Evo-B&B is a new evolutionary framework that jointly evolves branching and node selection policies for Mixed-Integer Linear Programming (MILP) using large language model guidance and end-to-end SCIP solving feedback. It introduces a behavior-aware multi-island mechanism that measures policy similarity based on branching and node-selection behaviors to preserve population diversity and reduce redundant evaluations. Experiments on four standard MILP benchmarks show that Evo-B&B achieves competitive solving performance, generalizes to larger instances, and that joint evolution and behavior-aware diversity management are effective.
arXiv:2609.36735v1 Announce Type: new Abstract: In branch-and-bound (B&B) for mixed-integer linear programming (MILP), branching variable selection critically impacts efficiency. Existing neural bran...
The paper introduces HoliDivEvolve, a holistic strategy for maintaining diversity in LLM-guided evolutionary search. It employs diverse initialization, trajectory‑adaptive island selection, and growth‑adaptive migration to ensure distinct search directions and effective knowledge sharing across islands. Experiments on varied benchmarks demonstrate competitive performance and provide controlled analyses of the design choices.
The paper introduces RePPO, a method that adapts Proximal Policy Optimization to jointly optimize a compact population of diverse, high‑quality policies using a population‑dependent, dynamically shaped reward. RePPO achieves a strong quality‑diversity tradeoff on standard continuous‑control tasks and scales to larger policy and behavior spaces, maintaining quality while significantly increasing diversity in pixel‑based tasks. The authors also demonstrate RePPO’s applicability to fine‑tuning large language models, showing for the first time that quality‑diversity reinforcement learning can scale to this domain.
arXiv:2607. 18252v1 Announce Type: new Abstract: Machine learning methods have shown that data-driven policies can accelerate mixed-integer linear programming (MILP) solvers, but many such approaches remain difficult to inspect, adapt, and deploy because the learned policy is represented as an external predictor or other opaque model.
TurboEvolve is a multi‑island evolutionary framework that enhances LLM‑driven program evolution by improving sample efficiency and stability within fixed API‑cost budgets. It introduces Verbalized Sampling to generate multiple diverse candidates in a single LLM call, an adaptive scheduler that adjusts the number of offspring based on progress, and island‑aware initialization that clusters seed programs and mixes subpopulations with elitist preservation. The authors also provide a reusable cross‑task solution‑pool dataset and demonstrate that TurboEvolve achieves better cost‑quality trade‑offs and more stable performance on mathematical discovery and GPU‑kernel optimization tasks compared to controlled baselines.
arXiv:2605. 30789v2 Announce Type: replace-cross Abstract: We identify a new dimension for enhancing rollout diversity in Group Relative Policy Optimization (GRPO) for LLMs.
arXiv:2606. 15197v1 Announce Type: cross Abstract: Optimization modeling is inherently hierarchical, requiring a precise sequence of symbolic commitments.
arXiv:2608. 10694v1 Announce Type: cross Abstract: Evolutionary optimization of LLM prompts and agentic programs (e.
Agentic ESOpt proposes using evolution strategies (ES) instead of reinforcement learning to fine‑tune large language‑model agents for long‑horizon tasks. ES offers model scalability, flexibility, and better long‑horizon credit assignment, enabling full‑parameter optimization with minimal GPU memory. The framework samples parameter perturbations, evaluates agents with rewards, and updates online, achieving notable performance gains on WebArena‑Lite and in test‑time prompt‑parameter co‑evolution.
Black-box combinatorial optimization requires systematically identifying high-quality solutions under a limited evaluation budget, yet the unknown objective function provides little guidance for deciding where the search should explore next. We introduce SCOPE, a general framework for Synthetic Conditional Objectives for Policy Evolution in Black-Box Combinatorial Optimization.
arXiv:2610.09943v1 Announce Type: cross Abstract: Reinforcement learning (RL) fine-tuning improves vision-language-action (VLA) policies through closed-loop experience, yet generalization beyond the...
Collab‑Solver introduces a multi‑agent policy learning framework for mixed‑integer linear programming (MILP) that enables collaborative optimization of multiple solver modules. By modeling the interaction between cut selection and branching as a Stackelberg game, the approach employs a two‑phase learning paradigm—data‑communicated policy pretraining followed by coordinated policy refinement. Experiments on synthetic and large‑scale real‑world MILP datasets show that the jointly learned policies markedly improve solving performance and generalize well across diverse instance sets.