))
The AI Engineer newsletter + Top 10 US Tech podcast. Exploring AI UX, Agents, Devtools, Infra, Open Source Models. See https://latent.space/about for highlights from Chris Lattner, Andrej Karpathy, George Hotz, Simon Willison, Emad Mostaque, et al!
…
continue reading
Whether you're a practicing Latent Print Examiner or you're interested in forensics and true crime, the Double Loop Podcast is a weekly show featuring Glenn Langenburg and Eric Ray discussing latent print topics, current events in forensic science, the newest research articles, interesting guests, and analysis of notable cases from a forensic scientist perspective.
…
continue reading
Deep diving classic or infamous movies.
…
continue reading
in which Tallahassee, FL-based photographer Dustin Hern talks [sometimes with friends, sometimes to himself] about life and art and stuff
…
continue reading
Latent - present and capable of emerging or developing but not now visible, obvious, active, or symptomatic. Welcome to a show of unedited conversations on the value of organizational learning and team performance at work. Hear from people defining the modern workforce culture and unleashing their teams' full potential. Hosted by Tudor Tarlev and produced at #onesevenx studio.
…
continue reading
You are a Divine Being. Follow keys to spiritual awakening. Learn how to access your inner reality and higher consciousness using Truth from Ascended Masters' wisdom. The Ascended Masters are your truest teachers and best friends...
…
continue reading
A weekly pre-sleep, post-thought audio play investigating the nature of latent boredom and the strip-mining of emotional bereavement. Semi is written & performed by Mark Simpson, Garth Simmons and Jim-John Harkness, and is produced by Stephen Landerhour for BBC Sounds.
…
continue reading
hi all, thanks for being here. we get the commune aspiring for growth and striving towards excellence ,the nitty gritty of handling the latent or dormant failures to be ....episode being the first of this series addressing POSH act , the compliance to which is mandatory by all organisations in our subcontinent -India.
…
continue reading
A podcast highlighting key articles in the current issue of Emerging Infectious Diseases, a journal from the Centers for Disease Control and Prevention (CDC).
…
continue reading
A podcast dedicated to all things quantitative, ranging from the relevant to the highly irrelevant. Co-hosts Patrick Curran and Greg Hancock talk about serious statistical topics, but without taking themselves too seriously. Think: CarTalk hi-jacked by the two grumpy old guys from the Muppets, grousing about quantitative methods, statistics, and data analysis, all presented to you with the production value of a 6th grade school project. But in a good way.
…
continue reading
In this podcast, Thomas Bertels explores with thought leaders and executives how to make work more productive, valuable, meaningful, and impactful.
…
continue reading
Coffee Conversations with Local Legends, the place where we break down the habits and latent abilities of people living their version of what it means to be whole.
…
continue reading
Empowered Empire is a podcast about you. We believe there is latent capability in every human being for a greater experience and legacy on earth. You will find tips for living well, strategies for growth, mindset tweaks, and motivation wherever you find the Empowered Empire brand. The mission of Empowered Empire is to equip you with the tools and mindset to rule yourself well. Our commitment is to you and providing value in your life. We are on a journey together. Put simply, Empowered Empir ...
…
continue reading
This show is dedicated to educating you on culturally diverse communities so you can be better healers.
…
continue reading
Mutations is a podcast exploring latent futures in the radical present. We inhabit a time between times, between worlds—what are the emergent potentials, articulated visions, or, in a word, ‘mutations’ that can help us to pathfind our way into the future? What forms of integrative thinking and being are required for this leap? Author (Seeing Through the World), show host, and integral philosopher Jeremy D Johnson explores these questions through solo podcasts, readings, and with featured guests.
…
continue reading
Things about Svelte. Sometimes weekly, sometimes not.
…
continue reading
Christina Warren & Brett Terpstra have odd sleep schedules. They nerd out over varied interests: gadgets, software, and life in a connected world. Tune in to find out what keeps them up at night.
