arXiv:2506. 07459v4 Announce Type: replace Abstract: Protein generative models have shown remarkable promise in protein design, yet their success rates remain constrained by reliance on curated sequence-structure datasets and by misalignment between supervised objectives and real design goals.
arXiv:2606. 18961v1 Announce Type: new Abstract: Protein language models (PLMs) have emerged as powerful tools for controllable biomolecular design, yet their post-training adaptation typically relies on costly wet-lab validation or curated preference datasets.
arXiv:2607. 19044v1 Announce Type: new Abstract: Leveraging large language models (LLMs) for molecular generation has shown remarkable potential in chemical and drug design.
AgentFold is a multi‑agent framework that treats protein‑folding model design as a closed‑loop search over executable code variants. Starting from the ESMFold codebase, the agents generate hypotheses, modify and debug code, evaluate model variants, and store both successes and failures in structured memory, guided by an MCTS‑style policy that allocates GPU resources. In an engineering‑scale experiment, AgentFold explored about 80 variants using 5,000 GPU‑hours and 170 million LLM tokens, improving the best lDDT score by 7.5% over independent Codex proposals and outperforming a random‑search baseline, while also uncovering empirical design patterns such as the benefits of early, soft, learnable priors.
Protein language models (PLMs) have emerged as powerful tools for controllable biomolecular design, yet their post-training adaptation typically relies on costly wet-lab validation or curated preference datasets. To overcome this supervision bottleneck, we introduce unsupervised reward optimization of PLMs, a comprehensive framework for steerable protein generation without ground-truth labels.
SymFold introduces a symmetric dual‑path architecture that combines protein language models (PLMs) and multimodal protein language models (MPLMs) to iteratively guide protein sequence generation for inverse folding. By leveraging pretrained sequence evolution knowledge from PLMs and structural knowledge from MPLMs, the method improves upon the traditional serial pipeline where structure encoders produce coarse sequences refined by PLMs. Experiments on standard inverse‑folding benchmarks show state‑of‑the‑art performance, and ablation studies confirm the effectiveness of the symmetric design.
The paper introduces MODS, a modular framework that performs multi‑objective protein optimization by steering diffusion models at inference time. MODS samples Chebyshev scalarization weights and a surrogate ensemble member to create a scalar steering reward, allowing the generation of sequences that balance multiple properties. After each batch of experimental evaluations, MODS updates its surrogates without retraining the generative model, and it achieves competitive hypervolume results across six protein optimization tasks, outperforming baselines on several.
Fraglingo is an autoregressive fragment-based molecular generator that jointly models fragment identity and attachment in a continuous latent space. It predicts attachment-aware fragment embeddings using a wildcard-anchored readout that captures the growing molecule’s active attachment site, then retrieves the next fragment via latent-space nearest-neighbor search. This approach allows new fragments to be added at inference time without retraining and achieves stronger joint property control on benchmarks while maintaining high validity, uniqueness, and novelty.
arXiv:2609.00189v1 Announce Type: new Abstract: Goal-directed optimization is essential for steering molecular generators to propose candidates with desired properties. However, it is often implement...
The paper introduces IDiom, an autoregressive protein language model trained on a large dataset of intrinsically disordered protein regions (IDRs) from AlphaFold, and demonstrates that it can generate sequences matching natural IDR composition, motifs, and disorder. It further presents RL‑SAE, a reinforcement learning approach that uses sparse autoencoder features to steer generation toward specific functional patterns, achieving high activation of targeted features and improved predicted subcellular localization and transcriptional activity. The combination of IDiom and RL‑SAE allows interpretable, composable IDR design by explicitly controlling function‑associated sequence features.
arXiv:2607. 02834v1 Announce Type: new Abstract: Molecular optimization often starts from a pretrained generative model that captures a broad prior over valid molecular structures.
MiMo‑V2.6 is an omni‑modal reinforcement learning framework that scales compute to advance large foundation models toward self‑improvement. It trains on a broad multimodal corpus before RL, then scales RL along three axes: larger batches and throughput, diverse environments across code, general, visual, and cyber domains, and enhanced grader compute for better reward signals. The system incorporates infrastructure such as a frozen MoE router, multi‑layer defense against reward hacking, unified trajectory representation, high‑concurrency rollout, and training‑inference consistency, and it is open‑source to support reproducibility and further research.
The paper introduces MCTH (Monte Carlo Tree Hallucination), an inference-only framework that performs all‑atom biomolecular sequence‑structure co‑design by treating pretrained folding and inverse‑folding models as black‑box operators. MCTH uses Monte Carlo Tree Search to allocate a fixed inference budget across competing design trajectories, incorporating model confidence, uncertainty, and cross‑expert consensus. Experiments across protein‑RNA, protein‑DNA, protein‑protein, and protein‑ligand design show that adaptive search outperforms simpler sampling strategies, and evaluations with AlphaFold3 and Chai‑1 demonstrate transferability beyond the search‑time oracle.