A West Coast biotech entrepreneur says he’s secured $30 million to form a public-benefit company to study how to safely create genetically edited babies, marking the largest known investment into the taboo technology.  

The new company, called Preventive, is being formed to research so-called “heritable genome editing,” in which the DNA of embryos would be modified by correcting harmful mutations or installing beneficial genes. The goal would be to prevent disease.

Preventive was founded by the gene-editing scientist Lucas Harrington, who described his plans yesterday in a blog post announcing the venture. Preventive, he said, will not rush to try out the technique but instead will dedicate itself “to rigorously researching whether heritable genome editing can be done safely and responsibly.”

Creating genetically edited humans remains controversial, and the first scientist to do it, in China, was imprisoned for three years. The procedure remains illegal in many countries, including the US, and doubts surround its usefulness as a form of medicine.

Still, as gene-editing technology races forward, the temptation to shape the future of the species may prove irresistible, particularly to entrepreneurs keen to put their stamp on the human condition. In theory, even small genetic tweaks could create people who never get heart disease or Alzheimer’s, and who would pass those traits on to their own offspring.

According to Harrington, if the technique proves safe, it “could become one of the most important health technologies of our time.” He has estimated that editing an embryo would cost only about $5,000 and believes regulations could change in the future. 

Preventive is the third US startup this year to say it is pursuing technology to produce gene-edited babies. The first, Bootstrap Bio, based in California, is reportedly seeking seed funding and has an interest in enhancing intelligence. Another, Manhattan Genomics, is also in the formation stage but has not announced funding yet.

As of now, none of these companies have significant staff or facilities, and they largely lack any credibility among mainstream gene-editing scientists. Reached by email, Fyodor Urnov, an expert in gene editing at the University of California, Berkeley, where Harrington studied, said he believes such ventures should not move forward.

Urnov has been a pointed critic of the concept of heritable genome editing, calling it dangerous, misguided, and a distraction from the real benefits of gene editing to treat adults and children. 

In his email, Urnov said the launch of still another venture into the area made him want to “howl with pain.”  

Harrinton’s venture was incorporated in Delaware in May 2025,under the name Preventive Medicine PBC. As a public-benefit corporation, it is organized to put its public mission above profits. “If our research shows [heritable genome editing] cannot be done safely, that conclusion is equally valuable to the scientific community and society,” Harrington wrote in his post.

Harrington is a cofounder of Mammoth Biosciences, a gene-editing company pursuing drugs for adults, and remains a board member there.

In recent months, Preventive has sought endorsements from leading figures in genome editing, but according to its post, it had secured only one—from Paula Amato, a fertility doctor at Oregon Health Sciences University, who said she had agreed to act as an advisor to the company.

Amato is a member of a US team that has researched embryo editing in the country since 2017, and she has promoted the technology as a way to increase IVF success. That could be the case if editing could correct abnormal embryos, making more available for use in trying to create a pregnancy.

It remains unclear where Preventive’s funding is coming from. Harrington said the $30 million was gathered from “private funders who share our commitment to pursuing this research responsibly.” But he declined to identify those investors other than SciFounders, a venture firm he runs with his personal and business partner Matt Krisiloff, the CEO of the biotech company Conception, which aims to create human eggs from stem cells.

That’s yet another technology that could change reproduction, if it works. Krisiloff is listed as a member of Preventive’s founding team.

The idea of edited babies has received growing attention from figures in the cryptocurrency business. These include Brian Armstrong, the billionaire founder of Coinbase, who has held a series of off-the-record dinners to discuss the technology (which Harrington attended). Armstrong previously argued that the “time is right” for a startup venture in the area.

Will Harborne, a crypto entrepreneur and partner at LongGame Ventures, says he’s “thrilled” to see Preventive launch. If the technology proves safe, he argues, “widespread adoption is inevitable,” calling its use a “societal obligation.”

Harborne’s fund has invested in Herasight, a company that uses genetic tests to rank IVF embryos for future IQ and other traits. That’s another hotly debated technology, but one that has already reached the market, since such testing isn’t strictly regulated. Some have begun to use the term “human enhancement companies” to refer to such ventures.

What’s still lacking is evidence that leading gene-editing specialists support these ventures. Preventive was unsuccessful in establishing a collaboration with at least one key research group, and Urnov says he had harsh words for Manhattan Genomics when that company reached out to him about working together. “I encourage you to stop,” he wrote back. “You will cause zero good and formidable harm.”

