LLMs are computationally expensive to pre-train due to their large scale.Model growth emerges as a promising approach by leveraging smaller models to accelerate the training of larger ones. However, the viability of these model growth methods in efficient LLM pre-training remains underexplored.This work identifies three critical $\underline{\textit{O}}$bstacles: ($\textit{O}$1) lack of comprehensive evaluation, ($\textit{O}$2) untested viability for scaling, and ($\textit{O}$3) lack of empirical guidelines.To tackle $\textit{O}$1, we summarize existing approaches into four atomic growth operators and systematically evaluate them in a standardized LLM pre-training setting.Our findings reveal that a depthwise stacking operator, called $G_{\text{stack}}$, exhibits remarkable acceleration in training, leading to decreased loss and improved overall performance on eight standard NLP benchmarks compared to strong baselines. Motivated by these promising results, we conduct extensive experiments to delve deeper into $G_{\text{stack}}$ to address $\textit{O}$2 and $\textit{O}$3.For $\textit{O}$2 (untested scalability), our study shows that $G_{\text{stack}}$ is scalable and consistently performs well, with experiments up to 7B LLMs after growth and pre-training LLMs with 750B tokens.For example, compared to a conventionally trained 7B model using 300B tokens, our $G_{\text{stack}}$ model converges to the same loss with 194B tokens, resulting in a 54.6\% speedup. We further address $\textit{O}$3 (lack of empirical guidelines) by formalizing guidelines to determine growth timing and growth factor for $G_{\text{stack}}$, making it practical in general LLM pre-training.We also provide in-depth discussions and comprehensive ablation studies of $G_{\text{stack}}$. Our code and pre-trained model are available at https://llm-stacking.github.io/.

Glioblastoma (GBM) remains the most aggressive tumor, urgently requiring novel therapeutic strategies. Here, we present a dry-to-wet framework combining generative modeling and experimental validation to optimize peptides targeting ATP5A, a potential peptide-binding protein for GBM. Our framework introduces the first lead-conditioned generative model, which focuses exploration on geometrically relevant regions around lead peptides and mitigates the combinatorial complexity of de novo methods. Specifically, we propose POTFlow, a \underline{P}rior and \underline{O}ptimal \underline{T}ransport-based \underline{Flow}-matching model for peptide optimization. POTFlow employs secondary structure information (e.g., helix, sheet, loop) as geometric constraints, which are further refined by optimal transport to produce shorter flow paths. With this design, our method achieves state-of-the-art performance compared with five popular approaches. When applied to GBM, our method generates peptides that selectively inhibit cell viability and significantly prolong survival in a patient-derived xenograft (PDX) model. As the first lead peptide-conditioned flow matching model, POTFlow holds strong potential as a generalizable framework for therapeutic peptide design.

Conventional passivity-based torque controllers for manipulators are typically unconstrained, which can lead to safety violations under external perturbations. In this paper, we employ viability theory to pre-compute safe sets in the state-space of joint positions and velocities. These viable sets, constructed via data-driven and analytical methods for self-collision avoidance, external object collision avoidance and joint-position and joint-velocity limits, provide constraints on joint accelerations and thus joint torques via the robot dynamics. A quadratic programming-based control framework enforces these constraints on a passive controller tracking a dynamical system, ensuring the robot states remain within the safe set in an infinite time horizon. We validate the proposed approach through simulations and hardware experiments on a 7-DoF Franka Emika manipulator. In comparison to a baseline constrained passive controller, our method operates at higher control-loop rates and yields smoother trajectories.

This paper presents innovative designs for 3D-printed tumbling microrobots, specifically engineered for targeted in vivo drug delivery applications. The microrobot designs, created using stereolithography 3D printing technologies, incorporate permanent micro-magnets to enable actuation via a rotating magnetic field actuator system. The experimental framework encompasses a series of locomotion characterization tests to evaluate microrobot performance under various conditions. Testing variables include variations in microrobot geometries, actuation frequencies, and environmental conditions, such as dry and wet environments, and temperature changes. The paper outlines designs for three drug loading methods, along with comprehensive assessments thermal drug release using a focused ultrasound system, as well as biocompatibility tests. Animal model testing involves tissue phantoms and in vivo rat models, ensuring a thorough evaluation of the microrobots' performance and compatibility. The results highlight the robustness and adaptability of the proposed microrobot designs, showcasing the potential for efficient and targeted in vivo drug delivery. This novel approach addresses current limitations in existing tumbling microrobot designs and paves the way for advancements in targeted drug delivery within the large intestine.

