show episodes
 
Artwork
 
Join host, Diana Varma, as she explores the wonderful world of design and printing, typography and branding, books and publishing. In each episode Diana aims to answer many age-old, deep philosophical questions such as: Where the heck did Comic Sans originate? Why do the majority of printed books have blank pages at the end? What is the world's ugliest colour? What does a sassy three-year-old have to say about the visible colour spectrum? Ready to have some fun? Let's talk paper scissors.
  continue reading
 
Focusing on independent typography, episodes of the Interrogang highlight new and exciting developments in type design and include lively discussions, games, and interviews in the world of type, design, and artistic expression. Co-hosted by Joshua Dick and Kyle Read.
  continue reading
 
Artwork

1
Typography Dojo

Rachel Elnar @ TypeEd

Unsubscribe
Unsubscribe
Ogni mese
 
Typography Dojo is the place where we explore the traditions and techniques in the practice of type and typography with lessons and insights from visual design masters. Join me to learn more about typesetting, readability, hand lettering, editorial design, type design, Calligraphy, logo design and how they improve graphic communication.
  continue reading
 
Artwork

1
Station 13

Dani Wright & Alex May

Unsubscribe
Unsubscribe
Ogni mese
 
Dani Wright and Alex May used to make games together in Kyoto. Now they live continents apart. Join them as they meet once more at Station 13.
  continue reading
 
Artwork

1
The Datacenter Show

Barb Darrow and Derrick Harris

Unsubscribe
Unsubscribe
Ogni mese
 
A fun podcast devoted to celebrating/commiserating our ridiculously overabundant world. Each episode features a guest expert to wax philosophic on how there's too much TV, movies, typography, politics, time, choices -- whatever!
  continue reading
 
Artwork
 
Wellfed is hosted by Jon Sorrentino, a graphic designer based in New York with a passion for creativity and an obsession with typography. Jon speaks to graphic designers, photographers, illustrators, and some of his creative heroes to discover the ingredients for a successful creative career. Learn more at wellfedpodcast.com or follow along on Instagram @wellfed.podcast & @jonsorre
  continue reading
 
Artwork

1
Overtime

Dribbble

Unsubscribe
Unsubscribe
Ogni settimana
 
Overtime, Dribbble's weekly podcast, explores the most interesting design news and gives you the tips you need to create your best work. No need to scroll copious news sites and design Twitter—host Meg Lewis and friends are primed to inform, entertain, and inspire you each week. Consider Overtime your audio lifeline to the design community, packaged in a perfectly punchy Shot. Subscribe and listen every Wednesday to stay ahead of the game.
  continue reading
 
Artwork
 
Welcome to Clokendagger's official podcast, Beneath the Brand, hosted by Joshua Doty & Olivia Viterna. Join us as we discuss how design and content impact everyday life. We'll break down what makes good design vs. bad design, how to master typography, and why consistent brand messaging creates better engagement.
  continue reading
 
Artwork
 
Design influences the world around us every day. We chat with designers, illustrators, photographers, and others to find out the stories that made them want to get into the creative world as a career, and talk with chefs, bartenders, restaurateurs and anyone else that uses their creative talent about how they use their creative talents to design their menus and restaurants. If you have suggestions for guests or questions about the show, please reach out to us on Twitter @feastondesign.
  continue reading
 
Artwork
 
Welcome the TypeCraft, the podcast that reveals the untold stories of type design! Season One of TypeCraft focuses on the world of Custom Type, the branding and lettering that drives so much business and marketing, but that we take so much for granted. We interviewed eight industry leaders over six episodes about their experiences, backgrounds, and perspectives creating custom typography and innovative brands. Credits: Interviews conducted by Kyle Read Audio edited by Andrew Sfiris Audio ass ...
  continue reading
 
Obsessed Show host Josh Miles interviews some of the hottest designers—from branding and illustration, to industrial design and architecture, to learn what makes them tick.
  continue reading
 
