025 Neeraj Arora, Ishita Chakraborty, and Yohei Nishimura Introduction: The authors' main idea is that a hybrid approach combining humans and large language models (LLMs) improves efficiency and effectiveness in marketing research. In qualitative research, they show that LLMs can help with both data generation and analysis; LLMs effectively create sample characteristics, generate synthetic respondents, and conduct and moderate in-depth interviews. The AI–human hybrid produces information-rich, coherent data that exceeds human-only data in depth and insightfulness and matc...
2025-11-1319 min
EDGE of the Web - The Best SEO Podcast for Today's Digital MarketerUnpacking LLMs.txt with Carolyn ShelbyErin welcomes Carolyn Shelby, the Principal SEO at Yoast and a renowned authority in technical and enterprise SEO. Carolyn brings decades of hands-on experience from her pioneering days in digital marketing, working with brands like Disney's ESPN, Tribune Publishing, and major nonprofits. The conversation kicks off with a surprising twist—Carolyn's unique title as Queen of the micronation Ladonia—before diving into her role at Yoast and their latest innovation: the LLMs.txt file generator. Carolyn explains how this new file helps websites communicate their most valuable content directly to large language models like ChatGPT and Google's AI...
2025-10-1642 min
Infinite Curiosity Pod with Prateek JoshiDiffusion LLMs - The Fastest LLMs Ever Built | Stefano Ermon, cofounder of Inception LabsStefano Ermon is the cofounder of Inception Labs and an associate professor at Stanford. Inception is developing a new type of AI models called Diffusion LLMs.Stefano's favorite book: If on a Winter's Night a Traveler (Author: Italo Calvino)(00:01) Introduction(00:38) What are autoregressive LLMs and how do they work(02:28) How diffusion LLMs rethink generation(04:02) The ceiling of autoregressive LLMs: cost, latency, reliability(06:19) Why diffusion LLMs are commercially viable now(09:12) Parallel refinement: how diffusion models generate text(12:05) Understanding diffusion steps and efficiency(13:49) Hardest engineering challenges at Inception(15:23) From...
2025-10-0939 min
Riedman Report: Risk, AI, Education, & SecurityEp 59. Can AI and LLMs like ChatGPT assess school shooting threats?Last week, I successfully defended my PhD dissertation. This episode is a short overview of my research on using LLMs like ChatGPT to assess threats of violence.My Dissertation: LLMs Versus Human Experts: Mixed Methods Analysis Measuring Variance in School Shooting Threat AssessmentsAbstract: This study measures the unwanted variability in expert judgments by testing six frontier large language models (LLMs) on fictitious but realistic school shooting scenarios derived from 1,000 real threats made in the United States (note: there are approximately 100,000 threats made nationwide to schools each year). Using prior decision science research, this dissertation measures...
2025-10-0327 min
Chaos Computer Club - recent events feedWhat opinions do LLMs/chatbots have about OpenStreetMap? (sotm2025)In this work we develop a reproducible pipeline for querying multiple LLMs/chatbots in order to access and analyse their opinion on OpenStreetMap by prompting these systems to answer a series of questions on OSM. People are turning to chatbots and LLMs for opinions and advice on practically every topic. We believe it is important that we begin to assess how chatbots and the LLMs provide information and opinion about OSM. Among other outputs, this work can providing evidence to the OSM community that can be used to shape future public engagement strategies about the project.
The work described in...
2025-10-0305 minChaos Computer Club - recent events feed (high quality)What opinions do LLMs/chatbots have about OpenStreetMap? (sotm2025)In this work we develop a reproducible pipeline for querying multiple LLMs/chatbots in order to access and analyse their opinion on OpenStreetMap by prompting these systems to answer a series of questions on OSM. People are turning to chatbots and LLMs for opinions and advice on practically every topic. We believe it is important that we begin to assess how chatbots and the LLMs provide information and opinion about OSM. Among other outputs, this work can providing evidence to the OSM community that can be used to shape future public engagement strategies about the project.
The work described in...
2025-10-0305 minChaos Computer Club - recent audio-only feedWhat opinions do LLMs/chatbots have about OpenStreetMap? (sotm2025)In this work we develop a reproducible pipeline for querying multiple LLMs/chatbots in order to access and analyse their opinion on OpenStreetMap by prompting these systems to answer a series of questions on OSM. People are turning to chatbots and LLMs for opinions and advice on practically every topic. We believe it is important that we begin to assess how chatbots and the LLMs provide information and opinion about OSM. Among other outputs, this work can providing evidence to the OSM community that can be used to shape future public engagement strategies about the project.
The work described in...
2025-10-0305 min
Chaos Computer Club - recent events feedWhat opinions do LLMs/chatbots have about OpenStreetMap? (sotm2025)In this work we develop a reproducible pipeline for querying multiple LLMs/chatbots in order to access and analyse their opinion on OpenStreetMap by prompting these systems to answer a series of questions on OSM. People are turning to chatbots and LLMs for opinions and advice on practically every topic. We believe it is important that we begin to assess how chatbots and the LLMs provide information and opinion about OSM. Among other outputs, this work can providing evidence to the OSM community that can be used to shape future public engagement strategies about the project.
The work described in...
2025-10-0305 minChaos Computer Club - recent events feed (low quality)What opinions do LLMs/chatbots have about OpenStreetMap? (sotm2025)In this work we develop a reproducible pipeline for querying multiple LLMs/chatbots in order to access and analyse their opinion on OpenStreetMap by prompting these systems to answer a series of questions on OSM. People are turning to chatbots and LLMs for opinions and advice on practically every topic. We believe it is important that we begin to assess how chatbots and the LLMs provide information and opinion about OSM. Among other outputs, this work can providing evidence to the OSM community that can be used to shape future public engagement strategies about the project.
The work described in...
2025-10-0305 min
Onchain Growth Club: Crypto & Web3 Marketing TalksFrom SEO to LLMs: The New Era of AI Marketing w/ Malte Landwehr, Peec AIMalte Landwehr, CMO of Peec AI, talks about the evolving landscape of marketing in the age of AI and large language models (LLMs). We discuss the transition from traditional SEO to AI-driven discoverability (GEO), the challenges of traffic attribution, and the current state of LLMs in the market. Malte shares insights on how brands can optimize their visibility in LLMs, the importance of understanding user behavior, and the practical tactics that can be employed to influence LLM outputs.—0:00 Introduction and Welcome04:53 Transitioning from SEO to AI a...
