paper-with-me

Papers

Q-Steer: Action-Value Guidance for Molecular Policy Optimization

2026-07-29 · Xinyu Wang, Jinbo Bi, Minghu Song arxiv

Oracle-limited molecular optimization gives reward only after a complete molecule is generated, while each rollout requires many local next-token decisions. This delayed-feedback interface makes molecular policy optimization myopic: an optimizer can learn that a molecule was good without knowing which intermediate actions made it good. We introduce Q-Steer, a rollout-time action-value steering primitive for molecular language models. Q-Steer uses an offline-trained and frozen prefix-action value scorer, PAVS-Q, that estimates the downstream reward of taking a candidate next token under a partial SMILES prefix, then adds a normalized value bonus to sampling logits. The optimizer update rule and online oracle budget are unchanged; the claim is fixed-online-oracle performance, not equal total compute. On PMO23 with a fixed 10,000-call online budget, complete factorial studies across two molecular language-model backbones and four optimizers show that Q-Steer improves mean valid-unique score in all eight backbone-optimizer cells, with positive macro mean-score gains between +0.033 and +0.049 and 18-20 task wins per cell. Mechanism controls show that action identity matters: prefix-broadcast values are nearly neutral, while shuffled action values harm performance. These results support Q-Steer as a reusable rollout-time action-value wrapper that improves average molecular optimization reward across optimizer families and policy backbones without changing the online oracle budget.

📄 PDF Abstract BibTeX arXiv:2607.26391

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Scalable and Cost-Efficient de Novo Template-Based Molecular Generation

2025-06-10 · Piotr Gaiński, Oussama Boussif, Andrei Rekesh, Dmytro Shevchuk 외

Template-based molecular generation offers a promising avenue for drug design by ensuring generated compounds are synthetically accessible through predefined reaction templates and building blocks. In this work, we tackl…

DiversityDrug Design

EmbodiSteer: Steering Embodiment-Agnostic Visuomotor Policies with Joint-Space Guidance for Zero-Shot Cross-Embodiment Deployment

2026-06-11 · Shihefeng Wang, Kangchen Lv, Mingrui Yu, Xiang Li arxiv

Scalable robot imitation learning relies on large-scale heterogeneous data from diverse robots or body-free data, making Cartesian end-effector actions a key interface for embodiment-agnostic policy learning. However, en…

Collision Avoidance

Sample-Efficient Diffusion-based Reinforcement Learning with Critic Guidance

2026-05-28 · Shutong Ding, Zejia Zhong, Zhongyi Wang, Ke Hu 외 arxiv

Recent advances in reinforcement learning (RL) have achieved great successes by leveraging the multimodality and exploration capability of diffusion policies. Among these approaches, one representative branch focuses on …

Reinforcement Learning

Advantage-Guided Diffusion for Model-Based Reinforcement Learning

2026-04-10 · Daniele Foffano, Arvid Eriksson, David Broman, Karl H. Johansson 외 arxiv

Model-based reinforcement learning (MBRL) with autoregressive world models suffers from compounding errors, whereas diffusion world models mitigate this by generating trajectory segments jointly. However, existing diffus…

Reinforcement Learning

DREAMSTEER: Latent World Models Can Steer VLA Policies During Deployment Without Any Finetuning

2026-07-03 · Hanchen Cui, Sergio Arnaud, Arjun Majumdar, Daniel Dugas 외 arxiv

Pretrained vision-language-action (VLA) policies show promising zero-shot generalization, but often fail under deployment-time distribution shift, leading to decreased robustness and inconsistent instruction following. W…

Zero-shot GeneralizationInstruction Following