Explore the untold histories behind the images, patterns, and designs we take for granted. Each week we’ll bring you a different story about a curiosity from the creative world - interviewing experts and creatives as we dive into the unusual origins of everything from Comic Sans to the polka dot. Curious Minds is an original podcast by Domestika dedicated to sharing and celebrating creativity in all its forms. To find blogs and images for each episode, go to domestika.org/podcasts. And to fi ...
  continue reading
 
Artwork

1
GUNWASH

Heritage Radio Network

Unsubscribe
Unsubscribe
Ogni mese
 
HEYWHATSUP™? GUNWASH is the future of talk radio; a weekly 90 minute explosion of dark humor, eye-opening interviews, dancehall reggae and psychedelic sounds. Tune in for unguarded insight from the artists, musicians, troublemakers and curators shaping culture in the 21st century. Design, typography, astrophysics, dancehall reggae, conspiracy theories, illustration, drums, nuclear engineering, disco, Polo Ralph Lauren, crime, the 1990's, iconography, iced coffee, late night downtown, drummer ...
  continue reading
 
The Neon Moiré Show is the podcast of neonmoire.com, an online curated design conference calendar of the world's most interesting design conferences and events on our digitized world. The agenda will focus on graphic, typography and interaction design conferences, not to forget design film festivals and an overview of the top design summer schools. The show is hosted by Thomas Dahm.
  continue reading
 
Artwork

1
The Weekly Typographic

The League of Moveable Type

Unsubscribe
Unsubscribe
Ogni mese
 
Welcome to the Weekly Typographic — a fun podcast talking about this week's most amazing articles in the world of typography, keeping you up-to-date on the wonderful world of type. Each week we'll discuss the cool news we've found, what we're working on, and answer burning questions about design & fontmaking.
  continue reading
 
Artwork

1
Between | Screens Podcast

Between | Screens Podcast

Unsubscribe
Unsubscribe
Ogni mese
 
Code | Design | Micro podcast. Expect to find content about the following topics: Ruby, Rails, Sinatra, Sass, Javascript, Design Patterns, Jekyll, Client-Side Web Apps, Rust, Typography, Git, Responsive Design, Prototyping, Web Design, Type Design and more … The elevator pitch for this micro-podcast is to conduct extensive interviews with professional designers & developers and distill multiple small digestible knowledge bombs out of these conversations. The podcast is free of charge, leaves ...
  continue reading
 
Artwork

1
How You Create

How You Create

Unsubscribe
Unsubscribe
Ogni mese
 
Ben Terry and Joshua Jean-Marie host candid conversations with creators, designers, storytellers, and more about what it takes to survive and thrive as a creative in this modern world. We pull back the curtain and explore the guest's creative journey and the practical lessons they've learned along the way. You will hear stories from creatives who have created some incredible work about what’s it’s like to succeed, what’s it’s like to fail, and what motivates and inspires them to keep moving ...
  continue reading
 
Artwork

1
Arxiv Papers

Igor Melnyk

Unsubscribe
Unsubscribe
Ogni giorno+
 
Running out of time to catch up with new arXiv papers? We take the most impactful papers and present them as convenient podcasts. If you're a visual learner, we offer these papers in an engaging video format. Our service fills the gap between overly brief paper summaries and time-consuming full paper reads. You gain academic insights in a time-efficient, digestible format. Code behind this work: https://github.com/imelnyk/ArxivPapers Support this podcast: https://podcasters.spotify.com/pod/s ...
  continue reading
 
Artwork
 
A Change of Brand, produced by Matchstic, is a podcast about the world’s most loved consumer companies and their rebrand glory, drama, or disaster. In this series we will go deeper than color palettes and typography, talking with brand leaders and agency creatives to retell the story behind the change. Getting into the nitty-gritty, we learn about their leading-edge work, creative process, and approach to change management. Regardless of fame or fail, our mission is to tell these stories to ...
  continue reading
 
A daily update on the latest AI Research Papers. We provide a high level overview of a handful of papers each day and will link all papers in the description for further reading. This podcast is created entirely with AI by PocketPod. Head over to https://pocketpod.app to learn more.
  continue reading
 