2025-09-3039 min
The Edward ShowLLMs.txt Is Dead? Why AI Bots Aren't ListeningE772: LLMs.txt was hyped as the next big thing in SEO - but are AI bots even using it? I break down fresh research from Adobe's SEO Strategist, Flavio Longato, that exposes the reality: GPT, Claude, and Perplexity aren't crawling your LLMs.txt at all. I'll share the data, explain why this "AI SEO" tactic is dead on arrival, and reveal two unique strategies that actually work to get your brand into AI search results. Whether you're a marketer, SEO, or business owner trying to future-proof your search strategy, this episode is y...
2025-08-1507 min
The Sports PR HuddleEpisode 9: Sports Comms Pros Need To Be Maximizing LLMsIn this episode of the Sports PR Huddle entitled"Sports Comms Pros Need To Be Maximizing LLMs", Ashley Mann, COO of The Colab, joins the show. Ashley discusses the transformative impact of Large Language Models (LLMs) on the sports public relations industry. After exploring Ashley's career journey now on the agency side, the discussion shifts to the importanceof authenticity in communication, and how LLMs can be leveraged for effective PR strategies. The conversation emphasizes the need for transparency, communityengagement, and the integration of data-driven approaches in PR. As with all modern dialogue in AI, ethical c...
2025-08-0747 min
IDEA México PodcastDominando LLMs: Capacitación Certificada IDEA México"¡Bienvenidos a un episodio especial de IDEA México! 🚀 Los Modelos de Lenguaje Grandes (LLMs) como GPT, Llama, y otros, están redefiniendo industrias enteras. Pero, ¿cómo puedes ir más allá de ser un simple usuario y realmente 'Dominar los LLMs' para innovar y liderar?En este episodio, te presentamos nuestra Capacitación Certificada en LLMs de IDEA México, diseñada para profesionales, desarrolladores y entusiastas que buscan una comprensión profunda y habilidades prácticas en esta revolucionaria tecnología.Descubre:✅ ¿Qué son los LLMs y por qué su dominio es crucial en el panor...
2025-06-1315 minThe Edward ShowLLMs.txt - What It Is and Do You Need One?E708: LLMs.txt files are going viral in the SEO world. This is a new file type aimed at helping AI tools understand your site better. But is it worth adding? I explain exactly what LLMs.txt is, how it works, and how it fits into the evolving SEO + AI landscape. I also checked some of the biggest SEO and tech sites to see how they're approaching it - and the results will surprise you. Topics covered: - What LLMs.txt actually does - How AI tools...
2025-06-1209 min
The Growth System16. Cómo posicionarte en ChatGPT y LLMs (SEO para LLM's) con Alejandro GonzálezYa no basta con posicionar un artículo en Google: ahora hay que pensar en cómo aparecer en las respuestas de ChatGPT, Claude, Perplexity y otros modelos de lenguaje.En este episodio me senté con Alejandro González, fundador de Ranki y uno de los mejores practitioners SEO en español, para entender cómo funcionan los LLMs, qué señales toman en cuenta, y qué podemos hacer para aparecer en este canal.Hablamos de experimentos reales, herramientas emergentes como el llms.txt, cómo medir si tu marca está siendo mencionada por los modelos, y...
2025-06-121h 12
Steven Data TalkEP3 Can AI Find Hidden Dangers in Your Code? 🤖💻 LLMs vs. Software Vulnerabilities! (feat. Astrid)(EN)Steven Data Talk - Can AI Find Hidden Dangers in Your Code? 🤖💻 LLMs vs. Software Vulnerabilities! (feat. Astrid)(EN Dubbed)Ever wondered how cutting-edge AI like Large Language Models (LLMs) are revolutionizing software security? In this episode of Steven Data Talk, we chat with Astrid, an expert researcher from University College London (UCL), specializing in using LLMs to detect code vulnerabilities.Join us as we objectively explore:The New Frontier: How LLM-based vulnerability detection differs from traditional machine learning approaches (like those using NLP or Graph Neural Networks).LLM Advantages: Exploring the potential of zero-shot/few-shot learning, flexibility, and genera...
2025-05-0516 min
Data Science Deep Dive#71: Predictive LLMs: Skalierung, Reproduzierbarkeit & DeepSeekIn dieser Folge geht's um die Frage: Macht Größe von Large Language Models (LLMs) bei Predictive Analytics wirklich einen Unterschied? Wir vergleichen Open-Source-Modelle mit bis zu 70 Milliarden Parametern – und siehe da, das 8B-Modell schlägt das große Schwergewicht. Außerdem berichten wir vom Finetuning auf einer AWS-Maschine mit 8 A100-GPUs und den Herausforderungen in Bezug auf die Reproduzierbarkeit. Auch das viel diskutierte DeepSeek-Modell haben wir im Autopreis-Benchmark antreten lassen. Und wie immer fragen wir uns: Was ist praktisch und was ist overkill?
**Zusammenfassung**
Modellgröße ≠ bessere Prognosen: Das Llama-3.1-8B übertraf das größere 7...
2025-05-0126 minData Science Deep Dive#71: Predictive LLMs: Skalierung, Reproduzierbarkeit & DeepSeekIn dieser Folge geht's um die Frage: Macht Größe von Large Language Models (LLMs) bei Predictive Analytics wirklich einen Unterschied? Wir vergleichen Open-Source-Modelle mit bis zu 70 Milliarden Parametern – und siehe da, das 8B-Modell schlägt das große Schwergewicht. Außerdem berichten wir vom Finetuning auf einer AWS-Maschine mit 8 A100-GPUs und den Herausforderungen in Bezug auf die Reproduzierbarkeit. Auch das viel diskutierte DeepSeek-Modell haben wir im Autopreis-Benchmark antreten lassen. Und wie immer fragen wir uns: Was ist praktisch und was ist overkill?
**Zusammenfassung**
Modellgröße ≠ bessere Prognosen: Das Llama-3.1-8B übertraf das größere 7...
2025-05-0126 min
The Ross Simmonds ShowBrands Can Influence LLMsIn this episode of Create Like The Greats, Ross dives into the evolving landscape of digital marketing in the era of artificial intelligence and large language models (LLMs). He challenges brands to rethink their approach by understanding that today’s most influential "audiences" aren't just humans—they're machines. Ross outlines how AI tools like ChatGPT, Perplexity, Google’s AI Overviews, and others scrape public content to generate recommendations and answers for users, and how brands can strategically position their content to become part of this intelligence loop. Whether it’s SEO, Reddit commentary, or owning bottom-of-funnel content, Ross shares powerful...
2025-04-2613 min
Robots TalkingLLMs and Probabilistic Beliefs? Watch Out for Those Answers! EP 33LLMs and Rational Beliefs: Can AI Models Reason Probabilistically?