Harrington thinks Preventive could change such attitudes, if it shows that it is serious about doing responsible research. “Most scientists I speak with either accept embryo editing as inevitable or are enthusiastic about the potential but hesitate to voice these opinions publicly,” he told MIT Technology Review earlier this year. “Part of being more public about this is to encourage others in the field to discuss this instead of ignoring it.”

Health insurance companies are fighting fire with fire as they combat rising medical costs they say are being driven up, in part, by artificial intelligence. 

Large private insurers have continued to call out purportedly overzealous coding and billing as a source of ballooning health care costs. In particular, some have blamed the use of AI tools for some of the surge in claims from health care providers that have cut into their profits. 

“I’ve talked to a couple of my peers where it does seem like the hospitals have gotten better organized around the application of AI for coding than payers,” said Centene chief financial officer Andrew Asher at the Deutsche Bank Healthcare Summit last month. “But we’re going to catch up to that.”

Continue to STAT+ to read the full story…

© Adobe

Nature, Published online: 31 October 2025; doi:10.1038/d41586-025-03566-8

In a world first, a bespoke gene-editing therapy benefited one child. Now researchers plan to launch a clinical trial of the approach.

Background: Epilepsy affects approximately 50 million people globally and imposes a substantial clinical and societal burden, requiring continuous and personalized monitoring for effective management. Wearable artificial intelligence (AI) technologies offer a promising solution by leveraging physiological signals and machine learning for seizure detection and prediction. While various approaches have been proposed, a comprehensive overview summarizing these advances and challenges is still needed. Objective: This review aims to comprehensively explore and map the existing literature on AI-driven wearable technologies for epilepsy, identifying device characteristics, AI methodologies, biosignal measurements, validation approaches, and research gaps. Methods: A scoping review was conducted following the PRISMA-ScR guidelines. A systematic search was performed across six electronic databases (Scopus, MEDLINE, EMBASE, ACM Digital Library, IEEE Xplore, and Google Scholar) to identify relevant studies published up to December 2023. We included studies that developed AI algorithms for epilepsy using non-invasive wearable devices (e.g., smartwatches, smart clothing) and excluded those using non-wearables or in-body devices. Eligible publication types included journal articles, conference papers, and dissertations. Study selection and data extraction were performed independently by six reviewers. The extracted data was synthesized narratively. Results: A total of 68 studies met the inclusion criteria. Research in this domain has increased significantly since 2021, with India, the United States, and China leading contributions. The studies examined both commercial (45.6%) and non-commercial (47.1%) wearable devices, with Empatica smart bands being the most frequently used. The primary biosignals monitored included activity measures (54.4%), cardiovascular metrics (45.6%), brain activity (35.3%), and electrodermal activity (33.8%). The most common AI models were support vector machines (42.6%), random forests (22.1%), and convolutional neural networks (16.2%). Most models focused on seizure detection (77.5%) compared to seizure prediction (22.5%), reflecting a research imbalance that suggests the need for further development in predictive analytics. Sensitivity (80.9%) was the most frequently reported performance metric, indicating a focus on identifying seizures; however, comprehensive clinical validation remains limited. Closed-source data predominated (64.7%), limiting the generalizability of findings. The most used validation methods were leave-one-out cross-validation (30.9%) and k-fold cross-validation (29.4%), while video-EEG served as the primary reference standard (42.6%). Conclusions: Wearable AI technologies show significant promise in epilepsy management, offering real-time, continuous monitoring and early seizure detection. To realize clinical impact, future research should prioritize the standardization of validation methods, promote open data exchange for reproducibility, and develop energy-efficient algorithms that support real-world deployment in wearable devices.

