Podcast Episodes
Back to Search
SuperThoughts: Reasoning Tokens in Superposition
SuperThoughts is a novel framework designed to accelerate the Chain-of-Thought (CoT) reasoning process in large language models by processing tokens …
2 months, 1 week ago
First-Explore PPO : Learning Meta-Exploration with Proximal Policy Optimization
This research paper introduces First-Explore Proximal Policy Optimization (FE-PPO), a new reinforcement learning algorithm designed to improve how ag…
2 months, 2 weeks ago
Self-Distillation for Data-Scarce Language Model Pretraining
This research paper investigates self-distillation as a powerful regularization technique for pretraining language models when high-quality data is i…
2 months, 2 weeks ago
Meta-Harness for Agent-State Construction
eta-Harness is an advanced optimization system designed to improve how language-model agents process and compress long interaction histories into use…
2 months, 2 weeks ago
ExpRL: Using Reference Solutions as Rewards for LLM Mid-Training
Exploratory RL (ExpRL) is an automated mid-training method designed to enhance the reasoning capabilities of large language models before they underg…
2 months, 2 weeks ago
Valid Inference with Synthetic Data via Task Exchangeability
This paper introduces a statistical framework for making valid scientific discoveries using synthetic data, specifically addressing concerns that art…
2 months, 2 weeks ago
GRPO is Secretly a Process Reward Model
This paper establishs that Group Relative Policy Optimization (GRPO), while appearing to use only final outcome rewards, inherently functions as a Pr…
2 months, 3 weeks ago
Agentic Interactions
This paper explores how AI agents inherit and potentially amplify human heterogeneity when tasked with negotiating on behalf of individuals. By compa…
2 months, 3 weeks ago
A Unifying View of Attention Sinks: Two Algorithms, Two Solutions
This research investigates the nature of attention sinks, which are specific tokens in Transformer models that attract disproportionate attention. Th…
2 months, 3 weeks ago
From AGI to ASI
This report from Google DeepMind explores the hypothetical transition from Artificial General Intelligence (AGI), which matches human capability, to …
2 months, 3 weeks ago