Large Language Models (LLMs) have shown remarkable capabilities in various tasks, from generating text to aiding in decision-making. As these models become more integrated into our lives, the need for them to represent and reason about uncertainty in a trustworthy and explainable way is paramount. This raises a crucial question: can LLMs truly have rational probabilistic beliefs?
This article delves into the findings of recent research that investigates the ability of current LLMs to adhere to fundamental properties of probabilistic reasoning. Understanding these capabilities and limitations is essential...
2025-04-2114 min
SEOPRESSO PODCAST - Der SEO Podcast mit Björn DarkoRanking in LLMs mit Norman Nielsen (Omio) | Ep. 195In dieser Episode von SEOPRESSO diskutieren Björn Darko und Norman Nielsen die Herausforderungen und Chancen von LLM Rankings im Kontext von SEO. Sie beleuchten die Volatilität der Ergebnisse, die Bedeutung von Sichtbarkeit und Brand Awareness, sowie die Notwendigkeit, Content-Strategien an die neue Antwort-Ökonomie anzupassen. Zudem wird die Rolle von menschlichem Content und die Sichtbarkeit von Daten für LLMs thematisiert. Abschließend werfen sie einen Blick auf die Zukunft der LLMs und deren Monetarisierung.TakeawaysLLMs liefern zunehmend direkte Antworten anstelle klassischer Suchergebnisse.Die Volatilität der LLM-Ergebnisse macht ein statis...
2025-03-2827 min
Experts In Polo ShirtsAre LLMs Living Up to the Hype?Send us Fan MailAre LLMs living up to the hype, or are they failing to deliver?In this episode of Experts in Polo Shirts, we examine the real-world impact of large language models (LLMs) beyond the marketing promises. From Google Gemini’s rocky launch to businesses struggling with AI adoption, we explore the pitfalls, misconceptions, and challenges facing LLMs today.Key Topics:Is Big Tech too big to innovate?Why some AI chatbots are easily manipulated, and ho...
2025-03-201h 11
Devsig PodcastFine-Tuning LLMs: Data Labelling Strategies for GPT-4oThe article "Data Labeling Strategies for Fine-tuning LLMs | Toptal®" discusses data labeling strategies for fine-tuning Large Language Models (LLMs) to enhance their capabilities in specialized industries and tasks. Here's a summary of the key points: Fine-tuning LLMs: Fine-tuning a pre-trained LLM with domain-specific data can extend its capabilities to specialized industries, using smaller datasets than building a model from scratch. The key requirement for fine-tuning is high-quality training data with accurate labeling. Benefits of Fine-tuned LLMs: Fine-tuned LLMs have proven valuable across industries like healthcare, finance, and legal. For example, they are used for transcribing doctor-patient interactions, analyzing market trends, a...
2025-02-2519 min
Devsig PodcastFine-Tuning LLMs: Data Labelling Strategies for GPT-4oThe article "Data Labeling Strategies for Fine-tuning LLMs | Toptal®" discusses data labeling strategies for fine-tuning Large Language Models (LLMs) to enhance their capabilities in specialized industries and tasks. Here's a summary of the key points: Fine-tuning LLMs: Fine-tuning a pre-trained LLM with domain-specific data can extend its capabilities to specialized industries, using smaller datasets than building a model from scratch. The key requirement for fine-tuning is high-quality training data with accurate labeling. Benefits of Fine-tuned LLMs: Fine-tuned LLMs have proven valuable across industries like healthcare, finance, and legal. For example, they are used for transcribing doctor-patient interactions, analyzing m...
2025-02-2519 min
Flying High with FlutterLLMs in Action with Immanuel TrummerIn this episode of Flying High with Flutter, we’re joined by Immanuel Trummer, the author of LLMs in Action and an associate professor at Cornell University specializing in large-scale data analysis. Immanuel shares his insights on large language models (LLMs), how they work, their potential future, and the challenges of privacy and AI. Whether you’re curious about how GPT prompting works or intrigued by the ethics and implications of AI in real-world applications, this episode is packed with valuable knowledge for developers and AI enthusiasts alike!🔥 Exclusive Manning Offer for Podcast Listeners 🔥Get 4...
2025-02-1947 min
mbanerjeepalmer+listennotes 's Listen LaterSimon Willison: Using LLMs for Python Development Podcast: The Real Python Podcast (LS 48 · TOP 1% what is this?)Episode: Simon Willison: Using LLMs for Python DevelopmentPub date: 2025-01-24Get Podcast Transcript →powered by Listen411 - fast audio-to-text and summarizationWhat are the current large language model (LLM) tools you can use to develop Python? What prompting techniques and strategies produce better results? This week on the show, we speak with Simon Willison about his LLM research and his exploration of writing Python code with these rapidly evolving tools.
Simon has been researching LLMs over the...
2025-02-141h 22
The Tech TrekHow AI (LLMs) Are Revolutionizing Data ProductsIn this episode,Sasha Bartashnik shares her insights on howlarge language models (LLMs) are transforming the development ofdata products, making advanced AI-driven solutions moreaccessible and scalable. We dive into thechallenges of traditional data tools, theadvantages and risks of LLM integration, and how businesses shouldadapt to the changing landscape of AI-driven decision-making.Key Takeaways🔹What Are Data Products? – Any software that processes or surfaces data to users, including dashboards and AI-powered insights.🔹Challenges in Building Data Products – Team complexity, data quality, and model training require specialized knowledge and resources.🔹How LLMs Help – They speed up d...
2025-02-1324 min
What's AI Podcast by Louis-François Bouchard7 Reasons Why Learning to Use LLMs Is a Game-ChangerI think the first though about LLMs and generative AI, is often, “Cool tech buzzwords, but do I really need to know this?”
YES. Here’s why diving into LLMs is practically essential...
🚀 1. They transform how we work
Think about all the repetitive, boring tasks in your day. You can (almost) automate them, building tools that make you 10x more productive. That’s what LLMs can do.
If you can't, someone else can. If it's too complex, it will be possible soon.
🧠 2. Reaching their full potential isn’t automatic
LLMs don’t come with a magic "win button," even if ChatGPT by itse...
