Full-Parameter Fine-Tuning Method of LLMs for Sports Injury Prevention and Treatment

Firearm Injury and Mortality Prevention in Pediatric Health-care Settings.

We propose a memory-efficient finetuning algorithm for large language models (LLMs) that supports finetuning LLMs with 65B parameters in 2/3/4-bit precision on as little as one 24GB GPU. Our method, modular low-rank adaptation (ModuLoRA), integrates any user-specified weight quantizer with finetuning via low-rank adapters (LoRAs). Our approach relies on a simple quantization-agnostic backward pass that adaptively materializes low-precision LLM weights from a custom black-box quantization module. This approach enables finetuning 2-bit and 3-bit LLMs for the first time-leveraging state-of-the-art 2-bit QuIP# quantization and 3-bit OPTQ quantization-outperforming finetuning that relies on less sophisticated 4-bit and 8-bit methods. In our experiments, ModuLoRA attains competitive performance on text classification, natural language inference, and instruction following tasks using significantly less memory than existing approaches, and we also surpass the state-of-the-art ROUGE score on a popular summarization task. We release ModuLoRA together with a series of low-precision models as part of LLMTools, a user-friendly library for quantizing, running, and finetuning LLMs on consumer GPUs.

Injuries, practices and perceptions of Australian wheelchair sports participants

BackgroundThe annual injury prevalence of referees in community sports is very high and yet few are undertaking injury prevention (IP) strategies (1). For IP strategies to have the best chances...

Athletes' and coaches' lack of knowledge about health and injury progression is also a major factor in the occurrence of injuries among table tennis athletes. The purpose of this study was to identify common injury patterns, risk factors that influence injury progression, and effective injury prevention and management strategies. The method of this study was a literature review regarding the development of health and injuries in table tennis players obtained from sciencedirect, Eric, and Google Schoolar. Data analysis techniques were conducted by citing studies from around the world, identifying common injury patterns, risk factors that influence injury development, and effective injury prevention and management strategies. Key findings highlighted the importance of a holistic approach that encompasses physical, mental and social aspects in maintaining optimal health and performance for table tennis athletes. This study also emphasizes the need for integration of recent findings in research to support innovation in injury prevention strategies.

The field of Artificial Intelligence has witnessed remarkable progress in recent years, especially with the emergence of large language models (LLMs) based on the transformer architecture. Cloud-based LLMs, such as OpenAI’s ChatGPT, offer impressive capabilities but come with concerns regarding latency and privacy due to network dependencies. This article presents an approach to LLM inference that allows LLMs with billions of parameters to be executed directly on mobile devices without network connectivity. The article showcases a fine-tuned GPT LLM with 3 billion parameters that can operate on devices with as low as 4GB of memory. Through the integration of native code and model quantization techniques, the application not only serves as a general-purpose assistant but also facilitates mobile interactions with our text-to-actions feature. With text-to-actions, the assistant capabilities are extended beyond just text conversation, enabling the communication between the low-level LLM and the device’s operating system, autonomously performing tasks such as making calls, searching the web, or scheduling events. The article provides insights into the training pipeline, implementation details, test results, and future directions of on-device LLM inference. This technology opens up possibilities for empowering users with sophisticated AI capabilities while preserving their privacy and eliminating latency concerns.

ObjectivesThe injury prevalence in Gaelic games refereeing is high, however few are adopting injury prevention programmes. This study aims to identify the barriers and facilitators to injury prevention strategy success and determine Ladies Gaelic Football referees’ preferences for injury prevention strategies and education. DesignSemi-structured interviews were conducted with 11 Ladies Gaelic Football referees (10 men, 1 woman). Two were club level, two were provincial level and 7 were national level referees. Interviews were audio-recorded, transcribed verbatim, and reflexive thematic analysis was completed. This analysis involved examining the data repeatedly and gradually developing sub-themes, themes, and categories related to each core concept. ResultsThe barriers to injury prevention success included negative attitudes, accessibility issues, lack of education, the state of refereeing and undesirable injury prevention strategy characteristics. Injury prevention promotion, suitable strategy characteristics and open communication were believed to facilitate success. Referees gave their preferences for injury prevention programmes, strategy logistics, and stakeholder roles along with their preferred topics, delivery, educators, characteristics, rollout, and timing for injury prevention education. ConclusionsReducing referee injury is critical to the success of Ladies Gaelic Football and other community sports. Governing bodies must develop and support injury prevention programmes and education for referees. These should be designed according to referees’ preferences and consider the barriers and facilitators referees have identified to maximise adoption.