The award winning Big Web Show features special guests and topics like web publishing, art direction, content strategy, typography, web technology, and more. It's everything web that matters. Hosted by Jeffrey Zeldman.
  continue reading
 
Artwork

1
Design Better

The Curiosity Department, LLC

Unsubscribe
Unsubscribe
Ogni mese+
 
Design Better co-hosts Eli Woolery and Aarron Walter explore the intersection of design, technology, and the creative process through conversations with inspiring guests across many creative fields. Whether you’re design curious or a design pro, Design Better is guaranteed to inspire and inform. Episodes are released semi-weekly. Vanity Fair calls Design Better, “sharp, to the point, and full of incredibly valuable information for anyone looking to better understand how to build a more innov ...
  continue reading
 
Artwork

1
Let's Pop Art

Gootrunkztrunk

Unsubscribe
Unsubscribe
Ogni mese
 
Art, art, and more art. As the world of art is evolving, changing, and heading towards Web 3, we dive into art with an open mindset. We see art through our own unique lens that weaves through the design world. That is the amazing thing about art, it impacts people in many different ways. The art world is vast. And with the advancements in technology, we are heading into unchartered waters. It is a very exciting time in history and we plan on having some fun on the show. We are excited to tak ...
  continue reading
 
Artwork

1
Beechtree Marketing

Beechtree Marketing

Unsubscribe
Unsubscribe
Ogni settimana
 
"Digital Growth Journeys with Beechtree Marketing" dives into the heart of digital marketing, presented by the experts at Beechtree Marketing of Cherry Hill, New Jersey. This podcast distills the essence of digital transformation, offering key strategies, trends, and insights into SEO, content marketing, and social media. Born from a pivot in 2020, Beechtree Marketing leverages its expertise to help clients across various sectors amplify their online presence, ensuring each episode is packed ...
  continue reading
 
Artwork

1
Analogous Analysis

British Academy of Graphic Design

Unsubscribe
Unsubscribe
Ogni mese
 
Analogous Analysis is the ultimate podcast for budding graphic designers and design fans! Join us as three experienced graphic designers delve into the dynamic realm of graphic design, addressing its ever-changing nature and what it truly means to be a graphic designer today. Each episode of this podcast, hosted by the British Academy of Graphic Design, will immerse you in an intimate roundtable discussion with our professional panel of graphic designers, who will share their unique experien ...
  continue reading
 
Artwork

1
Product Hunt Radio

Product Hunt

Unsubscribe
Unsubscribe
Ogni settimana
 
Product Hunt Radio is a a weekly podcast with the people creating and exploring the future. Tune in every week with Ryan Hoover and Abadesi Osunsade as they're joined by founders, investors, journalists, and makers to discuss the latest in tech.
  continue reading
 
Artwork

1
Mapbox

Mapbox

Unsubscribe
Unsubscribe
Ogni mese
 
Mapbox is the location data platform for mobile and web applications. We’re changing the way people move around cities and explore our world.
  continue reading
 
Artwork

1
The Innovative Creative

L.A.S. Podcast Network

Unsubscribe
Unsubscribe
Ogni mese
 
From the innovative mind of Devin Green comes The Innovative Creative, a podcast aiming to give you practical, ready-to-apply advice for your design work - all easily digestible and ready to consume on your commute to work!
  continue reading
 
Artwork

1
Know Your Nonsense

the Nonsensical Podcast Network

Unsubscribe
Unsubscribe
Ogni mese
 
Two self-proclaimed "experts" with varying competency, studying the nonsense we want and answering the questions nobody asked. Join us, engage with us, learn with us, and by all means, correct us, for we are completely, totally, enthusiastically fallible. Timothy Holloway: aspiring archeologist, teller of tales (both real and otherwise), and full-time food fanatic Riley Judd: aspiring game maker, begrudging graphic designer, and entertainment media addict. Join us, your hosts, as we kill som ...
  continue reading
 