2025-01-2809 minData Science Deep Dive#64: Predictive LLMs: Übertreffen Open-Source-Modelle jetzt OpenAI und XGBoost bei Preisprognosen?Teil 2 unseres Preisprognose-Experiments für Gebrauchtfahrzeuge: Können Open-Source-LLMs wie Llama 3.1, Mistral und Leo-HessianAI mit GPT-3.5 mithalten? Wir haben fleißig gefinetuned, bis die Motoren qualmten – und es zeigt sich, dass die Unterschiede gar nicht mehr so groß sind. Mit ausreichend vielen Trainingsbeobachtungen nähern sich die Open-Source-Modelle den Ergebnissen von GPT-3.5 an und können es in einzelnen Metriken sogar übertreffen. Für das Finetuning größerer Modelle sind jedoch auch leistungsfähige GPUs notwendig, was die Ressourcenanforderungen deutlich erhöht. In der Folge beleuchten wir, welchen Mehrwert diese Open-Source-LLMs für praxisnahe Use Cases liefern und welche Herausforderung...
2025-01-2340 minData Science Deep Dive#64: Predictive LLMs: Übertreffen Open-Source-Modelle jetzt OpenAI und XGBoost bei Preisprognosen?Teil 2 unseres Preisprognose-Experiments für Gebrauchtfahrzeuge: Können Open-Source-LLMs wie Llama 3.1, Mistral und Leo-HessianAI mit GPT-3.5 mithalten? Wir haben fleißig gefinetuned, bis die Motoren qualmten – und es zeigt sich, dass die Unterschiede gar nicht mehr so groß sind. Mit ausreichend vielen Trainingsbeobachtungen nähern sich die Open-Source-Modelle den Ergebnissen von GPT-3.5 an und können es in einzelnen Metriken sogar übertreffen. Für das Finetuning größerer Modelle sind jedoch auch leistungsfähige GPUs notwendig, was die Ressourcenanforderungen deutlich erhöht. In der Folge beleuchten wir, welchen Mehrwert diese Open-Source-LLMs für praxisnahe Use Cases liefern und welche Herausforderung...
2025-01-2340 min
Knowledge Graph InsightsFran Alexander: Alien vs Predator and LLMs vs Knowledge Graphs – Episode 15Fran Alexander
When Fran Alexander looks at the current AI landscape she sees some interesting parallels between the Alien vs Predator science fiction franchise and the way RAG and other architectures are combining LLMs and knowledge graphs.
We talked about:
the analogy she draws between the Alien and Predator science fiction franchise with LLMs and knowledge graphs
how the human-esque (if malevolent) cognitive and behavioral nature of Predators aligns more with knowledge graphs and how the unpredictable and stochastic nature of Aliens aligns more with LLMs
how the eloquence of LLM outputs can deceive humans
the lack of explainability and...
2024-12-0734 min
Daily Paper Cast (Test)What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective🤗 Daily Paper Upvotes: 42
Authors: Ming Li, Yanhong Li, Tianyi Zhou
Categories: cs.CL, cs.AI, cs.LG
Arxiv: http://arxiv.org/abs/2410.23743v1
Title: What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective
Abstract: What makes a difference in the post-training of LLMs? We investigate the training patterns of different layers in large language models (LLMs), through the lens of gradient, when training with different responses and initial models. We are specifically interested in how fast vs. slow thinking affects the layer-wise gradients, given the recent popularity of training LLMs on reasoning paths such as...
2024-11-0303 minDaily Paper Cast (Test)What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective🤗 Daily Paper Upvotes: 42
Authors: Ming Li, Yanhong Li, Tianyi Zhou
Categories: cs.CL, cs.AI, cs.LG
Arxiv: http://arxiv.org/abs/2410.23743v1
Title: What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective
Abstract: What makes a difference in the post-training of LLMs? We investigate the training patterns of different layers in large language models (LLMs), through the lens of gradient, when training with different responses and initial models. We are specifically interested in how fast vs. slow thinking affects the layer-wise gradients, given the recent popularity of training LLMs on reasoning paths such as...
2024-11-0303 min
Psych Tech @ WorkLLMs, Talent Assessments, & Hiring- Research Meets Practice"We’re generating assessments faster than ever, but our real test is ensuring that these tools are fair and reliable across diverse candidate groups."–Louis HickmanIn this episode I welcome my friend, super dad, and ex- professional wrestler Louis Hickman for a killer conversation about the ins and outs of using LLMs to create and score assessments.Louis is a professor at Virginia Tech specializing in research on AI and large language models in assessment and hiring processes. He knows a thing or two about this stuff and we waste no time tackl...
2024-10-311h 02
Swetlana AI PodcastLLMs & Ideology Of Their CreatorsIn this episode we're discussing this paper:"Large Language Models Reflect the Ideology of their Creators"byMaarten Buyl, Alexander Rogiers, Sander Noels, Iris Dominguez-Catena, Edith Heiter, Raphael Romero, Iman Johary, Alexandru-Cristian Mara, Jefrey Lijffijt, Tijl De Biehttps://arxiv.org/pdf/2410.18417We will delve into a groundbreaking study exploring the ideological leanings of large language models (LLMs). Through analyzing LLMs' responses to prompts about diverse historical figures, researchers found that these models often mirror the perspectives of their creators, exhibiting biases influenced by...
2024-10-2811 min
Crazy WisdomEpisode #403: Unlocking AI’s Brain: Knowledge Graphs, LLMs, and the Future of ReasoningIn this episode of the Crazy Wisdom Podcast, host Stewart Alsop welcomes Chia Yang, co-founder of whyhow.ai, a company specializing in data infrastructure and AI-powered knowledge graphs. They discuss the pivotal role of knowledge graphs in AI, particularly in enhancing structured search and reasoning, contrasting them with more stochastic systems like large language models (LLMs). Chia explains how knowledge graphs allow for more structured, reliable connections between data, and how this impacts the development of production-grade AI systems. He also touches on the limitations of LLMs, the significance of neurosymbolic approaches, and the future of AI reasoning. For...
2024-10-2540 min
LA AZOTEA Podcast¿Pueden los LLMS razonar?Los modelos de lenguaje extenso (LLMs) han demostrado capacidades notables en diversas áreas, incluyendo el procesamiento del lenguaje natural y la respuesta a preguntas. Esto ha llevado a un debate sobre si los LLMs han alcanzado capacidades de razonamiento similares a las humanas o si estas habilidades son una ilusión. El razonamiento, una parte central de la inteligencia humana, implica habilidades como la deducción, la inducción, la abducción y el pensamiento analógico.
Algunos investigadores argumentan que los LLMs exhiben un comportamiento similar al razonamiento, especialmente cuando se utilizan técnicas como la "Cadena...
2024-10-2315 min
AIandBlockchainCan AI Really Think? A Deep Dive into LLMs and Mathematical ReasoningIn this episode, we explore the fascinating world of large language models (LLMs)—those AI tools that can write poems, code websites, and seemingly perform all sorts of impressive tasks. But how smart are these models, really? Do they actually understand what they're doing, or are they just really good at making us think they understand? This million-dollar question sets the stage for today’s discussion.
