A knowledge-first approach to LLM-driven automatic heuristic design in combinatorial optimization yields better discovery efficiency, transfer, and generalization than code-centric baselines by formalizing a distortion-compression trade-off.
Deepaco: Neural-enhanced ant systems for combinatorial optimization.Advances in neural information processing systems, 36:43706–43728
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
other 1
citation-polarity summary
fields
cs.AI 2years
2026 2verdicts
UNVERDICTED 2roles
other 1polarities
unclear 1representative citing papers
The CARM module boosts neural routing solvers by adaptively modulating embeddings with constraint variables, enabling better use of global observations and improved performance on constrained VRPs.
citing papers explorer
-
Back to the Beginning of Heuristic Design: Bridging Code and Knowledge with LLMs
A knowledge-first approach to LLM-driven automatic heuristic design in combinatorial optimization yields better discovery efficiency, transfer, and generalization than code-centric baselines by formalizing a distortion-compression trade-off.
-
Rethinking Constraint Awareness for Efficient State Embedding of Neural Routing Solver
The CARM module boosts neural routing solvers by adaptively modulating embeddings with constraint variables, enabling better use of global observations and improved performance on constrained VRPs.