Defense Advanced Research Projects Agency

Abstract The DARPA Transfer from Imprecise and Abstract Models to Autonomous Technologies (TIAMAT) program aims to address rapid and robust transfer of autonomy technologies across dynamic and complex environments, goals, and platforms. Existing methods for simulation‐to‐reality (sim‐to‐real) transfer often rely on high‐fidelity simulations and struggle with broad adaptation, particularly in time‐sensitive scenarios. Although many approaches have shown incredible performance at specific tasks, most techniques fall short when posed with unforeseen, complex, and dynamic real‐world scenarios due to the inherent limitations of simulation. In contrast to current research that aims to bridge the gap between simulation environments and the real world through increasingly sophisticated simulations and a combination of methods typically assuming a small sim‐to‐real gap—such as domain randomization, domain adaptation, imitation learning, meta‐learning, policy distillation, and dynamic optimization—TIAMAT takes a different approach by instead emphasizing transfer and adaptation of the autonomy stack directly to real‐world environments by utilizing a breadth of low(er)‐fidelity simulations to create broadly effective sim‐to‐real transfers. By abstractly learning from multiple simulation environments in reference to their shared semantics, TIAMAT's approaches aim to achieve abstract‐to‐real transfer for effective and rapid real‐world adaptation. Furthermore, this program endeavors to improve the overall autonomy pipeline by addressing the inherent challenges in translating simulated behaviors into effective real‐world performance.

Aims: Small bowel tumors are a heterogeneous group of malignancies, including gastrointestinal stromal tumors (GISTs), adenocarcinomas, neuroendocrine tumors (NETs), and myofibroblastic tumors, each with distinct prognostic implications. While tumor stage is a well-established prognostic factor, patient survival outcomes and systemic inflammatory markers also play a crucial role in disease progression. This study evaluates these factors comprehensively to enhance prognostic assessment in small bowel malignancies. Methods: This retrospective study analyzed 25 patients diagnosed with small bowel tumors, including various histological subtypes. The prognostic significance of tumor stage (T and N classification), systemic inflammatory markers (neutrophil-to-lymphocyte ratio [NLR], platelet-to-lymphocyte ratio [PLR], albumin, and C-reactive protein [CRP]), and overall survival was assessed. Kaplan-Meier survival analysis was conducted to evaluate the association between tumor stage, inflammatory markers, and patient outcomes. Statistical analyses included independent sample t-tests, Mann-Whitney U tests, and Chi-square tests. Results: The median age of the cohort was 63 years (range: 47–81). The most common histological subtype was GIST (52%), followed by adenocarcinoma (24%), NET (20%), and myofibroblastic tumors (4%). Kaplan-Meier survival analysis revealed a significant association between tumor stage and patient survival (p=0.036), with advanced-stage tumors (T3–T4) demonstrating significantly lower survival rates compared to early-stage tumors (T2). Lymph node involvement (N stage) was also a significant predictor of reduced survival (p=0.013). Although inflammatory markers such as NLR, PLR, albumin, and CRP were assessed, they did not show statistically significant associations with survival outcomes (p>0.05). Conclusion: This study highlights the importance of evaluating both tumor stage and patient survival when determining prognosis in small bowel tumors. The Kaplan-Meier analysis underscores the strong prognostic impact of tumor staging and lymph node involvement on survival outcomes. Although systemic inflammatory markers did not show significant prognostic value in this cohort, their role in risk stratification warrants further investigation in larger studies. These findings contribute to a broader understanding of small bowel tumor prognosis beyond staging alone, supporting the need for a multidimensional approach in clinical assessment and treatment planning.