We focus on a cutting-edge research paper, GSM Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models, which uses math as a lens to test the reasoning abilities of these A...
2024-10-1410 min
Mad Tech Talk#28 - Breaking the Echo Chamber: LLMs and the Future of Online SearchIn this episode of Mad Tech Talk, we delve into the intriguing research on how large language models (LLMs) can create "echo chambers" in online search, potentially reinforcing users' pre-existing beliefs and exacerbating opinion polarization. Based on two comprehensive studies, we explore the dynamics of information-seeking behaviors and the impact of opinionated LLMs on user perspectives.
Key topics covered in this episode include:
Impact on Information Diversity: Discuss how LLM-powered conversational search systems influence information diversity and opinion polarization. Understand the comparison between conventional web search and conversational search using LLMs on controversial to...
2024-10-0813 min
ReadOrListenPHP and LLMs — Web Scraping and Building an Events Database using ToolsThis is out of the book "PHP and LLMs". https://bit.ly/php_llmsScraping the web for data can feel like wrestling with a tangled mess of HTML. But what if you could skip the wrestling match and have a powerful AI assistant do the heavy lifting for you?On today's episode, we're exploring how Large Language Models are transforming the web scraping game. We'll unpack how LLMs, combined with cleverly designed tools, can intelligently extract and structure data from websites, all without you needing to become an HTML parsing expert. **...
2024-10-0609 minReadOrListenWhy LLMs - Chapter 2 of PHP and LLMs The BookThis is the start of 10 part series as the "PHP and LLMs" book uses NotebookLm to create discussions around each chapter.These episodes are created using https://notebooklm.google.com/🎉 Buy the book now it is 40% done https://bit.ly/php_llms📰 Join the news letter https://sundance-solutions.mailcoach.app/php-and-llmsWhy LLMs - Chapter 2Eight years ago, I did a Machine Learning and PHP video on YouTube, and it is still my most popular video. Back then, AWS made a service that started to make "Machine Learning" easy to host and create APIs a...
2024-09-2408 minReadOrListenPrompting Overview - Chapter 5 of PHP and LLMs Take 2This is the start of 10 part series as the "PHP and LLMs" book uses NotebookLm to create discussions around each chapter.These episodes are created using https://notebooklm.google.com/🎉 Buy the book now it is 40% done https://bit.ly/php_llms📰 Join the news letter https://sundance-solutions.mailcoach.app/php-and-llmsPrompting Overview - Chapter 5Prompting is the key to unlocking the power of Large Language Models (LLMs) in your PHP applications. It's the language we use to communicate with AI, enabling our code to leverage advanced natural language processing for tasks ranging from c...
2024-09-2110 minReadOrListenPrompting Overview - Chapter 5 of PHP and LLMsThis is the start of 10 part series as the "PHP and LLMs" book uses NotebookLm to create discussions around each chapter.These episodes are created using https://notebooklm.google.com/🎉 Buy the book now it is 40% done https://bit.ly/php_llms📰 Join the news letter https://sundance-solutions.mailcoach.app/php-and-llmsPrompting Overview - Chapter 5Prompting is the key to unlocking the power of Large Language Models (LLMs) in your PHP applications. It's the language we use to communicate with AI, enabling our code to leverage advanced natural language processing for tasks ranging from c...
2024-09-2110 min
mbanerjeepalmer+listennotes 's Listen LaterLLMs are like your weird, over-confident intern | Simon Willison (Datasette) Podcast: Software Misadventures (LS 29 · TOP 10% what is this?)Episode: LLMs are like your weird, over-confident intern | Simon Willison (Datasette)Pub date: 2024-09-10Get Podcast Transcript →powered by Listen411 - fast audio-to-text and summarizationKnown for co-creating Django and Datasette, as well as his thoughtful writing on LLMs, Simon Willison joins the show to chat about blogging as an accountability mechanism, how to build intuition with LLMs, building a startup with his partner on their honeymoon, and more. Segments: (00:00:00) The weird intern (00:01:50) The...
2024-09-161h 55
Notes to My Legal Self®Season 9, Episode 2: LLMs and Business of Law: How Will Lawyers Continue Making Money?! (with Damien Riehl)How will lawyers make money in a world where AI can do in minutes what used to take hours? In this episode of Notes to My (Legal) Self, we dive into the transformative effects of Large Language Models (LLMs) on legal practice with our special guest, Damien Riehl. Together, we explore the profound changes LLMs are bringing to the traditional hourly billing model and discuss how the legal industry can adapt to this new reality.
With LLMs increasing efficiency, we're seeing a shift toward flat fee billing models. Currently, flat fees make up about 10...
2024-09-1243 min
Software MisadventuresLLMs are like your weird, over-confident intern | Simon Willison (Datasette)Known for co-creating Django and Datasette, as well as his thoughtful writing on LLMs, Simon Willison joins the show to chat about blogging as an accountability mechanism, how to build intuition with LLMs, building a startup with his partner on their honeymoon, and more. Segments: (00:00:00) The weird intern (00:01:50) The early days of LLMs (00:04:59) Blogging as an accountability mechanism (00:09:24) The low-pressure approach to blogging (00:11:47) GitHub issues as a system of records (00:16:15) Temporal documentation and design docs (00:18:19) GitHub issues for team collaboration
2024-09-101h 55
Machine Learning Street Talk (MLST)Prof. Subbarao Kambhampati - LLMs don't reason, they memorize (ICML2024 2/13)Prof. Subbarao Kambhampati argues that while LLMs are impressive and useful tools, especially for creative tasks, they have fundamental limitations in logical reasoning and cannot provide guarantees about the correctness of their outputs. He advocates for hybrid approaches that combine LLMs with external verification systems.
MLST is sponsored by Brave:
The Brave Search API covers over 20 billion webpages, built from scratch without Big Tech biases or the recent extortionate price hikes on search API access. Perfect for AI model training and retrieval augmentated generation. Try it now - get 2,000 free queries monthly...
2024-07-291h 42
The Holistic Success ShowHow to improve the curriculum review process using large language models (LLMs)In this episode, Dr. Deborah Chang, Director of Curriculum Evaluation at UT Health San Antonio discusses how UT Health's School of Medicine is using large language models (LLMs) to improve their curriculum review process. Dr. Chang shares how AI has helped synthesize large amounts of data, providing detailed results and summaries to faculty members and committee members — in turn enhancing continuous quality improvement processes. Timestamp of our conversation: 01:44 - What motivated the use of LLMs? 03:08 - What were the challenges prior to using LLMs? 05:08 - Were there any reservations about using LLMs? 06:57 - Wha...