Certified athletic trainers are positioned to play an integral role in sport-related, recreation-related, and exercise-related injury research and prevention efforts. The Centers for Disease Control and Prevention1 have specifically identified certified athletic trainers and their potential to contribute to this important area of research in their recently published Injury Research Agenda for 2009–2018. From a public health perspective, identifying factors associated with injury is one of the initial steps in the injury prevention process. The ultimate goal of this line of research is to identify populations that are at greatest risk for subsequent injury and to develop effective screening and intervention strategies to reduce the incidence and burden of injury. Therefore, the manner in which risk factors are conceptualized in initial research has significant bearing on the development of subsequent screening and intervention strategies. In the current issue of the Journal of Athletic Training, Reinking et al2 examined the factors associated with exerciserelated leg pain in high school cross-country athletes. After a discussion of modifiable and nonmodifiable risk factors, the authors conceptualized risk factors for exertional leg pain as intrinsic (within the body) or extrinsic (outside the body) factors; sports injury researchers have traditionally conceptualized risk factors associated with injury in this manner.3–5 This study provides an opportunity to discuss how risk factors are conceptualized in the context of sports injury research. The seminal work by sports injury researchers such as van Mechelen et al3 and Meeuwisse5 provided a theoretic framework for sports injury research, and although these models have evolved since they were first described,6–9 they have consistently characterized risk factors for injury as intrinsic factors and extrinsic factors. It is important to consider that injuries often result from the complex interaction of multiple factors3,7,10 in a dynamically changing environment,6 but the characterization of risk factors as intrinsic or extrinsic limits the clinical importance and usefulness of results in relation to future injury screening and prevention efforts. More importantly, characterizing risk factors in this manner provides limited insight into whether something can be done to intervene and mitigate the contributing influence of any given factor or combination of factors with regard to subsequent injury.11 Injury epidemiologists and public health professionals, on the other hand, use a different approach to conceptualizing risk factors for disease or injury by focusing on those risk factors that are modifiable and those that are nonmodifiable. Conceptualizing risk factors as modifiable and nonmodifiable is important from a clinical and injury prevention perspective, because modifiable risk factors are amenable to intervention.11 Researchers12,13 of chronic diseases, such as cardiovascular disease, hypertension, and type 2 diabetes, have conceptualized risk factors as modifiable and nonmodifiable. Some of these risk factors include physical activity, diet, smoking, and obesity, to name a few. Subsequent intervention efforts have been developed and evaluated to address these factors, with various degrees of success. Conceptualizing risk factors as modifiable and nonmodifiable aligns with the ‘‘Translating Research into Injury Prevention Practice (TRIPP)’’ framework described by Finch,9 which applies a public health approach to sports injury prevention. Clinicians and injury researchers are most interested in modifiable risk factors associated with injury because they provide the vector for developing injury prevention interventions. Although nonmodifiable risk factors may not be useful as targets for intervention, they are particularly important in identifying populations that are at greatest risk for injury, so that injury prevention strategies can be directed to those with the most pressing need. This information can be used to develop risk profiles for specific injuries, to screen athletes to identify those at greatest risk for injury, and to guide injury prevention interventions that target modifiable risk factors. This information can also be used to develop interventions and social marketing campaigns that are culturally and ecologically appropriate for the populations at greatest risk for injury.14 When only nonmodifiable risk factors associated with injury are known, this information can be used to counsel athletes and parents about potential risks during the preparticipation screening process, so they can make informed decisions about participation. Recently, researchers have conceptualized risk factors as modifiable and nonmodifiable in the sports medicine literature. Emerging research10,15–17 into the risk factors associated with anterior cruciate ligament (ACL) injuries is one of the best examples in the sports medicine literature of how the conceptualization of risk factors has evolved. Early work15 in this area conceptualized risk factors as intrinsic and extrinsic, while recognizing the importance of focusing on factors that are modifiable. Subsequent consensus statements10,16 on the risk factors for ACL