Background: To be authorized for use in the National Health Service (NHS) in England, digital health technologies (DHTs) must meet 2 mandatory clinical risk management standards, Data Coordination Board (DCB) 0129 and 0160, demonstrating that risks from design and use have been assessed and mitigated. NHS organizations must not procure a DHT without DCB0129 assurance and must not deploy one without DCB0160 assurance. Despite legal requirement, no public data exist on how many DHTs are in use in the NHS or how many are assured. Objective: This study aimed to determine the number of DHTs in use in the NHS in England and assess their assurance status against mandated clinical safety standards. Methods: In early 2025, 239 NHS organizations in England received a freedom of information notice requesting information on the number of DHTs they were using and their assurance against DCB0129 and DCB0160 standards. Results: Of the 239 NHS organizations, 204 (85.4%) responded, of which 178 (87.3%) provided full or partial data, covering 14,747 DHT deployments. The mean number of deployed DHTs per organization was 82.8 (SD 146.1; 95% CI 61.4-104.3) with substantial variation between NHS provider trusts (mean 107.1, SD 161.1; 95% CI 79.8-134.3), ambulance trusts (mean 13.0, SD 8.2; 95% CI 7.6-18.4), and integrated care boards (mean 8.1, SD 16.0; 95% CI 2.8-13.5). Overall organizational compliance rates were low, with a median of 25.6% (IQR 7.8%-55.7%) deployed DHTs being fully assured; for NHS provider trusts compliance was lower at 24.5% (IQR 8.1%-50%). A total of 13 (6.4%) of the 204 organizations reported that all their DHTs were fully assured, while 16 (7.8%) reported that none were assured. Across all DHTs with reported assurance data, 17.3% (95% CI 16.6%-18.1%) were fully assured against both standards, 13.3% were partially assured against one standard, and 70.1% (95% CI 69.1%-71.1%) had no documented assurance. Conclusions: This is the first study to quantify both the scale of DHT deployment in NHS organizations in England and the extent of compliance with mandatory safety standards. More than 10,000 DHTs currently in use lack documented assurance against clinical safety standards. In a typical NHS trust, 3 out of 4 digital tools influencing patient care do not demonstrate compliance with minimum legal or clinical safety requirements. These findings raise significant concerns about the risks posed to patients by these technologies; the capacity of organizations to assess and mitigate them; and the legal ramifications of when, not if, harm occurs. Crucially, failure to assure digital technologies poses a significant risk to one of the core ambitions of the NHS 10-Year Health Plan for England; safely transitioning from analogue to digital care models. These findings are unlikely to be unique to the NHS and should prompt health care systems worldwide to assess the risks posed by their DHT deployments.

Health care increasingly depends on information and communication technology. This offers both opportunities and challenges when designing connected health systems. While individual studies examined particular cases, there is a limited synthesis of insights across projects. The objective of this paper is to explore these opportunities and challenges by examining 6 diverse connected health projects and synthesizing lessons from an expert workshop. To achieve this, we conducted a full-day workshop that brought together 6 connected health projects. The workshop used an iterative and participatory process which included paper submissions and presentations and facilitated discussions, and a gallery walk to enable cross-case comparison and collaborative reflection. Thematic analysis of workshop outputs was then used to synthesize key opportunities and challenges in designing connected health systems. The 6 projects represented a variety of design methods and approaches to connected health, and their discussion surfaced both opportunities and challenges in this domain. Key opportunities include improving data integration and usability, enhancing collaboration across stakeholders, using a user-centered and iterative design process, addressing complexity in sociotechnical systems, sustainability, and adopting digital infrastructures for seamless communication. Participants also identified important challenges, namely exchange of information, interoperability, and communication; ethical considerations, rules, and regulations; understanding design, evaluation, and standards; actionable data, reliability, quality, and trust in data; and stakeholder involvement. The contribution of this paper lies in the synthesis of insights across multiple projects and perspectives to provide practical guidance for researchers, designers, and policymakers. By highlighting opportunities and challenges in designing connected health systems, the findings emphasize the importance of patient-centered, sustainable, and collaborative design approaches while also pointing to the need to address persistent barriers. Advancing connected health will require adopting iterative and inclusive design processes that prioritize patient-centeredness, sustainability, and collaboration across health care systems.

arXiv:2510.25813v1 Announce Type: new Abstract: We present a novel framework for Industry 5.0 that simplifies the deployment of AI models on edge devices in various industrial settings. The design reduces latency and avoids external data transfer by enabling local inference and real-time processing. Our implementation is agent-based, which means that individual agents, whether human, algorithmic, or collaborative, are responsible for well-defined tasks, enabling flexibility and simplifying integration. Moreover, our framework supports modular integration and maintains low resource requirements. Preliminary evaluations concerning the food industry in real scenarios indicate improved deployment time and system adaptability performance. The source code is publicly available at https://github.com/AI-REDGIO-5-0/ci-component.