Artwork

1
The Overlap

Elle Trost & Alex Trost

Unsubscribe
Unsubscribe
Ogni mese
 
The Overlap is a podcast that explores the intersection of Design and Front End Development. On the show, we discuss all things design and development in a fun and easy-to-listen-to way. Some topics we cover are typography, React, productivity, focus, designing for print, freelancing tips, and much more.
  continue reading
 
Artwork

1
By Design

Scout Studio

Unsubscribe
Unsubscribe
Ogni mese
 
By Design is a podcast hosted by members from Scout, Northeastern University's student-led design studio. Each episode, we'll break down impactful topics through a series of stories and conversations with our design friends. We'll offer a range of perspectives from our guests in an effort to further our collective understanding of the world of design.
  continue reading
 
Loading …
show series
 
AI Papers Podcast for 04/20/2024 Dynamic Typography: Bringing Words to Life Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing MeshLRM: Large Reconstruction Model for High-Quality Mesh MoA: Mixture-of-Attention for Subject-Context Disentanglement in Personalized Image Generation EdgeFusion: On-Device Text-to-Image Generatio…
  continue reading
 
Automated Dynamic Typography scheme deforms letters to convey meaning and adds vibrant movements based on user prompts, maintaining legibility and coherence in text animations. https://arxiv.org/abs//2404.11614 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/po…
  continue reading
 
In the digital landscape, the Call-to-Action (CTA) serves as the critical juncture where user interaction transforms into tangible outcomes, such as sales or subscriptions. The crafting of a CTA extends beyond mere text; it encompasses a strategic blend of visual elements—color and typography—each wielding significant influence over conversion rate…
  continue reading
 
https://arxiv.org/abs//2404.16820 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…
  continue reading
 
https://arxiv.org/abs//2404.16820 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…
  continue reading
 
https://arxiv.org/abs//2404.16710 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…
  continue reading
 
https://arxiv.org/abs//2404.16710 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…
  continue reading
 
IN2 training addresses lost-in-the-middle challenge in large language models by emphasizing information utilization in long contexts, leading to improved performance on various tasks. https://arxiv.org/abs//2404.16811 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.co…
  continue reading
 
IN2 training addresses lost-in-the-middle challenge in large language models by emphasizing information utilization in long contexts, leading to improved performance on various tasks. https://arxiv.org/abs//2404.16811 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.co…
  continue reading
 
AI Papers Podcast for 04/26/2024 How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source SuitesInteractive3D: Create What You Want by Interactive 3D GenerationLayer Skip: Enabling Early Exit Inference and Self-Speculative DecodingTele-FLM Technical ReportSEED-Bench-2-Plus: Benchmarking Multimodal Large Language Mo…
  continue reading
 
Consistency in content marketing often plays the understated sibling to flashier strategies like viral marketing or influencer partnerships. Yet, it establishes its dominion with unmatched reliability and coherence. This foundational strategy distinguishes itself in the expansive universe of digital marketing, offering a unique proposition that set…
  continue reading
 
DESIGN, BUILD, AND MODIFY — In his mid-20s, Scott Dadich told his editor at Texas Monthly, Evan Smith, that he wanted his job. A move like that is a combination of arrogance, youth, and frankly, balls. But you should also know that Dadich is an engineer. And what do engineers do? Well, according to one definition in Merriam-Webster, they “skillfull…
  continue reading
 
The paper explores inductive bias in transformer models, showing language modeling training leads to hierarchical generalization, supported by pruning experiments and Bayesian analysis. https://arxiv.org/abs//2404.16367 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.…
  continue reading
 
The paper explores inductive bias in transformer models, showing language modeling training leads to hierarchical generalization, supported by pruning experiments and Bayesian analysis. https://arxiv.org/abs//2404.16367 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.…
  continue reading
 
AutoGluon-Multimodal (AutoMM) is an open-source AutoML library for multimodal learning, offering easy fine-tuning with three lines of code. It supports various modalities and excels in basic and advanced tasks. https://arxiv.org/abs//2404.16233 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts…
  continue reading
 
