Sep 30, 2026
Mixture of Self-Improving Branches For Agent Harness Optimization
Mixture of Self-Improving Branches For Agent Harness Optimization Agent performance depends not only on the underlying language model, but also on the softwa...
Explore Franklin AI
AI Research Papers
Understand new AI research with readable explanations and links to the original papers.
Sep 30, 2026
Probability is Not Enough: Exploring and Counting Divergent Tokens for Reasoning Uncertainty Quantification in LLMs
What the paper is about As the chain-of-thought reasoning capabilities of large language models improve, evaluating and calibrating their reasoning confidenc...
Sep 30, 2026
Which Attention Heads are like the Human Head? Not the Ones that Compute
Which Attention Heads are like the Human Head? Not the Ones that Compute This paper asks whether attention heads in large language models that resemble human...
Sep 30, 2026
Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI
What the paper is about Agent performance depends on both reasoning ability and the environment in which it acts.
Sep 30, 2026
DIET: Deletion-response Expert Trimming for Video Diffusion Transformers
DIET: Deletion-response Expert Trimming for Video Diffusion Transformers Mixture-of-experts (MoE) video models reduce computation by activating only a small...
Sep 30, 2026
Topological Coherence for Self-evolving Multi-agent Systems
Topological Coherence for Self-Evolving Multi-Agent Systems Multi-agent systems can divide a difficult task among specialized agents, but coordination become...
Sep 30, 2026
Stochastic World Models for Verifying Vision-Based Neural Feedback Systems
What the paper is about Verifying a vision-based neural feedback system requires a model of the observations its controller acts upon.
Sep 30, 2026
SelfSearch: Reward-Free Search for Self-Improving Agents
SelfSearch: Reward-Free Search for Self-Improving Agents investigates whether an AI agent can improve its own implementation without repeatedly testing candi...
Sep 30, 2026
AdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation
What the paper is about A small trainable advisor can steer a frozen language-model executor using natural-language advice.
Sep 30, 2026
BrainNet Studio: A Unified Toolkit for Brain Network Construction, Intelligent Analysis, and Visualization
BrainNet Studio: A Unified Toolkit for Brain Network Construction, Intelligent Analysis, and Visualization BrainNet Studio is a software toolkit for analyzin...
Good to know
How do I use these AI research explainers?
Start with the overview to understand the paper’s main idea, then follow its original paper link for the methods, results and limitations.
Can I suggest a research paper?
Use Submit Research to send the original paper URL and explain why it is useful. An editor reviews the suggestion before publication.