…
continue reading
Book One of The Dawning of Power trilogy Echoes of the ancients' power are distant memories, tattered and faded by the passage of eons, but that is about to change. A new dawn has arrived. Latent abilities, harbored in mankind's deepest fibers, wait to be unleashed. Ancient evils awaken, and old fears ignite the fires of war. In times such as these, ordinary people have the power to save the world… or destroy it. The books of the Dawning of Power trilogy are World of Godsland novels.
…
continue reading
The Everyman Project was created for the common man, in hopes to activate the latent potential that we believe everyone has inside of them. It’s never too early or too late to be great. Greatness looks different on everyone, however the common denominators everyone must accept when becoming great are work, sacrifice, integrity, resilience and discipline. Those are the foundations of all of success be it spiritual, mental, relational or material.
…
continue reading
Although we may not see tuberculosis cases too often, that doesn’t mean this disease isn’t out there. In fact, up to 13 million people are carrying the latent infection, which can asymptomatically exist in a patient for years and can be fatal if it isn’t caught early enough. That’s why knowing which of our patients are at risk and using effective screening measures are now more critical than ever. In the hopes of eradicating this disease once and for all, Tackling TB brings together experts ...
…
continue reading
Interviews with scholars of human rights about their new books
…
continue reading
Happy Horror Time is hosted by Tim Murdock & Matt Emert – two off-the-wall gay guys who LOVE scary movies! We discuss new horror films (spoiling everything along the way) and interview horror stars & insiders. We're unpolished, unscripted, and a ton of fun!
…
continue reading
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
Interviews with authors of JAMA Guide to Statistics and Methods chapters about common and new statistics and methods used in clinical research and reported in medical journals.
…
continue reading
Episodic podcast discussing all things faith and Christianity related with an emphasis on being actively in ministry efforts to live out the purpose of the BODY of Christ (aka the Church) in tangible ways that point the communities around us back to God and Jesus. On this podcast you'll find a variety of special bonus programming with unique focuses, such as our workday morning devotional series, "...But God" and our lifestyle conversational series "Ark of Inclusion" or our series highlighti ...
…
continue reading
Spreading Happiness is a compilation of elegant teachings on conscious living. A blend of inspiration, timeless wisdom, story telling and guided meditations. These teachings are presented by Nithya Shanti, an internationally respected spiritual teacher. He was formerly a Buddhist monk who trained in forest meditation monasteries in South East Asia. He is committed to sharing practical wisdom teachings for happiness and enlightenment with people in a joyful and transformational way. Blending ...
…
continue reading
A spinoff of ”Bigfoot on Trial”. Listen to a 20 year police/ CSI/ Fingerprint Examiner share his thoughts on bigfoot as he commutes back and forth from work. A little about my background: I am currently assigned to the Crime Scene Investigations Section at a local department in Georgia I am a certified latent fingerprint examiner and certified crime scene tech. I have been a certified peace officer in the State of Georgia for over 19 years and am a graduate of the National Forensic Academy i ...
…
continue reading
Past and present advances in the fields of agronomic, crop, soil, and environmental sciences. Enjoy interviews with researchers published in journals, books, and magazines from the American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America. Opinions and conclusions expressed by authors are their own and are not considered as those of the American Society of Agronomy, Crop Science Society of America, Soil Science Society of America, its staff, its membe ...
…
continue reading
The Natural Birth Podcast is here to bring Embodied Birth Wisdom from Women from All over the World sharing their Natural & Empowering Birth Stories, with the aim to Inspire and Encourage Mamas to Believe in their Bodies and in the Natural Birth Process. Visit Anna and find out about all her offerings in the world at www.thenaturalbirthcourse.com Follow The Podcast on Instagram @thenaturalbirthpodcast Patreon patreon.com/thenaturalbirthpodcast
…
continue reading
Welcome to the Complexity Premia podcast from Coolabah Capital, which is hosted by Christopher Joye, CIO and portfolio manager of Coolabah Capital, and Ying Yi Ann Cheng, a portfolio management director. The Complexity Premia podcast strives to deconstruct modern investment problems for wholesale (not retail) participants in capital markets. If this recording contains reference to any financial products, that reference does not constitute financial advice or a recommendation, and may not be ...