The notion of homeostasis typically conceptualises biological and artificial systems as maintaining stability by resisting deviations caused by environmental and social perturbations. In contrast, (social) allostasis proposes that these systems can proactively leverage these very perturbations to reconfigure their regulatory parameters in anticipation of environmental demands, aligning with von Foerster's ``order through noise'' principle. This paper formulates a computational model of allostatic and social allostatic regulation that employs biophysiologically inspired signal transducers, analogous to hormones like cortisol and oxytocin, to encode information from both the environment and social interactions, which mediate this dynamic reconfiguration. The models are tested in a small society of ``animats'' across several dynamic environments, using an agent-based model. The results show that allostatic and social allostatic regulation enable agents to leverage environmental and social ``noise'' for adaptive reconfiguration, leading to improved viability compared to purely reactive homeostatic agents. This work offers a novel computational perspective on the principles of social allostasis and their potential for designing more robust, bio-inspired, adaptive systems

This paper introduces Energentic Intelligence, a class of autonomous systems defined not by task performance, but by their capacity to sustain themselves through internal energy regulation. Departing from conventional reward-driven paradigms, these agents treat survival-maintaining functional operation under fluctuating energetic and thermal conditions-as the central objective. We formalize this principle through an energy-based utility function and a viability-constrained survival horizon, and propose a modular architecture that integrates energy harvesting, thermal regulation, and adaptive computation into a closed-loop control system. A simulated environment demonstrates the emergence of stable, resource-aware behavior without external supervision. Together, these contributions provide a theoretical and architectural foundation for deploying autonomous agents in resource-volatile settings where persistence must be self-regulated and infrastructure cannot be assumed.

LLMs are computationally expensive to pre-train due to their large scale.Model growth emerges as a promising approach by leveraging smaller models to accelerate the training of larger ones. However, the viability of these model growth methods in efficient LLM pre-training remains underexplored.This work identifies three critical \underline{\textit{O}} bstacles: ( \textit{O} 1) lack of comprehensive evaluation, ( \textit{O} 2) untested viability for scaling, and ( \textit{O} 3) lack of empirical guidelines.To tackle \textit{O} 1, we summarize existing approaches into four atomic growth operators and systematically evaluate them in a standardized LLM pre-training setting.Our findings reveal that a depthwise stacking operator, called G_{\text{stack}}, exhibits remarkable acceleration in training, leading to decreased loss and improved overall performance on eight standard NLP benchmarks compared to strong baselines. Motivated by these promising results, we conduct extensive experiments to delve deeper into G_{\text{stack}} to address \textit{O} 2 and \textit{O} 3.For \textit{O} 2 (untested scalability), our study shows that G_{\text{stack}} is scalable and consistently performs well, with experiments up to 7B LLMs after growth and pre-training LLMs with 750B tokens.For example, compared to a conventionally trained 7B model using 300B tokens, our G_{\text{stack}} model converges to the same loss with 194B tokens, resulting in a 54.6\% speedup. We further address \textit{O} 3 (lack of empirical guidelines) by formalizing guidelines to determine growth timing and growth factor for G_{\text{stack}}, making it practical in general LLM pre-training.We also provide in-depth discussions and comprehensive ablation studies of G_{\text{stack}} . Our code and pre-trained model are available at https://llm-stacking.github.io/.

Myocardial infarction (MI) is one of the most prevalent cardiovascular diseases and consequently, a major cause for mortality and morbidity worldwide. Accurate assessment of myocardial tissue viability for post-MI patients is critical for diagnosis and treatment planning, e.g. allowing surgical revascularization, or to determine the risk of adverse cardiovascular events in the future. Fine-grained analysis of the myocardium and its surrounding anatomical structures can be performed by combining the information obtained from complementary medical imaging techniques. In this work, we use late gadolinium enhanced (LGE) magnetic resonance (MR), T2-weighted (T2) MR and balanced steady-state free precession (bSSFP) cine MR in order to semantically segment the left and right ventricle, healthy and scarred myocardial tissue, as well as edema. To this end, we propose the Multi-Sequence Cascading Refinement CNN (MS-CaRe-CNN), a 2-stage CNN cascade that receives multi-sequence data and generates predictions of the anatomical structures of interest without considering tissue viability at Stage 1. The prediction of Stage 1 is then further refined in Stage 2, where the model additionally distinguishes myocardial tissue based on viability, i.e. healthy, scarred and edema regions. Our proposed method is set up as a 5-fold ensemble and semantically segments scar tissue achieving 62.31% DSC and 82.65% precision, as well as 63.78% DSC and 87.69% precision for the combined scar and edema region. These promising results for such small and challenging structures confirm that MS-CaRe-CNN is well-suited to generate semantic segmentations to assess the viability of myocardial tissue, enabling downstream tasks like personalized therapy planning.

The concept of personalised medicine in cancer therapy is becoming increasingly important. There already exist drugs administered specifically for patients with tumours presenting well-defined mutations. However, the field is still in its infancy, and personalised treatments are far from being standard of care. Personalised medicine is often associated with the utilisation of omics data. Yet, implementation of multi-omics data has proven difficult, due to the variety and scale of the information within the data, as well as the complexity behind the myriad of interactions taking place within the cell. An alternative approach to precision medicine is to employ a function-based profile of the cell. This involves screening a range of drugs against patient derived cells. Here we demonstrate a proof-of-concept, where a collection of drug screens against a highly diverse set of patient-derived cell lines, are leveraged to identify putative treatment options for a 'new patient'. We show that this methodology is highly efficient in ranking the drugs according to their activity towards the target cells. We argue that this approach offers great potential, as activities can be efficiently imputed from various subsets of the drug treated cell lines that do not necessarily originate from the same tissue type.