Artificial Intelligence

Download Report

Frontier AI research — LLM reasoning, agentic systems, interpretability, algorithm discovery

13Sources
12Concepts
5Entities
OverviewConceptsEntitiesSourcesPractice

Agent Evaluation Benchmarks

Steady Progress
evaluationbenchmarksassessment

Agent Memory Architectures

Active Frontier
memoryagent-infrastructureretrieval-augmented+1

Agent Safety & Alignment

Active Frontier
safetyalignmentred-teaming+1

Agentic Reasoning

Active Frontier
paradigmagentsreasoning

Chain-of-Thought Reasoning

Active Frontier
reasoningchain-of-thoughtmonitoring

Circuit Tracing

Active Frontier
interpretabilitycircuitsfeatures+1

Evolutionary Algorithm Discovery

Active Frontier
evolutionary-algorithmscode-generationalgorithm-discovery+1

LLM Tool Use

Active Frontier
tool-useagentsparadigms

Mechanistic Interpretability

Active Frontier
interpretabilitysafetytransparency

Multi-Agent Systems

Active Frontier
multi-agentcollaborationprotocols

Reinforcement Learning for Agents

Active Frontier
reinforcement-learningpolicy-learningoptimization

Vision-Language-Action Models

Active Frontier
multimodalrobotics-aiembodied-ai+1
Artificial Intelligence | KB | MenFem