arXiv Machine Learning June 26, 2026 By Mohammad Haddadnia, Yuvan Chali, Abhilash Jayaraj, Constance Kraay, Joana Reis, Felix Strieth-Kalthoff, Haribabu Arthanari

Target-Aware Bandit Allocation for Scalable Surrogate Optimization in Chemical Space

Read the original on arXiv Machine Learning →

arXiv:2606. 26657v1 Announce Type: new Abstract: Identifying high-utility candidates from massive discrete spaces under expensive evaluations is a recurring challenge across the sciences, with structure-based drug discovery as a prominent example.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

reinforcement-learning

arXiv Machine Learning

Jul 15

Sample Efficient Generative Optimization for Molecular Design

arXiv:2607. 12488v1 Announce Type: new Abstract: Molecular optimization in drug discovery, materials design, and catalysis requires searching vast chemical spaces under tight evaluation budgets, since high-fidelity oracles and experimental measurements are costly.

By Sarina Kopf, Cristina Nevado, Philippe Schwaller

benchmarks

arXiv AI

Jun 18

Bayesian Anytime Pareto Set Identification for Multi-Objective Multi-Armed Bandits

arXiv:2606. 18785v1 Announce Type: cross Abstract: Identifying Pareto optimal solutions is critical to support multi-objective decision-making.

By Lennert Saerens, Bram Silue, Eleni Litsa, Peter Vrancx, Pieter Libin

reinforcement-learning benchmarks

arXiv Machine Learning

Sep 7

Training Large Language Models for Small-Molecule Design with Synthetic Task Scaling

The paper explores how large language models (LLMs) can be trained for small-molecule drug design by using synthetic tasks that are cheaper to evaluate. By employing a curriculum that gradually increases task difficulty, the authors demonstrate that LLMs can learn design strategies that outperform larger models on structure-based lead optimization. This approach shows that scaling post‑training with synthetic tasks can effectively adapt LLMs to high‑cost experimental scenarios that are otherwise infeasible to train on directly.

By Frank Hu, Shriram Chennakesavalu, Zichen Wang, Patricia Suriana, Bodhi Vani, Kirill Shmilovich, Kangway Chuang, Colin Grambow

llms reinforcement-learning

arXiv AI

Sep 16

LLM-Guided Dynamic Action Spaces for Synthesizable Molecular Optimization

arXiv:2604.07669v3 Announce Type: replace-cross Abstract: Synthesizable molecular optimization seeks to improve target properties while ensuring that molecular modifications follow feasible synthetic...

By Tao Li, Kaiyuan Hou, Tuan Vinh, Fanglei Xue, Monika Raj, Zhichun Guo, Carl Yang

llms

arXiv Machine Learning

Aug 18

Large Discovery Models: Empirically-grounded Model-Based Open-Ended Search

arXiv:2608. 15669v1 Announce Type: new Abstract: Scientific discovery often involves optimising expensive-to-evaluate objectives over vast, structured, and open-ended hypothesis spaces, such as molecules, protein sequences, and computer programs.

By Zhongwei Yu, Yan Song, Xue Yan, Anjie Liu, Xingyu Lu, Yihang Chen, Huichi Zhou, Siyuan Guo, Luoyang Sun, Sihan Chen, Xiangning Yu, Jun Wang

llms

arXiv Computation and Language

Aug 25

Closed-Loop Bayesian Molecular Inverse Design with Semantic LLM Surrogates

arXiv:2608.22967v1 Announce Type: new Abstract: Practical molecular inverse design is rarely a one-shot generation problem; it often takes the form of closed-loop candidate-pool enrichment, where und...

By Yaoyao Xu, Xinjian Zhao, Xiaozhuang Song, Lei Bai, Tianshu Yu

llms rag

One email a morning, machine-written

Related stories

Sample Efficient Generative Optimization for Molecular Design

Bayesian Anytime Pareto Set Identification for Multi-Objective Multi-Armed Bandits

Training Large Language Models for Small-Molecule Design with Synthetic Task Scaling

LLM-Guided Dynamic Action Spaces for Synthesizable Molecular Optimization

Large Discovery Models: Empirically-grounded Model-Based Open-Ended Search

Closed-Loop Bayesian Molecular Inverse Design with Semantic LLM Surrogates