All Articles

Browse the full archive, newest first.

Tech

Briefing: HyEvo: Self-Evolving Hybrid Agentic Workflows for Efficient Reasoning

Strategic angle: Exploring the potential of self-evolving workflows to enhance reasoning efficiency in complex tasks.

Editorial Staff Mar 23

Tech

Briefing: ItinBench: Benchmarking Planning Across Multiple Cognitive Dimensions with Large Language Models

Strategic angle: Exploring the capabilities of large language models in reasoning and planning tasks.

Editorial Staff Mar 23

Tech

Briefing: Learning to Disprove: Formal Counterexample Generation with Large Language Models

Strategic angle: Exploring the capabilities of AI in generating counterexamples for mathematical reasoning.

Editorial Staff Mar 23

Tech

Briefing: PA2D-MORL: A New Approach to Multi-Objective Reinforcement Learning

Strategic angle: Introducing Pareto Ascent Directional Decomposition for enhanced decision-making in conflicting objectives.

Editorial Staff Mar 23

Tech

Briefing: PowerLens: Taming LLM Agents for Safe and Personalized Mobile Power Management

Strategic angle: A new approach to enhance battery life in mobile devices using LLM agents.

Editorial Staff Mar 23

Tech

Briefing: Stepwise: Neuro-Symbolic Proof Search for Automated Systems Verification

Strategic angle: A new approach to formal verification using interactive theorem proving.

Editorial Staff Mar 23

Tech

Briefing: Teaching an Agent to Sketch One Part at a Time

Strategic angle: A novel approach to training AI for vector sketching through multi-turn reinforcement learning.

Editorial Staff Mar 23

Life

Briefing: The President’s Malaria Initiative and Other U.S. Government Global Malaria Efforts

Strategic angle: This fact sheet provides a snapshot of global malaria efforts and examines the U.S. government’s role in addressing malaria worldwide.

Editorial Staff Mar 23

Tech

Briefing: When both Grounding and not Grounding are Bad -- A Partially Grounded Encoding of Planning into SAT (Extended Version)

Strategic angle: Exploring the complexities of classical planning problems and their representations.

Editorial Staff Mar 23

Tech

Continually Self-Improving AI

Exploring the limitations and potential of modern language model-based AI systems.

Editorial Staff Mar 23