Aim/Purpose: The study investigates the factors influencing the acceptance and utilisation of large language models (LLMs) (predictor variables of LLM usage), such as ChatGPT, in Learning design by instructional designers and university-teaching academics from various countries. Background: Large language models (LLMs) have exploded onto the scene, transforming the landscape of learning design. Instructional designers and university teaching academics have been overburdened with content creation for their teaching programmes, and the arrival of LLM models will help in this regard by developing more interactive content that drives student engagement and, in turn, contributes to student success. Since LLMs are a relatively new phenomenon, little is known about the factors influencing their acceptance in learning design; therefore, this research is needed, as learning design principles are the bedrock of student engagement and success. Methodology: A cross-sectional correlational quantitative study was employed. Data was collected using an online questionnaire posted on social media, including LinkedIn, from 203 instructional designers and university teaching academics. Purposive and snowball sampling methods were used to target instructional designers and university teaching academics at colleges and universities worldwide. Participants were asked to share the survey link with fellow instructional designers and university-teaching academics in their communities. The factor structure of the data was determined using exploratory factor analysis. Nonetheless, the factor structure derived from the LLMs did not entirely reflect the original configuration of the Unified Theory of Acceptance and Use of Technology (UTAUT3), as certain predictors appeared to coalesce, indicating LLMs’ unique nature in learning design. Confirmatory factor analysis was used to verify the fit of the data on the measurement model. First-order and second-order structural modelling were used to identify the structural relationships among the variables. Contribution: The study determines significant factors for the acceptance of LLMs by instructional designers and academic teaching staff in learning design, enabling possible opportunities for best practices in the field through interventions to optimize LLM usage. The study applies the technology acceptance model to the emerging LLM technology and extends the technology acceptance model by adding the trust construct as a predictor variable. Findings: The structural analysis results indicated that the ingrained LLM practices, LLM peer-driven expectations, innovative propensity towards LLM adoption, reliability and provider trust in LLMs, and ease of use and support influenced perceived LLM benefits and usage, but community standards and infrastructure had no influence. The second-order structural equation modelling indicated that perceived LLM benefits and usage and ingrained LLM habits contributed most to the learning design. Recommendations for Practitioners: Teaching academics and instructional designers must use LLMs in designing content, assessments, and interactive learning activities, and attend LLM training workshops on prompting and best practices in integrating LLMs into learning and teaching to see their benefits; hence, regular use of LLMs will then lead to trust and innovation in LLMs usage, enhancing learning design and improving student learning outcomes. Recommendation for Researchers: Researchers must use mixed methods approaches to have a deeper understanding of the factors influencing LLMs. Since habit and perceived LLM benefits and usage contributed the most variance to learning design, researchers must investigate strategies that optimise these factors in learning design, such as effective intervention strategies that can help form positive LLM habits. In addition, the findings provide researchers with a starting point for future research. Further researchers must investigate interventions that optimise the influence of personal innovativeness and trust that contributed the least variance to learning design, hence unlocking the potential of LLMs in learning design through innovation, responsible, and ethical use. Impact on Society: The use of LLMs in learning design has a high possibility of transforming education, specifically the learning design landscape. Using LLMs will free up more time for teaching academics and instructional designers so that they spend more time on higher-order thinking skill demands. Consequently, the students will be exposed to more engaging and interactive content, resulting in improved learning outcomes. Future Research: Future research must include context-derived external variables in technology acceptance models, such as levels of prompting competencies, to provide a deeper understanding of LLMs. In addition, future research must be based on the application and impact of LLMs on student engagement and success, and their attainment of 21st-century skills.

BackgroundOver the last decade, injury rates in Ladies Gaelic Football (LGF) have remained high, indicating that injury prevention (IP) strategies are desperately needed (1). Despite efforts to implement IP strategies,...

