WMLLM is a self‑evolving optimization‑agent framework that uses a predict‑then‑act world modeling approach. The agent first predicts promising optimization directions and then generates candidates, refining both its world model and strategy through multi‑turn refinement, population‑based search, and reinforcement learning. Experiments on black‑box tasks, particularly multi‑objective molecular optimization, demonstrate that WMLLM improves sample efficiency and achieves state‑of‑the‑art results within a limited evaluation budget.
arXiv:2606. 06473v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly applied to long-horizon tasks such as scientific discovery and machine learning engineering (MLE), where sustained self-evolution becomes a key capability.
arXiv:2606. 00008v1 Announce Type: new Abstract: Multi-objective molecular optimization requires searching vast chemical spaces under conflicting objectives, where early design decisions strongly constrain downstream outcomes.
arXiv:2605. 08756v2 Announce Type: replace Abstract: Automatic heuristic design (AHD) has emerged as a promising paradigm for solving NP-hard combinatorial optimization problems (COPs).
The paper introduces MCCE, a hybrid framework that combines a frozen closed‑source large language model (LLM) with a lightweight, trainable model for multi‑objective discrete optimization. By maintaining a trajectory memory and refining the small model through reinforcement learning, the two models jointly enhance global exploration and learning. Experiments on drug‑design benchmarks demonstrate that MCCE achieves state‑of‑the‑art Pareto front quality, outperforming existing baselines.
arXiv:2609.36683v1 Announce Type: new Abstract: Molecular optimization is inherently iterative: a candidate is proposed, evaluated against several objectives, and revised while preserving a relations...
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.
The paper introduces Mol-E, an evolutionary algorithm that leverages large language models trained on molecular data to generate candidate molecules. Mol-E achieves state‑of‑the‑art performance on the Practical Molecular Optimization benchmark, scoring 17.500 in the task‑agnostic regime and 20.551 in the task‑informed regime. It also outperforms baseline methods in multi‑property optimization tasks involving docking against DRD2, MK2, and AChE.
arXiv:2607. 19044v1 Announce Type: new Abstract: Leveraging large language models (LLMs) for molecular generation has shown remarkable potential in chemical and drug design.
arXiv:2607. 17281v1 Announce Type: cross Abstract: Auto-bidding plays an essential role in online advertising, automatically adjusting bids for advertisers to optimize their commercial goals.
arXiv:2606. 30639v1 Announce Type: new Abstract: World models offer a principled way to equip long-horizon LLM agents with foresight: predictions of action consequences before execution.
arXiv:2606. 02372v1 Announce Type: new Abstract: Equipping language agents with world models enables them to anticipate environment dynamics and evaluate candidate actions before execution.
CoMAP introduces a framework that jointly evolves textual world models and agent policies through a closed‑loop interaction. At each decision step the world model forecasts future state feedback for candidate actions, while the agent reflects on the reliability of this feedback to refine its action. The resulting on‑policy trajectories are used to self‑distill and update the world model, improving prediction accuracy and long‑horizon decision‑making across embodied planning, web navigation, and tool‑use benchmarks.