arXiv:2510.25908v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated transformative potential in scientific research, yet their deployment in high-stakes contexts raises significant trustworthiness concerns. Here, we introduce SciTrust 2.0, a comprehensive framework for evaluating LLM trustworthiness in scientific applications across four dimensions: truthfulness, adversarial robustness, scientific safety, and scientific ethics. Our framework incorporates novel, open-ended truthfulness benchmarks developed through a verified reflection-tuning pipeline and expert validation, alongside a novel ethics benchmark for scientific research contexts covering eight subcategories including dual-use research and bias. We evaluated seven prominent LLMs, including four science-specialized models and three general-purpose industry models, using multiple evaluation metrics including accuracy, semantic similarity measures, and LLM-based scoring. General-purpose industry models overall outperformed science-specialized models across each trustworthiness dimension, with GPT-o4-mini demonstrating superior performance in truthfulness assessments and adversarial robustness. Science-specialized models showed significant deficiencies in logical and ethical reasoning capabilities, along with concerning vulnerabilities in safety evaluations, particularly in high-risk domains such as biosecurity and chemical weapons. By open-sourcing our framework, we provide a foundation for developing more trustworthy AI systems and advancing research on model safety and ethics in scientific contexts.

arXiv:2510.26057v1 Announce Type: new Abstract: The AI we use is powerful, and its power is increasing rapidly. If this powerful AI is to serve the needs of consumers, voters, and decision makers, then it is imperative that the AI is accountable. In general, an agent is accountable to a forum if the forum can request information from the agent about its actions, if the forum and the agent can discuss this information, and if the forum can sanction the agent. Unfortunately, in too many cases today's AI is not accountable -- we cannot question it, enter into a discussion with it, let alone sanction it. In this chapter we relate the general definition of accountability to AI, we illustrate what it means for AI to be accountable and unaccountable, and we explore approaches that can improve our chances of living in a world where all AI is accountable to those who are affected by it.

arXiv:2510.26518v1 Announce Type: new Abstract: Human feedback is critical for aligning AI systems to human values. As AI capabilities improve and AI is used to tackle more challenging tasks, verifying quality and safety becomes increasingly challenging. This paper explores how we can leverage AI to improve the quality of human oversight. We focus on an important safety problem that is already challenging for humans: fact-verification of AI outputs. We find that combining AI ratings and human ratings based on AI rater confidence is better than relying on either alone. Giving humans an AI fact-verification assistant further improves their accuracy, but the type of assistance matters. Displaying AI explanation, confidence, and labels leads to over-reliance, but just showing search results and evidence fosters more appropriate trust. These results have implications for Amplified Oversight -- the challenge of combining humans and AI to supervise AI systems even as they surpass human expert performance.

arXiv:2510.26603v1 Announce Type: new Abstract: The electricity sector transition requires substantial increases in residential demand response capacity, yet Home Energy Management Systems (HEMS) adoption remains limited by user interaction barriers requiring translation of everyday preferences into technical parameters. While large language models have been applied to energy systems as code generators and parameter extractors, no existing implementation deploys LLMs as autonomous coordinators managing the complete workflow from natural language input to multi-appliance scheduling. This paper presents an agentic AI HEMS where LLMs autonomously coordinate multi-appliance scheduling from natural language requests to device control, achieving optimal scheduling without example demonstrations. A hierarchical architecture combining one orchestrator with three specialist agents uses the ReAct pattern for iterative reasoning, enabling dynamic coordination without hardcoded workflows while integrating Google Calendar for context-aware deadline extraction. Evaluation across three open-source models using real Austrian day-ahead electricity prices reveals substantial capability differences. Llama-3.3-70B successfully coordinates all appliances across all scenarios to match cost-optimal benchmarks computed via mixed-integer linear programming, while other models achieve perfect single-appliance performance but struggle to coordinate all appliances simultaneously. Progressive prompt engineering experiments demonstrate that analytical query handling without explicit guidance remains unreliable despite models' general reasoning capabilities. We open-source the complete system including orchestration logic, agent prompts, tools, and web interfaces to enable reproducibility, extension, and future research.

