SwarmEvolve uses latent space agent swarms for faster program optimization
SwarmEvolve reaches best solutions in 3-8 minutes via latent-space KV cache sharing instead of 60 minutes.
Stories per day · 30 days
2 RSS
Relatedagent-swarms1autonomous-driving1circuit-design1generative-retrieval1huggingface1hyperparameter-optimization1information-retrieval1latent-space1llm1modular-inversion1neural-architecture-search1open-weight1quantum-computing1reinforcement-learning1semantic-identifiers1toffoli1
10 stories, newest first · one card per story
Stories tagged #evolutionary-search
HPNAS-RL jointly searches RL architectures and hyperparameters for driving
In a driving simulator the joint search beat standalone NAS, producing higher-reward policies with fewer parameters.
LLM evolutionary search improves generative retrieval semantic indexes
On NQ320K, jointly evolving SID construction and collision handling lifts Recall@1 from 55.8% to 59.3%.
Discovering state-of-the-art quantum circuits with large language models
LLM-guided evolutionary search finds quantum circuits that cut Toffoli and qubit costs on seven of eight tasks.
FrugalEvo: Towards Cost-Aware LLM-Guided Program Evolution
FrugalEvo pairs a costly strategy model with cheap code refinement and beats evolutionary baselines at far lower spend.
EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery
EvoDuet co-evolves web queries and solutions, raising LLM discovery scores on 21 scientific tasks.
EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery
EvoDuet co-evolves candidate solutions and web-search queries to improve LLM scientific discovery.
MILO: Automated Harness Discovery via Orchestrated Multi-Agent Evolution
MILO co-evolves agent harnesses and search strategy, beating prior harnesses on Terminal-Bench, PaperBench, and DeepSWE.
Guiding Worker Self-Selection in Crowdsourcing Contests: An LLM-Augmented Algorithmic Approach
Researchers introduce GRAF, a greedy framework for crowdsourcing contest self-selection, and LLMScore, an LLM-driven method that auto-designs its scoring algorithm.
LLM-Driven Algorithm Design for Quantum Circuit Synthesis based on Binary Decision Diagrams
QuantumEvo uses an LLM to evolve BDD variable-ordering heuristics, achieving a 70.9% tie-or-win rate on quantum circuit cost versus baseline methods.