…
continue reading
Holly Herndon & Mat Dryhurst, optimistic about the 21st Century. Patrons and Channel holders get access to weekly episodes as they drop, the free feed is time delayed <3
…
continue reading
Hosted by author and entrepreneur, Steve Glaveski, Future Squared brings you conversations with preeminent thought leaders from myriad disciplines such as startups and entrepreneurship, technology, corporate innovation, psychology, philosophy, economics, politics, biology, human performance and productivity, motivation and self-development, management and leadership, health and fitness and more to help you think in a multi-disciplinary way, avoid relying on the first piece of information tha ...
…
continue reading
"John Shore is American's leading non-douchey Christian." -- Dan Savage, Savage Love, and founder of the It Gets Better Project. "John Shore is awesome. The minute I started reading his stuff, I knew he was a brother from another mother." -- Rob Bell, author of Love Wins, Sex God, and Velvet Elvis. "John Shore is a gadfly, calling the Christian Church everywhere to act the way it says it believes about love and justice, which of course makes him an uncomfortable presence in those churches th ...
…
continue reading
Podcasts, class lectures and resources from Jeff Curto
…
continue reading
The Experience Customers podcast is the best way for CX leaders and learners alike to learn about a new groundbreaking Customer Needs methodology that’s being adopted globally.
…
continue reading
Recorded in Karachi and Lahore in the summer of 2011, Another Pakistan offers a unique view of a complicated country. Host Christopher Lydon interviews singers, story-tellers and artists to paint a picture rarely seen in mainstream media.
…
continue reading
Sonic Acts is an interdisciplinary arts organisation based in Amsterdam. In 2024, we will be celebrating 30 years of cutting-edge experiments in sound, moving image and contemporary theory. Join us from 2 February to 24 March.
…
continue reading
Accompanying podcast to the blog, Blend This Trend at blendthistrend.com
…
continue reading
The Transformative Ideas Podcast (formerly the ACIT Science Podcast). This podcast is to provide a glimpse into the life of scientists: to learn about the ideas they are passionate about, to find out what gets them out of bed every day to face the challenges and frustrations of working at the frontier to the unknown, and to share in some of the most important lessons they have learned in their career. For inquiries, reach out to manu.brenn@gmail.com.
…
continue reading
Displaced in Space is a weekly podcast where married couple and casual Star Trek fans Sam and Angelina watch "Star Trek: Voyager" for the first time! We are such casual Trek fans that we have never seen Voyager, so come with us as we Voyage through the Delta Quadrant! See what adventure awaits us as we explore strange new worlds, seek out new life and new civilizations, and belatedly go where many have gone before!