2024-07-1718 min
Experiencing Data w/ Brian T. O’Neill147 - LLMs need UX: How to Increase Your B2B Product’s Value with AI (Part 1)Let’s talk about design for AI (which more and more, I’m agreeing means GenAI to those outside the data space). The hype around GenAI and LLMs—particularly as it relates to dropping these in as features into a software application or product—seems to me, at this time, to largely be driven by FOMO rather than real value. In this “part 1” episode, I look at the importance of solid user experience design and outcome-oriented thinking when deploying LLMs into enterprise products. Challenges with immature AI UIs, the role of context, the constant game of understanding what accuracy means (and h...
2024-07-1025 min
Two Voice DevsEpisode 197 - Alexa Skill Development in the Age of LLMsWhat should people developing with LLMs learn from a decade of experience building Alexa skills? How will Alexa skill developers leverage the latest #GenerativeAI and #CoversationalAI tools as they continue to build #VoiceFirst and multimodal skills?
Join Allen and Mark on Two Voice Devs as they delve into the evolving landscape of Alexa skill development in the era of large language models (LLMs). Sparked by a thought-provoking discussion on the Alexa forums, they explore the potential benefits and challenges of integrating LLMs into skills.
Key topics and timestamps:
2024-07-0540 min
Vanishing GradientsEpisode 30: Lessons from a Year of Building with LLMs (Part 2)Hugo speaks about Lessons Learned from a Year of Building with LLMs with Eugene Yan from Amazon, Bryan Bischof from Hex, Charles Frye from Modal, Hamel Husain from Parlance Labs, and Shreya Shankar from UC Berkeley.These five guests, along with Jason Liu who couldn't join us, have spent the past year building real-world applications with Large Language Models (LLMs). They've distilled their experiences into a report of 42 lessons across operational, strategic, and tactical dimensions (https://applied-llms.org/), and they're here to share their insights.We’ve split this roundtable into 2 episodes and, in this second episode, we'll ex...
2024-06-261h 15
Vanishing GradientsEpisode 29: Lessons from a Year of Building with LLMs (Part 1)Hugo speaks about Lessons Learned from a Year of Building with LLMs with Eugene Yan from Amazon, Bryan Bischof from Hex, Charles Frye from Modal, Hamel Husain from Parlance Labs, and Shreya Shankar from UC Berkeley.These five guests, along with Jason Liu who couldn't join us, have spent the past year building real-world applications with Large Language Models (LLMs). They've distilled their experiences into a report of 42 lessons across operational, strategic, and tactical dimensions (https://applied-llms.org/), and they're here to share their insights.We’ve split this roundtable into 2 episodes and, in this first episode, we'll ex...
2024-06-261h 30
AI and IWhat Do LLMs Tell Us About the Nature of Language—And Ourselves? - Ep. 23 with Robin SloanAn interview with best-selling sci-fi novelist Robin SloanOne of my favorite fiction writers, New York Times best-selling author Robin Sloan, just wrote the first novel I’ve seen that’s inspired by LLMs.
The book is called Moonbound, and Robin originally wanted to write it with language models. He tried doing this in 2016 with a rudimentary model he built himself, and more recently with commercially available LLMs. Both times Robin found himself unsatisfied with the creative output generated by the models. AI couldn’t quite generate the fiction he was looking for—the kind that pushes the b...
2024-06-1253 min
AI & IWhat Do LLMs Tell Us About the Nature of Language—And Ourselves? - Ep. 23 with Robin SloanAn interview with best-selling sci-fi novelist Robin SloanOne of my favorite fiction writers, New York Times best-selling author Robin Sloan, just wrote the first novel I’ve seen that’s inspired by LLMs.The book is called Moonbound, and Robin originally wanted to write it with language models. He tried doing this in 2016 with a rudimentary model he built himself, and more recently with commercially available LLMs. Both times Robin found himself unsatisfied with the creative output generated by the models. AI couldn’t quite generate the fiction he was looking for—th...
2024-06-1253 min
Vanishing GradientsEpisode 28: Beyond Supervised Learning: The Rise of In-Context Learning with LLMsHugo speaks with Alan Nichol, co-founder and CTO of Rasa, where they build software to enable developers to create enterprise-grade conversational AI and chatbot systems across industries like telcos, healthcare, fintech, and government.What's super cool is that Alan and the Rasa team have been doing this type of thing for over a decade, giving them a wealth of wisdom on how to effectively incorporate LLMs into chatbots - and how not to. For example, if you want a chatbot that takes specific and important actions like transferring money, do you want to fully entrust the conversation to one...
2024-06-101h 05
New Paradigm: AI Research SummariesA Summary of 'LLMs achieve adult human performance on higher-order theory of mind tasks' by Google DeepMind, Johns Hopkins University & TheA Summary of Google DeepMind, Johns Hopkins University & The University of Oxford's 'LLMs achieve adult human performance on higher-order theory of mind tasks' Available at: https://arxiv.org/abs/2405.18870 This summary is AI generated, however the creators of the AI that produces this summary have made every effort to ensure that it is of high quality. As AI systems can be prone to hallucinations we always recommend readers seek out and read the original source material. Our intention is to help listeners save time and stay on top of trends and new discoveries. You can find the introductory section of...
2024-06-0611 min
Rabbit FoodLLMs need searchSummary
LLMs and vector databases are powerful tools in information retrieval, but they still need a search engine to perform optimally. Vectors provide predictions based on the most likely context within the vector space, but without additional context, the interpretation can be difficult. LLMs understand language patterns and allow for semantic search without exact terms. Vector databases use coordinates to find content matches and determine relevance, but they lack the user's context. Elasticsearch as a vector database allows for additional context and combines multiple search modalities for better results.
Keywords:
...
2024-06-0506 min
The UnionProtecting Your Company Data When Using LLMsWhile LLMs offer undeniable benefits, integrating them into the workplace poses significant risks to company data. Here’s why:Data Leakage: It’s easy for employees to paste confidential company information into LLM prompts inadvertently. This could include anything an employee can access: financial reports, trade secrets, customer data in text, documents, or even data in spreadsheets. Ownership Concerns: When company data is used to create content using LLMs, there’s a risk of losing ownership rights or control over intellectual property. Who owns the content created by LLMs? The company that provides the data or the...
