Single-Agent vs Multi-Agent Systems Design

Core architectural choice in the paper: single agents determine actions dynamically based on context, while multi-agents follow pre-defined workflows with specialized roles.

Neighborhood — ranked by edge-count

Frameworks (1)

framework

SFR-DeepResearch
about
The paper's core contribution: an RL-based framework for training autonomous single-agent LLMs to perform deep research with web search, browsing, and code execution.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

Multiscale Agent-Based Modelmethod0.742
Who Is The Creator When Multiple Agents Contributequestion0.738
Single agents can generalize better to unseen tasks because they are not constrained by predefined heuristic-based workflows.claim0.732
Architectural belief motivating single-agent design choice; suggests flexibility provides better out-of-distribution performance.
Multi-Agent Reflectionconcept0.731
Approach using multiple LLM agents for generation and critique, a key prior approach to improving reflection.
Evolved vs. designed origin is not a definitive guide to agency; rationally engineered systems can be true cognitive agents.claim0.723
Rejects the notion that only evolutionary history confers intrinsic motivation or 'true' cognition.
Agent-based computational modelmethod0.721
The computational approach used to simulate morphogenesis with cells as agents on a 2D grid; allows quantitative testing of stress-sharing hypothesis.
A naive agent equipped with reduced priors from an experienced agent performs perfectly with maximum confidence from the first trial.finding0.719
Demonstration that model-level priors (not parameter-level knowledge) suffice for immediate transfer
The boundary between evolved and designed agents is becoming blurred through biorobotics and synthetic morphology.claim0.718