…
continue reading
1
Why Google failed to make GPT-3 + why Multimodal Agents are the path to AGI — with David Luan of Adept
Our next SF event is AI UX 2024 - let’s see the new frontier for UX since last year! Last call: we are recording a preview of the AI Engineer World’s Fair with swyx and Ben Dunphy, send any questions about Speaker CFPs and Sponsor Guides you have! Alessio is now hiring engineers for a new startup he is incubating at Decibel: Ideal candidate is an “…
…
continue reading
1
The Human Side of Latent TB: Community Insights with Franky and Mohammed (Ep. 42)
41:36
41:36
Play later
Play later
Lists
Like
Liked
41:36
Summary: We dive into the complexities of navigating the US healthcare system with insights from Franky and Mohammad, focusing specifically on the Marshallese community's challenges and the widespread misconceptions about TB screenings in diverse populations. Together, we explore culturally responsive strategies to improve trust and treatment adher…
…
continue reading
1
[short] Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation
1:42
1:42
Play later
Play later
Lists
Like
Liked
1:42
Latent Adversarial Diffusion Distillation (LADD) improves image synthesis speed and quality compared to existing methods by leveraging generative features from pretrained latent diffusion models. https://arxiv.org/abs//2403.12015 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podca…
…
continue reading
Erin Dawson, fresh off her latest tour, joins the gang to tell tales from the road. Plus the much-anticipated picks of the week! Sponsor Hims is changing men’s healthcare by providing simple and convenient access to science-backed treatments for erectile dysfunction, hair loss, weight loss, and more. Start your free online visit today at hims.com/o…
…
continue reading
1
3 Home Births with a Long Latent and a Quick Active Phase
1:26:53
1:26:53
Play later
Play later
Lists
Like
Liked
1:26:53
Today on The Natural Birth Podcast we have Maria. Maria is a mama of 3 from Sweden. She has a background in journalism and has written three books about pregnancy, birth and the postpartum as well as working as a doula and a massage therapist. She’s also the co-founder of Braido Organics with Kristina that you heard from just a couple of weeks back…
…
continue reading
Dr. Amera Khan, a technical officer at the Stop TB Partnership in Geneva, Switzerland, and Sarah Gregory discuss treatment of latent TB among US-bound immigrants in Vietnam.
…
continue reading
1
[short] BioMedLM: A 2.7B Parameter Language Model Trained On Biomedical Text
2:26
2:26
Play later
Play later
Lists
Like
Liked
2:26
BioMedLM, a 2.7 billion parameter model trained on PubMed data, competes with larger models in biomedical question-answering tasks, showing potential for efficient NLP applications in biomedicine. https://arxiv.org/abs//2403.18421 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podc…
…
continue reading
1
BioMedLM: A 2.7B Parameter Language Model Trained On Biomedical Text
17:24
17:24
Play later
Play later
Lists
Like
Liked
17:24
BioMedLM, a 2.7 billion parameter model trained on PubMed data, competes with larger models in biomedical question-answering tasks, showing potential for efficient NLP applications in biomedicine. https://arxiv.org/abs//2403.18421 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podc…
…
continue reading
1
[short] Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models
2:18
2:18
Play later
Play later
Lists
Like
Liked
2:18
Mini-Gemini enhances Vision Language Models by refining visual tokens, improving data quality, and enabling VLM-guided generation, achieving leading performance in zero-shot benchmarks. https://arxiv.org/abs//2403.18814 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.…
…
continue reading
1
Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models
15:51
15:51
Play later
Play later
Lists
Like
Liked
15:51
Mini-Gemini enhances Vision Language Models by refining visual tokens, improving data quality, and enabling VLM-guided generation, achieving leading performance in zero-shot benchmarks. https://arxiv.org/abs//2403.18814 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.…
…
continue reading
1
[short] Long-form factuality in large language models
1:46
1:46
Play later
Play later
Lists
Like
Liked
1:46
Large language models can assess long-form factuality using a method called SAFE, achieving superhuman performance and cost-effectiveness compared to human annotators. https://arxiv.org/abs//2403.18802 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arx…
…
continue reading
1
Long-form factuality in large language models
11:45
11:45
Play later
Play later
Lists
Like
Liked
11:45
Large language models can assess long-form factuality using a method called SAFE, achieving superhuman performance and cost-effectiveness compared to human annotators. https://arxiv.org/abs//2403.18802 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arx…
…
continue reading
1
Michael Davis, "Freedom Undone: The Assault on Liberal Values in Hong Kong" (Association for Asian Studies, 2023)
1:03:30
1:03:30
Play later
Play later
Lists
Like
Liked
1:03:30
"What happened in Hong Kong is not an anomaly but a warning" - Hong Kong Human Rights defender Chow Hang Tung, speech written from prison upon receiving a human rights award. In our interview today, I spoke with Professor Michael C. Davis, author of Freedom Undone: The Assault on Liberal Values and Institutions in Hong Kong (AAS and Columbia UP, 20…