2024-05-2218 min
New Paradigm: AI Research SummariesA Summary of Predibase's 'LoRA Land: 310 Fine-tuned LLMs that Rival GPT-4, A Technical Report'A Summary of Predibase's 'LoRA Land: 310 Fine-tuned LLMs that Rival GPT-4, A Technical Report' Available at: https://arxiv.org/abs/2405.00732 This summary is AI generated, however the creators of the AI that produces this summary have made every effort to ensure that it is of high quality. As AI systems can be prone to hallucinations we always recommend readers seek out and read the original source material. Our intention is to help listeners save time and stay on top of trends and new discoveries. You can find the introductory section of this recording provided below... This is a summary of "...
2024-05-1115 min
Stories from the HackeryUsing LLMs Beyond the Chatbot | Stories From The HackeryAs we continue to discuss generative AI on Nashville Software School’s podcast, Stories from the Hackery, Founder and CEO John Wark and lead Data instructor Michael Holloway, dive into various techniques for leveraging large language models (LLMs) like generative AI. They explore the potential of using hosted public LLMs via chatbot interfaces and discuss strategies for embedding LLMs into applications. One such technique discussed is the use of a prompt engineering, which involves wrapping the LLM API to tailor user prompts for more effective responses.
They also discuss more advanced techniques like retrieval-augmented generation (RAG), which involves using external da...
2024-05-0856 min
New Paradigm: AI Research SummariesA Summary of Tencent AI Lab's 'Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing'A Summary of Tencent AI Lab's 'Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing' Available at: https://arxiv.org/abs/2404.12253 This summary is AI generated, however the creators of the AI that produces this summary have made every effort to ensure that it is of high quality. As AI systems can be prone to hallucinations we always recommend readers seek out and read the original source material. Our intention is to help listeners save time and stay on top of trends and new discoveries. You can find the introductory section of this recording provided below... This is a summary...
2024-04-2210 minNew Paradigm: AI Research SummariesA Summary of Microsoft Research's 'The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits'This is a summary of the AI research paper: The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits Available at: https://arxiv.org/abs/2402.17764 And is also available here: https://huggingface.co/papers/2402.17764 This summary is AI generated, however the creators of the AI that produces this summary have made every effort to ensure that it is of high quality. As AI systems can be prone to hallucinations we always recommend readers seek out and read the original source material. Our intention is to help listeners save time and stay on top of...
2024-04-2007 minChatGPT: News on Open AI, MidJourney, NVIDIA, Anthropic, Open Source LLMs, Machine LearningUnstructured's $25M Data Prep Journey with CEO Brian RaymondIn this episode, we dive into Unstructured's impressive journey, having raised $25 million to enhance data preparation for LLMs (Large Language Models). Join us for an enlightening conversation with Unstructured's CEO, Brian Raymond, as we explore the significance of this funding, the role it plays in refining data for LLMs, and the exciting developments that lie ahead. Discover how Unstructured is shaping the landscape of data optimization for cutting-edge language models.
Get on the AI Box Waitlist: https://AIBox.ai/Join our ChatGPT Community: https://www.facebook.com/groups/739308654562189/Follow me on Twitter: https://twitter.com/jaeden...
2024-04-1133 min
LessWrong (Curated & Popular)LLMs for Alignment Research: a safety priority?A recent short story by Gabriel Mukobi illustrates a near-term scenario where things go bad because new developments in LLMs allow LLMs to accelerate capabilities research without a correspondingly large acceleration in safety research.This scenario is disturbingly close to the situation we already find ourselves in. Asking the best LLMs for help with programming vs technical alignment research feels very different (at least to me). LLMs might generate junk code, but you can keep pointing out the problems with the code, and the code will eventually work. This can be faster than doing it myself, in cases...
2024-04-0720 min
The Georgian Impact Podcast | AI, ML & MoreTesting LLMs for trust and safetyWe all get a few chuckles when autocorrect gets something wrong, but there's a lot of time-saving and face-saving value with autocorrect. But do we trust autocorrect? Yeah. We do, even with its errors. Maybe you can use ChatGPT to improve your productivity. Ask it to a cool question and maybe get a decent answer. That's fine. After all, it's just between you and ChatGPT. But, what if you're a software company and you're leveraging these technologies? You could be putting generative AI output in front of your users.
On this episode of the Georgian Impact Podcast...
2024-03-1521 minChatGPT: News on Open AI, MidJourney, NVIDIA, Anthropic, Open Source LLMs, Machine LearningDynamoFL Secures $15.1M Funding to Shield Enterprises from Data Leaks to LLMsIn this episode, we dive into the exciting world of DynamoFL, the data protection solution designed to keep enterprises from leaking sensitive data to large language models (LLMs). Join us as we explore how DynamoFL raised an impressive $15.1 million in funding to further enhance its mission of safeguarding sensitive information. Discover the innovative technologies and strategies that are reshaping data security in the age of advanced AI.
Get on the AI Box Waitlist: https://AIBox.ai/Join our ChatGPT Community: https://www.facebook.com/groups/739308654562189/Follow me on Twitter: https://twitter.com/jaeden_ai
See Pri...
2024-02-2010 min
Real AI. Now.Inside LLMs: A Deep Dive (with Paulo Nunes, Marc Giombetti & Pascal Guldener)This is a very special episode! We’re going deep on the hot-topic of Large Language Models for this one and decided to make it even better by having not only an LLMs expert, but also both Real AI. Now. hosts in the conversation. Do we have your curiosity?
Our special guest is Pascal Guldener, NLP Engineer at Two Impulse. He has been working for several years now in the area of software engineering with application to NLP. He not only has actively worked on NLP projects, from recommender systems to libraries, but also took the front se...
2024-02-191h 18
ILTA Voices#0003: (CCT) Incorporating Large Language Models (LLMs) in the Legal ArenaIn this podcast interview, the speaker will share experiences and challenges in utilizing Large Language Models in the legal arena.
Questions the moderator asked the speaker:
General
-What are some of the key challenges that legal departments face in adopting and integrating LLMs into their workflows?
-How can legal departments overcome the initial resistance or skepticism that may exist towards using LLMs?
-How can legal departments ensure that LLMs are being used responsibly and ethically, without perpetuating biases or producing inaccurate or misleading results?
People
-How can legal departments train and empower their lawyers...
2024-02-0919 min
The Gradient: Perspectives on AISubbarao Kambhampati: Planning, Reasoning, and Interpretability in the Age of LLMsIn episode 110 of The Gradient Podcast, Daniel Bashir speaks to Professor Subbarao Kambhampati.Professor Kambhampati is a professor of computer science at Arizona State University. He studies fundamental problems in planning and decision making, motivated by the challenges of human-aware AI systems. He is a fellow of the Association for the Advancement of Artificial Intelligence, American Association for the Advancement of Science, and Association for Computing machinery, and was an NSF Young Investigator. He was the president of the Association for the Advancement of Artificial Intelligence, trustee of the International Joint Conference on Artificial Intelligence, and a...