AutoGluon-Multimodal (AutoMM) is an open-source AutoML library for multimodal learning, offering easy fine-tuning with three lines of code. It supports various modalities and excels in basic and advanced tasks. https://arxiv.org/abs//2404.16233 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts…
  continue reading
 
The paper introduces EXPO, a method to enhance large language models' alignment with human preference by extrapolating from weaker models, showing improved performance without additional training. https://arxiv.org/abs//2404.16792 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podc…
  continue reading
 
The paper introduces EXPO, a method to enhance large language models' alignment with human preference by extrapolating from weaker models, showing improved performance without additional training. https://arxiv.org/abs//2404.16792 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podc…
  continue reading
 
Retrieval-augmented generation (RAG) addresses issues in language models, but conflicts can arise between parametric knowledge and context, affecting knowledge updates. https://arxiv.org/abs//2404.16032 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/ar…
  continue reading
 
Retrieval-augmented generation (RAG) addresses issues in language models, but conflicts can arise between parametric knowledge and context, affecting knowledge updates. https://arxiv.org/abs//2404.16032 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/ar…
  continue reading
 
This paper addresses uncertainty estimation and calibration for large language models, proposing a supervised approach utilizing labeled data to enhance reliability and accuracy in LLM outputs. https://arxiv.org/abs//2404.15993 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcast…
  continue reading
 
This paper addresses uncertainty estimation and calibration for large language models, proposing a supervised approach utilizing labeled data to enhance reliability and accuracy in LLM outputs. https://arxiv.org/abs//2404.15993 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcast…
  continue reading
 
AI Papers Podcast for 04/25/2024 Hyper-SD: Trajectory Segmented Consistency Model for Efficient Image SynthesisA Multimodal Automated Interpretability AgentSEED-X: Multimodal Models with Unified Multi-granularity Comprehension and GenerationMultiBooth: Towards Generating All Your Concepts in an Image from TextLearning H-Infinity Locomotion Control…
  continue reading
 
Tldr; We’re transitioning to weekly episode releases, publishing an enhanced newsletter, and simplifying our premium subscription which will now be on Substack. Read the details below … Upgrade to paid Weekly episodes For years, we’ve been on a semi-weekly cadence for new episodes, but this schedule is proving too limiting for the amount of intervi…
  continue reading
 
AI Papers Podcast for 04/24/2024 OpenELM: An Efficient Language Model Family with Open-source Training and Inference FrameworkMulti-Head Mixture-of-ExpertsPegasus-v1 Technical ReportAlign Your Steps: Optimizing Sampling Schedules in Diffusion ModelsSnapKV: LLM Knows What You are Looking for Before Generation…
  continue reading
 
OpenELM, a state-of-the-art open language model, enhances accuracy using layer-wise scaling. Released with complete training framework, it empowers open research community. Available on GitHub and HuggingFace. https://arxiv.org/abs//2404.14619 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts:…
  continue reading
 
OpenELM, a state-of-the-art open language model, enhances accuracy using layer-wise scaling. Released with complete training framework, it empowers open research community. Available on GitHub and HuggingFace. https://arxiv.org/abs//2404.14619 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts:…
  continue reading
 
The paper introduces DUP prompting strategy to improve Large Language Models' performance on complex reasoning tasks, outperforming Zero-Shot CoT on diverse datasets, achieving state-of-the-art results. https://arxiv.org/abs//2404.14963 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https:…
  continue reading
 
The paper introduces DUP prompting strategy to improve Large Language Models' performance on complex reasoning tasks, outperforming Zero-Shot CoT on diverse datasets, achieving state-of-the-art results. https://arxiv.org/abs//2404.14963 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https:…
  continue reading
 
SnapKV is a fine-tuning-free method that efficiently reduces Key-Value cache size in Large Language Models, maintaining performance while enhancing memory and time efficiency for long input sequences. https://arxiv.org/abs//2404.14469 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://…
  continue reading
 
SnapKV is a fine-tuning-free method that efficiently reduces Key-Value cache size in Large Language Models, maintaining performance while enhancing memory and time efficiency for long input sequences. https://arxiv.org/abs//2404.14469 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://…
  continue reading
 
