Is It Time to Rethink LLM Pre-Training? with Aditi Raghunathan - #747

Is It Time to Rethink LLM Pre-Training? with Aditi Raghunathan - #747

Today, we're joined by Aditi Raghunathan, assistant professor at Carnegie Mellon University, to discuss the limitations of LLMs and how we can build more adaptable and creative models. We dig into her ICML 2025 Outstanding Paper Award winner, “Roll the dice & look before you leap: Going beyond the creative limits of next-token prediction,” which examines why LLMs struggle with generating truly novel ideas. We dig into the "Roll the dice" approach, which encourages structured exploration by injecting randomness at the start of generation, and the "Look before you leap" concept, which trains models to take "leaps of thought" using alternative objectives to create more diverse and structured outputs. We also discuss Aditi’s papers exploring the counterintuitive phenomenon of "catastrophic overtraining," where training models on more data improves benchmark performance but degrades their ability to be fine-tuned for new tasks, and dig into her lab's work on creating more controllable and reliable models, including the concept of "memorization sinks," an architectural approach to isolate and enable the targeted unlearning of specific information. The complete show notes for this episode can be found at https://twimlai.com/go/747.

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(795)

From Voice Agents to AI Avatars with Alexander Smola - #777

From Voice Agents to AI Avatars with Alexander Smola - #777

Voice AI has gotten remarkably good, but natural conversation remains a high bar. Small delays, awkward interruptions, or the wrong tone can quickly break the illusion—and adding vision and visual pre...

17 Sep 1h 4min

Do AI Tokenomics Matter More Than Model Benchmarks? with Chris Potts - #776

Do AI Tokenomics Matter More Than Model Benchmarks? with Chris Potts - #776

As reasoning models consume more tokens and AI systems become more expensive to run, understanding what those tokens actually buy is becoming increasingly important. In this episode, Stanford professo...

9 Sep 59min

World Models and the Future of Spatial AI with Justin Johnson - #775

World Models and the Future of Spatial AI with Justin Johnson - #775

In this episode, Justin Johnson, co-founder of World Labs, joins us to discuss world models and the emerging field of spatial AI. We explore why many researchers see capabilities beyond language as an...

1 Sep 1h 6min

Why the Next AI Breakthrough May Come from Physics with Max Welling - #774

Why the Next AI Breakthrough May Come from Physics with Max Welling - #774

The conventional wisdom in AI is that the next breakthrough will come from more compute, more data, and larger models. But what if the next leap comes from somewhere else? In this episode, Max Wellin...

26 Aug 58min

Why Image Generation Needs More Than Bigger Models with Fatih Porikli - #773

Why Image Generation Needs More Than Bigger Models with Fatih Porikli - #773

Text-to-image models have become remarkably good at producing realistic images. But realism isn’t the same as correctness. Ask for several distinct people, a specific composition, or a high-resolution...

12 Aug 56min

Why Models Are AI’s Next Training Dataset with Damian Borth - #772

Why Models Are AI’s Next Training Dataset with Damian Borth - #772

For more than a decade, AI has advanced by training ever-larger models on ever-larger datasets. But as high-quality training data becomes harder to find and pretraining grows increasingly expensive, r...

27 Jul 47min

How AI Learns to Smell with Alex Wiltschko - #771

How AI Learns to Smell with Alex Wiltschko - #771

In this episode, Alex Wiltschko, founder and CEO of Osmo, joins the show to discuss his goal of giving computers a sense of smell and what it takes to build olfactory intelligence. We explore the sci...

8 Jul 59min

Why AI Agents Break the GenAI Security Model with Devvret Rishi - #770

Why AI Agents Break the GenAI Security Model with Devvret Rishi - #770

In this episode, Sam talks with Dev Rishi, GM of AI at Rubrik, about what happens when agents move beyond answering questions and start taking action across tools, systems, and business processes. We...

16 Jun 56min

Populært innen Politikk og nyheter

giver-og-gjengen-vg
aftenpodden
forklart
popradet
aftenpodden-usa
stopp-verden
rss-gukild-johaug
bt-dokumentar-2
det-store-bildet
dine-penger-pengeradet
fotballpodden-2
rss-espen-lee-usensurert
nokon-ma-ga
hanna-de-heldige
rss-ness
aftenbla-bla
e24-podden
rss-penger-polser-og-politikk
frokostshowet-pa-p5
unitedno