2024-02-081h 59
The AI and Digital Transformation PodcastLLMs are not magic: Finding ways to make AI generate trustworthy content | Tim LeersCan we rely on LLMs to repurpose our content in social media?
To end our first season of the AI and Digital Transformation Podcast, we talked to dataroots R&D engineer Tim Leers about two very popular topics in 2023: LLMs, and content creation.
In this age of content creation and social media, journalists now have an extra role to fill: sharing their work and the news using their social media accounts. Given the popular use of ChatGPT and Midjourney, people ask LLMs to repurpose their news content for social media purposes.
This comes...
2024-01-261h 01
NVIDIA AI PodcastNVIDIA’s Annamalai Chockalingam on the Rise of LLMs - Ep. 206Generative AI and large language models (LLMs) are stirring change across industries — but according to NVIDIA Senior Product Manager of Developer Marketing Annamalai Chockalingam, “we’re still in the early innings.”
In the latest episode of NVIDIA’s AI Podcast, host Noah Kravitz spoke with Chockalingam about LLMs: what they are, their current state and their future potential.
LLMs are a “subset of the larger generative AI movement” that deals with language. They’re deep learning algorithms that can recognize, summarize, translate, predict and generate language.
AI has been around for a while, but according to Chockalingam, three key factors enabled LLMs.
O...
2023-11-2338 min
AI-Powered Bot : ChatgptNVIDIA’s Annamalai Chockalingam on the Rise of LLMs - Ep. 206Generative AI and large language models (LLMs) are stirring change across industries — but according to NVIDIA Senior Product Manager of Developer Marketing Annamalai Chockalingam, “we’re still in the early innings.” In the latest episode of NVIDIA’s AI Podcast, host Noah Kravitz spoke with Chockalingam about LLMs: what they are, their current state and their future potential. LLMs are a “subset of the larger generative AI movement” that deals with language. They’re deep learning algorithms that can recognize, summarize, translate, predict and generate language. AI has been around for a while, but according to Chockalingam, three key factors enabled LLMs...
2023-11-0138 min
AI Unraveled: Latest AI News, ChatGPT, Gemini, Claude, DeepSeek, Gen AI, LLMs, Agents, Ethics, BiasAI Monthly Rundown September 2023: The Future of LLMs in Search!AI Monthly Rundown September 2023: The Future of LLMs in Search!https://youtu.be/9hmWPza7dQEExplore the latest developments in the AI world for September 2023. We delve into the burning question: Are Large Language Models (LLMs) poised to replace traditional search engines? Dive into this comprehensive rundown and discover the evolution and future of search in the age of AI.In today's episode, we'll cover the evolution of search and large language models, Amazon's investment in Anthropic and generative AI updates, Google's advancements in personalized route suggestions and language modeling, DeepMind's AlphaMissense system...
2023-09-3054 min
Bricks And Bytes#057 - Rohan Jawali - How LLMs And AI Will Shape the Next Decade Of Construction TechIn today’s episode of Bricks & Bytes, we have Rohan Jawali, the Co-Founder of Joist.ai
In this episode, we learn about the key lessons Rohan learned while building an AI-based startup, the significance and capabilities of LLMs in construction, how to write better prompts, and much more!
Tune in to find out about:
How LLM’s will change construction
How to write better prompts
Limits and the future of LLMs
The key areas that LLM’s disrupt
If you enjoy today’s episode, leave us a comment. And don’t forget to subscribe...
2023-09-2246 min
VUX WorldTalking LLMs with Voiceflow’s first Senior Conversation Design Advocate - Peter Isaacs!Peter has written and shared a lot about Large Language Models in recent posts and blogs. We wanted to get his perspective on various things, especially now that Voiceflow have incorporated LLMs into their tool. How can conversation designers use LLMs now? How does Peter use the LLM as a sounding board to help him craft prompts? Where might we be using them next? How can we document dynamic conversational designs (when LLMs can allow conversations to go off-rails)?Ben asked him these questions and more…0:00:00Start0:02:44About Pet...
2023-07-201h 08
Redefining CyberSecuritySafeguarding Against Malicious Use of Large Language Models: A Review of the OWASP Top 10 for LLMs | A Conversation with Jason Haddix | Redefining CyberSecurity with Sean MartinGuest: Jason Haddix, CISO and Hacker in Charge at BuddoBot Inc [@BuddoBot]On LinkedIn | https://www.linkedin.com/in/jhaddix/On Twitter | https://twitter.com/Jhaddix____________________________Host: Sean Martin, Co-Founder at ITSPmagazine [@ITSPmagazine] and Host of Redefining CyberSecurity Podcast [@RedefiningCyber]On ITSPmagazine | https://www.itspmagazine.com/itspmagazine-podcast-radio-hosts/sean-martin____________________________This Episode’s SponsorsImperva | https://itspm.ag/imperva277117988Pentera | https://itspm.ag/penteri67a___________________________Episode NotesIn this Redefining CyberSecurity Podcast, we provide an in-depth exploration of the...
2023-06-1451 min
On The Frontier#12 - Phillip Carter (Principal PM @ Honeycomb) - All The Hard Stuff when Building Products with LLMs, Actual Results from Leveraging AIPhillip Carter is a Principal Product Manager at Honeycomb, which develops a software debugging product for distributed systems. Phillip recently published one of the most interesting blog posts I’ve read titled “All the Hard Stuff Nobody Talks About when Building Products with LLMs”. The post is excellent and everyone should give it a read. In this episode, we dive deep into what the hard stuff actually is, the pros and cons of Large Language Models (LLMs) and what teams need to think about when using LLMs in their products. We also talk about the real world results that come f...
2023-06-1450 min
GeeksBlaBla#149 - Building Smart Apps with LLMsIn this episode, we discuss LLMs, how everything started, how they work, and how to use frameworks such as LangChain to develop intelligent applications with them.
Guests
Taha Bouhsine
Sifeddine Nahhas
Nouamane Tazi
Notes
0:00:00 - Introduction and welcoming
0:05:33 - History of LLMs
0:12:00 - The role of transformers in LLMs
0:21:00 - How LLMs differ from other AI methods
0:26:00 - Emergent Abilities of Large Language Models
0:42:00 - HaggingFace and the role or open-source in LLMs
0:47...
2023-06-081h 51