MH-MoE addresses low expert activation and lack of fine-grained analysis in SMoE by using a multi-head mechanism to enhance context understanding and expert activation. https://arxiv.org/abs//2404.15045 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/ar…
  continue reading
 
MH-MoE addresses low expert activation and lack of fine-grained analysis in SMoE by using a multi-head mechanism to enhance context understanding and expert activation. https://arxiv.org/abs//2404.15045 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/ar…
  continue reading
 
Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions FlowMind: Automatic Workflow Generation with LLMs Music Consistency Models How Good Are Low-bit Quantized LLaMA3 Models? An Empirical Study
  continue reading
 
LLMs are vulnerable to attacks due to equal priority given to all prompts. Proposed instruction hierarchy teaches models to ignore lower-priority instructions, enhancing robustness with minimal impact on capabilities. https://arxiv.org/abs//2404.13208 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple P…
  continue reading
 
LLMs are vulnerable to attacks due to equal priority given to all prompts. Proposed instruction hierarchy teaches models to ignore lower-priority instructions, enhancing robustness with minimal impact on capabilities. https://arxiv.org/abs//2404.13208 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple P…
  continue reading
 
https://arxiv.org/abs//2404.14367 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…
  continue reading
 
https://arxiv.org/abs//2404.14367 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…
  continue reading
 
Introducing phi-3-mini, a high-performing language model trained on a large dataset, with smaller versions phi-3-small and phi-3-medium showing even better performance. https://arxiv.org/abs//2404.14219 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/ar…
  continue reading
 
Introducing phi-3-mini, a high-performing language model trained on a large dataset, with smaller versions phi-3-small and phi-3-medium showing even better performance. https://arxiv.org/abs//2404.14219 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/ar…
  continue reading
 
AI Papers Podcast for 04/23/2024 PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generation TextSquare: Scaling up Text-Centric Visual Instruction Tuning Does Gaussian Splatting need SFM Initialization? How Far Can We Go with Practical Function-Level Program Repair? AutoCrawler: A Progressive Understanding Web Agent for Web Crawler…
  continue reading
 
Approach estimates latent knowledge in large language models using in-context learning, showing differences in factual knowledge across models and sizes. https://arxiv.org/abs//2404.12957 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id16…
  continue reading
 
Approach estimates latent knowledge in large language models using in-context learning, showing differences in factual knowledge across models and sizes. https://arxiv.org/abs//2404.12957 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id16…
  continue reading
 
HalluciBot predicts hallucination probability before generation in Large Language Models, aiding in query quality assessment and user accountability, potentially reducing computational waste. https://arxiv.org/abs//2404.12535 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.…
  continue reading
 
HalluciBot predicts hallucination probability before generation in Large Language Models, aiding in query quality assessment and user accountability, potentially reducing computational waste. https://arxiv.org/abs//2404.12535 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.…
  continue reading
 
Evaluating language models' in-context learning performance faces challenges. A stronger random baseline is proposed, improving evaluation accuracy and predicting held-out performance effectively. https://arxiv.org/abs//2404.13020 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podc…
  continue reading
 
Evaluating language models' in-context learning performance faces challenges. A stronger random baseline is proposed, improving evaluation accuracy and predicting held-out performance effectively. https://arxiv.org/abs//2404.13020 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podc…
  continue reading
 
AI Papers Podcast for 04/21/2024Ctrl-Adapter: An Efficient and Versatile Framework for Adapting Diverse Controls to Any Diffusion ModelHQ-Edit: A High-Quality Dataset for Instruction-based Image EditingTango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference OptimizationTextHawk: Exploring Efficient Fine-Grained Percept…
  continue reading
 
The paper compares Direct Preference Optimization (DPO) and Proximal Policy Optimization (PPO) in aligning large language models with human feedback, showing PPO outperforms DPO in various RLHF testbeds. https://arxiv.org/abs//2404.10719 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https…
  continue reading
 
Loading …

Guida rapida