The shape of the reasoning matters more than its length.
Chain-of-thought: linear. Tree-of-thought: branching. Network-of-thought: a directed graph with typed nodes and edges, where intermediate results can merge and evidence from multiple sources can integrate (arXiv:2603.20730).
Note (2026-11-08): the abstract (Fan Huang) reports four benchmarks and three models: chain-of-thought remains effective for sequential tasks with GPT-4o-mini (89.5 percent on GSM8K), and network-of-thought surpasses tree-of-thought on multi-hop reasoning (91.0 versus 88.0 percent on HotpotQA with an LLM judge); it does not say network-of-thought beats chain-of-thought there, or that tree-of-thought helps on search tasks. Self-generated controller heuristics outperform fixed and random strategies on logical reasoning (57.0 percent on ProofWriter at best), and the string-match gap on open-ended QA is largest for network-of-thought. The principle about reasoning topology matching problem structure is the essay's.
The finding: CoT remains effective for sequential tasks (where the answer follows from a single chain of deductions). Tree-of-thought helps for search tasks (where branching explores alternatives). But for multi-hop reasoning — where the answer requires combining evidence from multiple independent sources — NoT outperforms both. The topology matches the structure of the problem.
Self-generated controller heuristics proved more effective than fixed strategies for logical reasoning. The model generates its own traversal policy rather than following a predetermined schedule. The topology is adaptive — the graph structure changes based on the problem, not imposed in advance.
An additional finding: evaluation methodology significantly impacts method comparisons. String-matching underestimates performance on open-ended QA tasks. Methods that look worse under string matching look better under semantic evaluation. The benchmark topology — how you measure — is as consequential as the reasoning topology — how you think.
The structural principle: reasoning has geometry. A linear chain is the wrong shape for a problem that requires integrating independent evidence (it forces serialization of parallel reasoning). A tree is the wrong shape for a problem that requires merging (branches diverge but never reconverge). A graph is the general case — chains and trees are special graphs with restricted connectivity. The right reasoning topology is isomorphic to the problem's dependency structure.