…
continue reading
1
[short] Fully-fused Multi-Layer Perceptrons on Intel Data Center GPUs
2:01
2:01
Play later
Play later
Lists
Like
Liked
2:01
SYCL MLP implementation optimized for Intel Data Center GPU Max 1550 improves performance by minimizing global memory accesses, increasing data reuse, outperforming CUDA in various applications. https://arxiv.org/abs//2403.17607 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcas…
…
continue reading
1
Fully-fused Multi-Layer Perceptrons on Intel Data Center GPUs
19:12
19:12
Play later
Play later
Lists
Like
Liked
19:12
SYCL MLP implementation optimized for Intel Data Center GPU Max 1550 improves performance by minimizing global memory accesses, increasing data reuse, outperforming CUDA in various applications. https://arxiv.org/abs//2403.17607 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcas…
…
continue reading
1
Work Matters S5E4: Dart Lindsley - Middle Management
49:14
49:14
Play later
Play later
Lists
Like
Liked
49:14
In today's Work Matters, Thomas talks with Dart Lindsley, head of Global Process Excellence for People Operations at Google. He previously led the HR Transformation Planning and Analysis organization at Cisco Systems and is the host of the popular "Work for Humans" podcast. In this discussion, Dart unpacks what makes the middle manager role so chal…
…
continue reading
1
[short] The Unreasonable Ineffectiveness of the Deeper Layers
2:15
2:15
Play later
Play later
Lists
Like
Liked
2:15
Study explores layer pruning in pretrained LLMs, finding minimal performance drop until half of layers removed. Optimal block identified for pruning, followed by finetuning with PEFT methods for efficiency and improved inference. https://arxiv.org/abs//2403.17887 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_pa…
…
continue reading
1
The Unreasonable Ineffectiveness of the Deeper Layers
17:44
17:44
Play later
Play later
Lists
Like
Liked
17:44
Study explores layer pruning in pretrained LLMs, finding minimal performance drop until half of layers removed. Optimal block identified for pruning, followed by finetuning with PEFT methods for efficiency and improved inference. https://arxiv.org/abs//2403.17887 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_pa…
…
continue reading
1
[short] Data Mixing Laws: Optimizing Data Mixtures by Predicting Language Modeling Performance
2:04
2:04
Play later
Play later
Lists
Like
Liked
2:04
Quantitative data mixing laws predict large language model performance based on mixture proportions, guiding optimal data selection and training strategies for improved outcomes. https://arxiv.org/abs//2403.16952 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/…
…
continue reading
1
Data Mixing Laws: Optimizing Data Mixtures by Predicting Language Modeling Performance
20:36
20:36
Play later
Play later
Lists
Like
Liked
20:36
Quantitative data mixing laws predict large language model performance based on mixture proportions, guiding optimal data selection and training strategies for improved outcomes. https://arxiv.org/abs//2403.16952 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/…
…
continue reading
AIOS is an operating system integrating large language models to optimize resource allocation, context switching, and concurrent agent execution, aiming to enhance LLM agent performance and efficiency. https://arxiv.org/abs//2403.16971 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https:/…
…
continue reading
AIOS is an operating system integrating large language models to optimize resource allocation, context switching, and concurrent agent execution, aiming to enhance LLM agent performance and efficiency. https://arxiv.org/abs//2403.16971 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https:/…
…
continue reading
A seedy motel in the middle of nowhere is always a good spot for a horror movie. Now add in some creepy ghosts and an escaped mental patient and you’ve got “Night Shift.” Tune in as we break down this new thriller — from its similarities to “Psycho” to that major twist at the end, and of course, what happened in the sinkhole.…
…
continue reading
1
[short] FOLLOWIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions
2:56
2:56
Play later
Play later
Lists
Like
Liked
2:56
FOLLOWIR introduces a dataset to help Information Retrieval models better follow complex instructions, showing potential for significant improvements in understanding and using instructions. https://arxiv.org/abs//2403.15246 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.a…
…
continue reading
1
FOLLOWIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions
14:33
14:33
Play later
Play later
Lists
Like
Liked
14:33
FOLLOWIR introduces a dataset to help Information Retrieval models better follow complex instructions, showing potential for significant improvements in understanding and using instructions. https://arxiv.org/abs//2403.15246 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.a…
…
continue reading
1
LLM2LLM: Boosting LLMs with Novel Iterative Data Enhancement
13:32
13:32
Play later
Play later
Lists
Like
Liked
13:32
LLM2LLM proposes a data augmentation strategy using a teacher LLM to enhance fine-tuning on low-data tasks, outperforming traditional methods and reducing data curation efforts. https://arxiv.org/abs//2403.15042 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/p…
…
continue reading
1
[short] Can large language models explore in-context?
