Search:
Match:
2 results
Paper#llm🔬 ResearchAnalyzed: Jan 3, 2026 18:22

Unsupervised Discovery of Reasoning Behaviors in LLMs

Published:Dec 30, 2025 05:09
1 min read
ArXiv

Analysis

This paper introduces an unsupervised method (RISE) to analyze and control reasoning behaviors in large language models (LLMs). It moves beyond human-defined concepts by using sparse auto-encoders to discover interpretable reasoning vectors within the activation space. The ability to identify and manipulate these vectors allows for controlling specific reasoning behaviors, such as reflection and confidence, without retraining the model. This is significant because it provides a new approach to understanding and influencing the internal reasoning processes of LLMs, potentially leading to more controllable and reliable AI systems.
Reference

Targeted interventions on SAE-derived vectors can controllably amplify or suppress specific reasoning behaviors, altering inference trajectories without retraining.

Research#AI Algorithms📝 BlogAnalyzed: Dec 29, 2025 08:26

Masked Autoregressive Flow for Density Estimation with George Papamakarios - TWiML Talk #145

Published:May 28, 2018 19:20
1 min read
Practical AI

Analysis

This article summarizes a podcast episode discussing George Papamakarios's research on Masked Autoregressive Flow (MAF) for density estimation. The episode explores how MAF utilizes neural networks to estimate probability densities from input data. It touches upon related research like Inverse Autoregressive Flow, Real NVP, and Masked Auto-encoders, highlighting the foundational work that contributed to MAF. The discussion also covers the characteristics of probability density networks and the difficulties encountered in this area of research. The article provides a concise overview of the podcast's content, focusing on the technical aspects of MAF and its context within the field of density estimation.
Reference

George walks us through the idea of Masked Autoregressive Flow, which uses neural networks to produce estimates of probability densities from a set of input examples.