Abstract Background Sarcopenia is one of the syndromes that cause falls, fractures, and morbidity in geriatric patients. Early diagnosis of sarcopenia is important as it is known that muscle functions improve with early intervention. We aimed to investigate whether urinary amino acid levels are a biomarker of sarcopenia. Methods The study included ninety-one patients aged 45–65 who applied to our outpatient clinic. The patients underwent physical examinations, blood tests and sixteen different urine amino acid levels were analyzed. Anthropometric measurements were made. Physical performances were evaluated. Muscle strengths were measured. Muscle masses were analyzed. Patients were divided into 4 groups: pre-sarcopenic, sarcopenic, severe sarcopenic and non-sarcopenic. Statistical significance level was determined as p < 0.05. Results A total of ninety-one patients, fifty-three female and thirty-eight males, were included in the study. Three patients had pre-sarcopenia, eleven had sarcopenia, and two met the criteria for severe sarcopenia, while seventy-five patients were non sarcopenic. A significant difference was found between the sarcopenia and non-sarcopenia groups in terms of glutamine and valine levels (p < 0.001 for both). In the ROC analysis, the cut-off value of glutamine and valine levels in detecting sarcopenia was determined as 492 micromole/L and 209 micromole/L (AUC:0.875;0.968 respectively). In correlation analysis between urine amino acid levels and muscle strength-mass a negative correlation was found between leucine and muscle strength and muscle mass. Conclusions We found that patients with sarcopenia had high urinary glutamine and valine levels and that urinary leucine levels were associated with both muscle strength and muscle mass.

Today's digital phased array architectures face the challenges of exponential growth in processing and digital data flow as arrays increase in size, element count, instantaneous bandwidth, and other factors. Traditional methods of beamforming, adaptive beamforming, and heretofore unexplored array functions cannot be sustained due to processing and data flow bottlenecks common to current digital architectures, which remain linked to inherently analog methods of beam-oriented array processing. DARPA's Scalable On-Array Processing (SOAP) [1] aims to look at new architectures that support new forms of on-array distributed processing architectures and data flow management to enable digital arrays to realize their full potential.

To achieve a truly closed-loop direct recycling process for lithium-ion batteries, all component materials must be recovered. To date, direct recycling method development has primarily focused on the high-value transition-metal cathode materials, while the inherently lower-value graphite has been challenging to recover in a cost-effective manner. However, end-of-life graphite contains a unique engineered value due to the presence of the solid electrolyte interphase (SEI). Growth of the SEI during the cell’s active lifetime stabilizes the electronically reactive graphite surface through an irreversible consumption of Li, and thus necessitates both excess lithiation of the cathode and a costly and time-intensive formation procedure during manufacturing. An optimized pre-formed SEI that capitalizes on existing SEI components from end-of-life batteries has the potential to significantly reduce cathode lithiation requirements and eliminate the critical bottleneck of formation cycling during cell remanufacturing. Further, retaining Li at the anode obviates the need for a separate Li leaching and recovery step, improving the overall efficiency of the direct recycling line. In this work, we present a novel approach to “upcycling” spent graphite through use of tailored chemical treatment to remove adverse (i.e., highly resistive and/or poorly passivating) SEI species while retaining beneficially passivating components. We have explored a rational set of solvents spanning a range of polarity, proticity, and molecular size to evaluate structure-property-performance relationships between applied solvent(s), removed and remaining SEI species, and electrochemical response of the resulting graphite product. Further, we have developed and optimized a robust and holistic analysis procedure that couples symmetric-cell electrochemical testing, multi-modal materials characterization, and advanced electrochemical modeling. These analysis results inform a set of correlative metrics for graphite performance relative to both solvent properties and upcycled SEI composition. We demonstrate effective tunability in the residual SEI composition by varying solvent identity and concentration, and report on several promising solvent systems that achieve comparable performance to pristine graphite. Figure 1

The DARPA Robotic Autonomy in Complex Environments with Resiliency (RACER) program is focusing on developing highspeed offroad ground autonomy for military applications. The presentation will discuss the progress made to advance the state of the art in this domain and address the challenges that remain.

The DARPA Optomechanical Thermal Imaging (OpTIm) program seeks to develop a novel modality of low-SWaP quantum-limited infrared detection. Building upon proof-of-concept studies and detailed noise analyses, the OpTIm device concept amalgamates high performance optomechanical sensors, quantum-limited all-optical readout techniques, and spectrally resolved metamaterial-based IR absorbers to realize several order-of-magnitude improvements in both IR sensitivity and detection speed. I will give a brief overview of the OpTIm program concept, the over-arching goals of this program, and a summary of innovations that have resulted from the program thus far. I will also describe the numerous technological applications and opportunities that are engendered by the novel capabilities of OpTIm imaging systems.