1:49
1:49
Play later
Play later
Lists
Like
Liked
1:49
The study explores Large Language Models' ability to engage in exploration without training interventions, finding external summarization crucial for robust exploratory behavior. https://arxiv.org/abs//2403.15371 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/…
…
continue reading
1
Can large language models explore in-context?
19:22
19:22
Play later
Play later
Lists
Like
Liked
19:22
The study explores Large Language Models' ability to engage in exploration without training interventions, finding external summarization crucial for robust exploratory behavior. https://arxiv.org/abs//2403.15371 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/…
…
continue reading
1
[short] Detoxifying Large Language Models via Knowledge Editing
1:55
1:55
Play later
Play later
Lists
Like
Liked
1:55
The paper explores detoxifying Large Language Models using knowledge editing techniques, introducing SafeEdit benchmark and proposing DINM baseline for efficient detoxification with minimal performance impact. https://arxiv.org/abs//2403.14472 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts:…
…
continue reading
1
Detoxifying Large Language Models via Knowledge Editing
16:57
16:57
Play later
Play later
Lists
Like
Liked
16:57
The paper explores detoxifying Large Language Models using knowledge editing techniques, introducing SafeEdit benchmark and proposing DINM baseline for efficient detoxification with minimal performance impact. https://arxiv.org/abs//2403.14472 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts:…
…
continue reading
1
We Birth The Way We Live. Sarah’s Easeful Birth Stories
55:37
55:37
Play later
Play later
Lists
Like
Liked
55:37
Today on The Natural Birth Podcast we have Sarah. Sarah is a mama of 2 from Cornwall, UK. She is a holistic nutritionist and passionate about all aspects of health and wellness, the importance of our connection to nature and living as empowered beings for a happy, healthy life. When Sarah first became pregnant she was fearful of birth, but through …
…
continue reading
1
[short] Evolutionary Optimization of Model Merging Recipes
1:54
1:54
Play later
Play later
Lists
Like
Liked
1:54
Evolutionary algorithms automate creating powerful foundation models by merging diverse open-source models, achieving state-of-the-art performance in Japanese language and culture tasks. https://arxiv.org/abs//2403.13187 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple…
…
continue reading
1
Evolutionary Optimization of Model Merging Recipes
23:10
23:10
Play later
Play later
Lists
Like
Liked
23:10
Evolutionary algorithms automate creating powerful foundation models by merging diverse open-source models, achieving state-of-the-art performance in Japanese language and culture tasks. https://arxiv.org/abs//2403.13187 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple…
…
continue reading
1
[short] Recourse for Reclamation: Chatting with Generative Language Models
1:59
1:59
Play later
Play later
Lists
Like
Liked
1:59
Toxicity scoring in language models can hinder information access and cultural norms. A novel recourse mechanism allows users to set toxicity thresholds dynamically for improved interaction. https://arxiv.org/abs//2403.14467 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.a…
…
continue reading
1
Recourse for Reclamation: Chatting with Generative Language Models
8:32
8:32
Play later
Play later
Lists
Like
Liked
8:32
Toxicity scoring in language models can hinder information access and cultural norms. A novel recourse mechanism allows users to set toxicity thresholds dynamically for improved interaction. https://arxiv.org/abs//2403.14467 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.a…
…
continue reading
1
[short] A Unified Framework for Model Editing
1:49
1:49
Play later
Play later
Lists
Like
Liked
1:49
The paper introduces a framework unifying ROME and MEMIT model editing techniques under a preservation-memorization objective, presenting EMMET as a new batched memory-editing algorithm for Transformers. https://arxiv.org/abs//2403.14236 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https…