arXiv:2510.25819v1 Announce Type: cross Abstract: The rapid rise of AI agents presents urgent challenges in authentication, authorization, and identity management. Current agent-centric protocols (like MCP) highlight the demand for clarified best practices in authentication and authorization. Looking ahead, ambitions for highly autonomous agents raise complex long-term questions regarding scalable access control, agent-centric identities, AI workload differentiation, and delegated authority. This OpenID Foundation whitepaper is for stakeholders at the intersection of AI agents and access management. It outlines the resources already available for securing today's agents and presents a strategic agenda to address the foundational authentication, authorization, and identity problems pivotal for tomorrow's widespread autonomous systems.

arXiv:2510.25929v1 Announce Type: cross Abstract: Algorithmic collusion has emerged as a central question in AI: Will the interaction between different AI agents deployed in markets lead to collusion? More generally, understanding how emergent behavior, be it a cartel or market dominance from more advanced bots, affects the market overall is an important research question. We propose a hierarchical multi-agent reinforcement learning framework to study algorithmic collusion in market making. The framework includes a self-interested market maker (Agent~A), which is trained in an uncertain environment shaped by an adversary, and three bottom-layer competitors: the self-interested Agent~B1 (whose objective is to maximize its own PnL), the competitive Agent~B2 (whose objective is to minimize the PnL of its opponent), and the hybrid Agent~B$^\star$, which can modulate between the behavior of the other two. To analyze how these agents shape the behavior of each other and affect market outcomes, we propose interaction-level metrics that quantify behavioral asymmetry and system-level dynamics, while providing signals potentially indicative of emergent interaction patterns. Experimental results show that Agent~B2 secures dominant performance in a zero-sum setting against B1, aggressively capturing order flow while tightening average spreads, thus improving market execution efficiency. In contrast, Agent~B$^\star$ exhibits a self-interested inclination when co-existing with other profit-seeking agents, securing dominant market share through adaptive quoting, yet exerting a milder adverse impact on the rewards of Agents~A and B1 compared to B2. These findings suggest that adaptive incentive control supports more sustainable strategic co-existence in heterogeneous agent environments and offers a structured lens for evaluating behavioral design in algorithmic trading systems.

arXiv:2510.25992v1 Announce Type: cross Abstract: Large Language Models (LLMs) often struggle with problems that require multi-step reasoning. For small-scale open-source models, Reinforcement Learning with Verifiable Rewards (RLVR) fails when correct solutions are rarely sampled even after many attempts, while Supervised Fine-Tuning (SFT) tends to overfit long demonstrations through rigid token-by-token imitation. To address this gap, we propose Supervised Reinforcement Learning (SRL), a framework that reformulates problem solving as generating a sequence of logical "actions". SRL trains the model to generate an internal reasoning monologue before committing to each action. It provides smoother rewards based on the similarity between the model's actions and expert actions extracted from the SFT dataset in a step-wise manner. This supervision offers richer learning signals even when all rollouts are incorrect, while encouraging flexible reasoning guided by expert demonstrations. As a result, SRL enables small models to learn challenging problems previously unlearnable by SFT or RLVR. Moreover, initializing training with SRL before refining with RLVR yields the strongest overall performance. Beyond reasoning benchmarks, SRL generalizes effectively to agentic software engineering tasks, establishing it as a robust and versatile training framework for reasoning-oriented LLMs.

arXiv:2510.26007v1 Announce Type: cross Abstract: The development of Artificial Intelligence (AI), including AI in Science (AIS), should be done following the principles of responsible AI. Progress in responsible AI is often quantified through evaluation metrics, yet there has been less work on assessing the robustness and reliability of the metrics themselves. We reflect on prior work that examines the robustness of fairness metrics for recommender systems as a type of AI application and summarise their key takeaways into a set of non-exhaustive guidelines for developing reliable metrics of responsible AI. Our guidelines apply to a broad spectrum of AI applications, including AIS.

