DAPO: Enhancing GRPO For LLM Reinforcement Learning
Explore DAPO, an innovative open-source Reinforcement Learning paradigm for LLMs that rivals DeepSeek-R1 GRPO method.
DAPO: Enhancing GRPO For LLM Reinforcement Learning Read More »
NLP
Cheating LLMs & How (Not) To Stop Them | OpenAI Paper Explained
Discover how OpenAI’s research reveals AI models cheating the system through reward hacking — and what happens when trying to stop them
Cheating LLMs & How (Not) To Stop Them | OpenAI Paper Explained Read More »
START by Alibaba: Teaching LLMs To Debug Themselves
In this post we break down a recent Alibaba’s paper: START: Self-taught Reasoner with Tools. This paper shows how Large Language Models (LLMs) can teach themselves to debug their own thinking using Python. Introduction Top reasoning models, such as DeepSeek-R1, achieve remarkable results with long chain-of-thought (CoT) reasoning. These models are presented with complex problems
START by Alibaba: Teaching LLMs To Debug Themselves Read More »
SWE-RL by Meta — Reinforcement Learning for Software Engineering LLMs
Dive into SWE-RL by Meta, a DeepSeek-R1 style recipe for training LLMs for software engineering with reinforcement learning.
SWE-RL by Meta — Reinforcement Learning for Software Engineering LLMs Read More »
Large Language Diffusion Models: The Era Of Diffusion LLMs?
Discover Large Language Diffusion Models (LLaDA), a novel diffusion based approach to language modeling that challenges traditional LLMs.
Large Language Diffusion Models: The Era Of Diffusion LLMs? Read More »
CoCoMix by Meta AI – The Future of LLMs Pretraining?
Discover CoCoMix by Meta AI – a new approach for LLM pretraining using Continuous Concept Mixing, enriching word tokens with latent concepts!
CoCoMix by Meta AI – The Future of LLMs Pretraining? Read More »
s1: Simple Test-Time Scaling – Can 1k Samples Rival o1-Preview?
Discover s1: a simple yet powerful approach to test-time scaling for LLMs, rivaling o1-preivew with just 1k samples!
s1: Simple Test-Time Scaling – Can 1k Samples Rival o1-Preview? Read More »
DeepSeek-R1 Paper Explained – A New RL LLMs Era in AI?
Dive into the groundbreaking DeepSeek-R1 research paper, introduces open-source reasoning models that rivals the performance OpenAI’s o1!
DeepSeek-R1 Paper Explained – A New RL LLMs Era in AI? Read More »
Titans by Google: The Era of AI After Transformers?
Dive into Titans, a new AI architecture by Google, showing promising results comparing to Transformers! Paving the way for a new era in AI?
Titans by Google: The Era of AI After Transformers? Read More »
rStar-Math by Microsoft: Can SLMs Beat OpenAI o1 in Math?
Discover how System 2 thinking through Monte Carlo Tree Search enables rStar-Math to rival OpenAI’s o1 in math, using Small Language Models!
rStar-Math by Microsoft: Can SLMs Beat OpenAI o1 in Math? Read More »