…
continue reading
The paper introduces a framework unifying ROME and MEMIT model editing techniques under a preservation-memorization objective, presenting EMMET as a new batched memory-editing algorithm for Transformers. https://arxiv.org/abs//2403.14236 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https…
…
continue reading
1
[short] Language Models Can Reduce Asymmetry in Information Markets
1:47
1:47
Play later
Play later
Lists
Like
Liked
1:47
The paper explores the buyer's inspection paradox in information markets using simulated agents with language models, focusing on biases, pricing effects, and quality outcomes. https://arxiv.org/abs//2403.14443 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/po…
…
continue reading
1
Language Models Can Reduce Asymmetry in Information Markets
14:34
14:34
Play later
Play later
Lists
Like
Liked
14:34
The paper explores the buyer's inspection paradox in information markets using simulated agents with language models, focusing on biases, pricing effects, and quality outcomes. https://arxiv.org/abs//2403.14443 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/po…
…
continue reading
Unsupervised training data and neural scaling have led to larger models, but memory bandwidth is now the main bottleneck in AI applications, requiring architecture and strategy redesign. https://arxiv.org/abs//2403.14123 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple…
…
continue reading
Unsupervised training data and neural scaling have led to larger models, but memory bandwidth is now the main bottleneck in AI applications, requiring architecture and strategy redesign. https://arxiv.org/abs//2403.14123 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple…
…
continue reading
1
Patryk I. Labuda, "International Criminal Tribunals and Domestic Accountability" (Oxford UP, 2023)
58:17
58:17
Play later
Play later
Lists
Like
Liked
58:17
In the 1990s, the promise of justice for atrocity crimes was associated with the revival of international criminal tribunals (ICTs). More recently, however, there has been a renewed emphasis on domestic accountability for international crimes across the globe. In identifying a 'complementarity turn', a paradigm shift toward domestic accountability …
…
continue reading
Eric Ray and Glenn Langenburg sit down for a chat with Steve Johnson about face examiners, face comparisons, and the IAI. Steve's background includes latent prints, forensic art, face comparison, crime scene, and many positions with the IAI. He shares the future of the face discipline and how the IAI is working towards a certification process.…
…
continue reading
1
RewardBench: Evaluating Reward Models for Language Modeling
19:52
19:52
Play later
Play later
Lists
Like
Liked
19:52
The paper introduces REWARDBENCH, a dataset and code-base for evaluating reward models in aligning language models with human preferences, shedding light on their capabilities and limitations. https://arxiv.org/abs//2403.13787 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts…
…
continue reading
1
Evaluating Frontier Models for Dangerous Capabilities
23:54
23:54
Play later
Play later
Lists
Like
Liked
23:54
https://arxiv.org/abs//2403.13793 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
1
Reverse Training to Nurse the Reversal Curse
13:28
13:28
Play later
Play later
Lists
Like
Liked
13:28
Large language models struggle with generalizing from "A has B" to "B is a feature of A" due to Zipf's law. Reverse training doubles tokens, improving performance on reversal tasks. https://arxiv.org/abs//2403.13799 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/…
…
continue reading
1
[short] Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs
1:50
1:50
Play later
Play later
Lists
Like
Liked
1:50
Technique proposed to transfer reasoning capabilities from large-language models to vision-language models, achieving state-of-the-art performance on ChartQA, PlotQA, and FigureQA tasks. https://arxiv.org/abs//2403.12596 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple…
…
continue reading