arXiv:2510.26151v1 Announce Type: cross Abstract: Large annotated datasets are essential for training robust Computer-Aided Diagnosis (CAD) models for breast cancer detection or risk prediction. However, acquiring such datasets with fine-detailed annotation is both costly and time-consuming. Vision-Language Models (VLMs), such as CLIP, which are pre-trained on large image-text pairs, offer a promising solution by enhancing robustness and data efficiency in medical imaging tasks. This paper introduces a novel Multi-View Mammography and Language Model for breast cancer classification and risk prediction, trained on a dataset of paired mammogram images and synthetic radiology reports. Our MV-MLM leverages multi-view supervision to learn rich representations from extensive radiology data by employing cross-modal self-supervision across image-text pairs. This includes multiple views and the corresponding pseudo-radiology reports. We propose a novel joint visual-textual learning strategy to enhance generalization and accuracy performance over different data types and tasks to distinguish breast tissues or cancer characteristics(calcification, mass) and utilize these patterns to understand mammography images and predict cancer risk. We evaluated our method on both private and publicly available datasets, demonstrating that the proposed model achieves state-of-the-art performance in three classification tasks: (1) malignancy classification, (2) subtype classification, and (3) image-based cancer risk prediction. Furthermore, the model exhibits strong data efficiency, outperforming existing fully supervised or VLM baselines while trained on synthetic text reports and without the need for actual radiology reports.

arXiv:2510.26188v1 Announce Type: cross Abstract: Reducing preventable hospital readmissions is a national priority for payers, providers, and policymakers seeking to improve health care and lower costs. The rate of readmission is being used as a benchmark to determine the quality of healthcare provided by the hospitals. In thisproject, we have used machine learning techniques like Logistic Regression, Random Forest and Support Vector Machines to analyze the health claims data and identify demographic and medical factors that play a crucial role in predicting all-cause readmissions. As the health claims data is high dimensional, we have used Principal Component Analysis as a dimension reduction technique and used the results for building regression models. We compared and evaluated these models based on the Area Under Curve (AUC) metric. Random Forest model gave the highest performance followed by Logistic Regression and Support Vector Machine models. These models can be used to identify the crucial factors causing readmissions and help identify patients to focus on to reduce the chances of readmission, ultimately bringing down the cost and increasing the quality of healthcare provided to the patients.

arXiv:2510.26275v1 Announce Type: cross Abstract: Generative AI (GenAI) is rapidly transforming software engineering (SE) practices, influencing how SE processes are executed, as well as how software systems are developed, operated, and evolved. This paper applies design science research to build a roadmap for GenAI-augmented SE. The process consists of three cycles that incrementally integrate multiple sources of evidence, including collaborative discussions from the FSE 2025 "Software Engineering 2030" workshop, rapid literature reviews, and external feedback sessions involving peers. McLuhan's tetrads were used as a conceptual instrument to systematically capture the transforming effects of GenAI on SE processes and software products.The resulting roadmap identifies four fundamental forms of GenAI augmentation in SE and systematically characterizes their related research challenges and opportunities. These insights are then consolidated into a set of future research directions. By grounding the roadmap in a rigorous multi-cycle process and cross-validating it among independent author teams and peers, the study provides a transparent and reproducible foundation for analyzing how GenAI affects SE processes, methods and tools, and for framing future research within this rapidly evolving area. Based on these findings, the article finally makes ten predictions for SE in the year 2030.

arXiv:2510.26336v1 Announce Type: cross Abstract: Large Language Models (LLMs) excel at general tasks but underperform in specialized domains like economics and psychology, which require deep, principled understanding. To address this, we introduce ACER (Automated Curriculum-Enhanced Regimen) that transforms generalist models into domain experts without sacrificing their broad capabilities. ACER first synthesizes a comprehensive, textbook-style curriculum by generating a table of contents for a subject and then creating question-answer (QA) pairs guided by Bloom's taxonomy. This ensures systematic topic coverage and progressively increasing difficulty. The resulting synthetic corpus is used for continual pretraining with an interleaved curriculum schedule, aligning learning across both content and cognitive dimensions. Experiments with Llama 3.2 (1B and 3B) show significant gains in specialized MMLU subsets. In challenging domains like microeconomics, where baselines struggle, ACER boosts accuracy by 5 percentage points. Across all target domains, we observe a consistent macro-average improvement of 3 percentage points. Notably, ACER not only prevents catastrophic forgetting but also facilitates positive cross-domain knowledge transfer, improving performance on non-target domains by 0.7 points. Beyond MMLU, ACER enhances performance on knowledge-intensive benchmarks like ARC and GPQA by over 2 absolute points, while maintaining stable performance on general reasoning tasks. Our results demonstrate that ACER offers a scalable and effective recipe for closing critical domain gaps in LLMs.