Sports injury rehabilitation should be sports-and condition-specific, address muscle and functional performance, and consider the joint health, while tissue type and mechanism should be considered in return to sport and injury prevention strategies.In the following section we outline an overview of recent research on common conditions such as patellofemoral pain (PFP), anterior cruciate ligament reconstruction (ACLR) and the athletic hip, which provides valuable insights into condition-specific rehabilitation approaches and recovery trajectories, and we outline an overview describing the importance of incorporating injury tissue type and mechanism into prevention strategies.Patellofemoral pain remains one of the most common injuries among runners, with muscle strengthening being an important component for stage-specific rehabilitation and return to sport.Evidence from recent research in PFP by He et al. (2025) indicates that quadriceps strength varies according to the duration of symptoms, with individuals exhibiting reduced strength in the short-term (< 3 months) but not in the long-term (>12 months). Interestingly, He et al. (2025) found that individuals with long-term symptom duration demonstrated comparable strength levels to healthy controls. In contrast, hamstring strength, hamstring-to-quadriceps ratios, and muscle symmetry appear largely unaffected by PFP duration (He et al., 2025). Although limitations exist-particularly the use of single-velocity isokinetic testing and concentric-only contractions-these findings enhance our understanding of strength adaptations across PFP stages and may guide more targeted, stagespecific rehabilitation strategies. With a high rate of hip and groin re-injuries, the authors stress there is a need for effective secondary and tertiary prevention strategies (Bizzini et al., 2025).Addressing the injury by tissue type and mechanism during the rehabilitation process may enable implementation of better injury preventative strategies to reduce re-injury risks. A recent narrative review on a synthesis of football injury types and prevention strategies by Zeng et al. ( 2025) proposed a theoretical basis for understanding injuries in athletes. As proposed by Zeng et al. (2025), skeletal muscle injuries, including muscle fiber and tendon injuries, may be mitigated through eccentric strength training, whereas joint injuries such as ligament damage and muscle imbalances require emphasis on neuromuscular control. Degenerative injuries demand systematic, often long-term management, with surgical intervention considered when appropriate and followed by individualized rehabilitation. Accidental injuries, including concussions and fractures, can be reduced through protective equipment, rule modifications, structured training programs and enhanced safety education. The authors conclude that by ensuring injury prevention strategies are mechanismspecific, injury risk in sports may be reduced (Zeng et al., 2025).Equally important-but often underemphasized-is the psychological dimension of RTS. In the following section we outline findings from two recent studies that emphasized the significance of psychological considerations in return to sport and injury prevention. 2025) recently reported higher psychological readiness associated with reduced kinesiophobia among adolescents and young adults 6 -12 months following ACLR, although nearly half of athletes from both groups reported poor psychological readiness. These findings underscore the need for targeted psychological interventions during rehabilitation to improve psychological readiness for return to sport. The authors recommend future research should further explore the influence of sex and age on emotional responses, confidence, and risk appraisal, while accounting for variability in rehabilitation protocols, surgical details, and actual RTS outcomes (Butler et al., 2025).In conclusion, while sports injuries may be inevitable, recurrent injuries are not. We presented evidence from six recent articles included in this topic on knowledge in injury rehabilitation and return to sport practices, advocating for a holistic rehabilitation model that integrates physical, psychological, and social factors alongside evidence-based physiotherapy. Such an approach not only supports a successful return to sport, but also reduces the risk of long-term re-injury, psychological distress, and performance decline. The ultimate goal of rehabilitation is not merely returning to sport-but returning well, and with injury-free participation and performance enhancement over time.Barzyk, P., Fiedler, C., Schlag, M., Heitner, A., Bender, J., & Paul, J. (2026)

Purpose(1) To quantify current practice at the most elite level of professional club football in Europe with regard to injury prevention strategy; (2) to describe player adherence and coach compliance...

