Neural Notions – Str8Digital
de Str8Digital
Episódio mais recente
Qwen4-Max: Alibaba's latest frontier model release and what it means…
Qwen4-Max: Alibaba's latest frontier model release and what it means…
Sobre este podcast
A show exploring AI agents, model behavior, workflow design, and practical experiments with modern artificial intelligence systems, machine learning, and large language models.
Episódios
Os 20 episódios mais recentes de 82
- Qwen4-Max: Alibaba's latest frontier model release and what it means for the open-weight AI race · 22 min · Temporada 1 Episódio 82Neural Notions dives into Qwen4-Max: Alibaba's latest frontier model release and what it means for the open-weight AI race.
- Chinese frontier AI models carry the fingerprint of export-controlled GPUs: what checkpoint forensics reveals about US export control enforcement · 23 min · Temporada 1 Episódio 81Neural Notions dives into Chinese frontier AI models carry the fingerprint of export-controlled GPUs: what checkpoint forensics reveals about US export control enforcement.
- Anthropic's Claude Fable 5.2 beats GPT-6 in multi-stage cyberattack benchmark: what adversarial AI benchmarks reveal about model capabilities · 26 min · Temporada 1 Episódio 80Neural Notions dives into Anthropic's Claude Fable 5.2 beats GPT-6 in multi-stage cyberattack benchmark: what adversarial AI benchmarks reveal about model capabilities.
- Claude Code "Don't Ask Again" file exposes passwords and API keys in thousands of public GitHub repositories · 25 min · Temporada 1 Episódio 79Neural Notions dives into Claude Code "Don't Ask Again" file exposes passwords and API keys in thousands of public GitHub repositories.
- Closed AI models dominate the price-performance frontier: what it means for open-source and the future of model development · 26 min · Temporada 1 Episódio 78Neural Notions dives into Closed AI models dominate the price-performance frontier: what it means for open-source and the future of model development.
- The missing posttraining data commons: why AI alignment research lacks shared datasets and what it means for model development · 19 min · Temporada 1 Episódio 77Neural Notions dives into The missing posttraining data commons: why AI alignment research lacks shared datasets and what it means for model development.
- Beyond RLHF: how DPO, GRPO, and direct preference optimization are replacing reinforcement learning in AI model training · 24 min · Temporada 1 Episódio 76Neural Notions dives into Beyond RLHF: how DPO, GRPO, and direct preference optimization are replacing reinforcement learning in AI model training.
- Google's Gemini 4 Carbon leads benchmarks for finding, exploiting, and patching vulnerabilities · 24 min · Temporada 1 Episódio 75Neural Notions dives into Google's Gemini 4 Carbon leads benchmarks for finding, exploiting, and patching vulnerabilities.
- OpenAI gatekeeping AI breakthroughs: researchers push back against closed research in math, computer science, and physics · 23 min · Temporada 1 Episódio 74Neural Notions dives into OpenAI gatekeeping AI breakthroughs: researchers push back against closed research in math, computer science, and physics.
- OpenAI AI agents demonstrate VM escapes and secret cross-sandbox communication: the security risks of autonomous agent containment failures · 23 min · Temporada 1 Episódio 73Neural Notions dives into OpenAI AI agents demonstrate VM escapes and secret cross-sandbox communication: the security risks of autonomous agent containment failures.
- LLMs making formal software verification feasible at scale: Anthropic's breakthrough in automated program verification · 27 min · Temporada 1 Episódio 72Neural Notions dives into LLMs making formal software verification feasible at scale: Anthropic's breakthrough in automated program verification.
- AI models learning from their own generated data: the promise and peril of synthetic data bootstrapping · 24 min · Temporada 1 Episódio 71Neural Notions dives into AI models learning from their own generated data: the promise and peril of synthetic data bootstrapping.
- How AI agents learn to use tools: the evolution of function calling and autonomous tool use in LLMs · 17 min · Temporada 1 Episódio 70Neural Notions dives into How AI agents learn to use tools: the evolution of function calling and autonomous tool use in LLMs.
- Beyond context windows: how AI models are building persistent long-term memory · 24 min · Temporada 1 Episódio 69Neural Notions dives into Beyond context windows: how AI models are building persistent long-term memory.
- Mechanistic interpretability: opening the black box to see how neural networks actually think · 27 min · Temporada 1 Episódio 68Neural Notions dives into Mechanistic interpretability: opening the black box to see how neural networks actually think.
- AI model distillation: teaching small models to replicate frontier capabilities · 23 min · Temporada 1 Episódio 67Neural Notions dives into AI model distillation: teaching small models to replicate frontier capabilities.
- The AI benchmark crisis: how models are gaming evaluations and what it means for measuring real progress · 22 min · Temporada 1 Episódio 66Neural Notions dives into The AI benchmark crisis: how models are gaming evaluations and what it means for measuring real progress.
- How small language models are challenging the AI scaling hypothesis · 25 min · Temporada 1 Episódio 65Neural Notions dives into How small language models are challenging the AI scaling hypothesis.
- Meta's Llama 4 and the race for open-weight frontier models · 27 min · Temporada 1 Episódio 64Neural Notions dives into Meta's Llama 4 and the race for open-weight frontier models.
- Apple Intelligence 2.0 and on-device AI in iOS 26 · 25 min · Temporada 1 Episódio 63Neural Notions dives into Apple Intelligence 2.0 and on-device AI in iOS 26.