BackgroundSports injuries remain a significant challenge to athletes' health, performance, and long-term participation, particularly in high-contact and endurance sports. Existing prevention frameworks often lack sport-specific adaptability and integration with rehabilitation.ObjectivesThis study aimed to develop and refine an Integrated Injury Prevention and Rehabilitation Model (IIPRM) through a qualitative, multi-phase consensus process.MethodsA three-stage, sequential qualitative design was used. Stage 1 was a scoping review of injury risks and prevention strategies in rugby and long-distance running to identify core concepts for the model. Stage 2 used a modified e-Delphi with experts (n=22; three rounds), with consensus predefined as ≥75% agreement; panel retention was 100% across Rounds 1–3 and 18.2% attrition in Round 4. Stakeholder focus groups (n=10) were used to refine clarity, feasibility, and contextual relevance. Stage 3 combined the final fourth Delphi round and focus-group feedback to build consensus and validate the model.ResultsAcross Delphi rounds, 82% of items reached the ≥75% consensus threshold by Round 2, with remaining items achieving consensus in Round 3. The final IIPRM is a cyclical, athlete-centred framework integrating primary and secondary prevention through education, pre-participation screening, risk-targeted interventions, functional reintegration, and continuous monitoring. Stakeholder input prompted the inclusion of psychological readiness screening, enhanced recovery education, and seasonal alignment of interventions.ConclusionThe IIPRM provides a conceptually practical and adaptable approach to injury prevention and rehabilitation, positioned to facilitate interdisciplinary collaboration and long-term athlete health. As the model has not yet been implemented or evaluated, future research should examine its feasibility, effectiveness, and transferability across sports.

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 217671, “Enhancing Information Retrieval in the Drilling Domain: Zero-Shot Learning With Large Language Models for Question Answering,” by Felix J. Pacis, SPE, University of Stavanger, and Sergey Alyaev and Gilles Pelfrene, SPE, NORCE, et al. The paper has not been peer reviewed. _ Finding information across multiple databases, formats, and documents remains a manual job in the drilling industry. Large language models (LLMs) have proven effective in data-aggregation tasks, including answering questions. However, using LLMs for domain-specific factual responses poses a nontrivial challenge. The expert-labor cost for training domain-specific LLMs prohibits niche industries from developing custom question-answering bots. The complete paper tests several commercial LLMs for information-retrieval tasks for drilling data using zero-shot in-context learning. In addition, the model’s calibration is tested with a few-shot multiple-choice drilling questionnaire. Introduction While LLMs have proven effective in various tasks ranging from sentiment analysis to text completion, using LLMs for question-answering tasks presents a challenge in providing factual responses. Pretrained LLMs only serve as a parameterized implicit knowledge base and cannot access recent data; thus, information is bounded by the time of training. Retrieval augmented generation (RAG) can address some of these issues by extending the utility of LLMs to specific data sources. Fig. 1 shows a simplified RAG-based LLM question/answer application. RAG involves two primary components: document retrieval (green boxes), which retrieves the most relevant context based on the query, and LLM response generation (blue boxes). During the response generation, LLM operates based on the prompt, query, and retrieved context without any change in the model parameters, a process the authors term as “in-context learning.” Methodology Two experiments have been conducted: The first one is a few-shot multiple-choice experiment evaluated using the SLB drilling glossary; the second is a zero-shot in-context experiment evaluated on drilling reports and company reports. Multiple-Choice Experiment. SLB Drilling Glossary. For the multiple-choice experiment, a publicly available drilling glossary served as a basis for evaluation. A total of 409 term/definition pairs were considered. Five term/definition pairs were chosen, serving as few-shot default values, while the remaining 404 pairs served as the multiple-choice questions. Four choices were given for each term/definition question pair, where one was the correct answer. The three incorrect choices were picked randomly from all possible terms minus the true answer. Zero-Shot In-Context Experiment. Norwegian Petroleum Directorate (NPD) Database. The authors explored the wellbore history of all individual exploration wells drilled in the Norwegian shelf in the NPD database. In this experiment, 12 exploration wells were randomly chosen for evaluation. In addition to these drilling reports, information about the stratigraphy of three additional wells was added. Annual Reports. Annual reports of two major operators in Norway for 2020 and 2021 also were considered. These consisted of short summaries that presented the main operational and economic results achieved by the company throughout the year. These reports were added to the evaluation to balance the higher technical content of the wellbore-history reports.

BackgroundParticipating in organised sports, such as Ladies Gaelic Football (LGF), during adolescence can lead to physical and mental health benefits (1). However, sports participation is also a major source of...