arXiv:2511.04698v1 Announce Type: cross Abstract: The early detection of mental health disorders from social media text is critical for enabling timely support, risk assessment, and referral to appropriate resources. This work introduces multiMentalRoBERTa, a fine-tuned RoBERTa model designed for multiclass classification of common mental health conditions, including stress, anxiety, depression, post-traumatic stress disorder (PTSD), suicidal ideation, and neutral discourse. Drawing on multiple curated datasets, data exploration is conducted to analyze class overlaps, revealing strong correlations between depression and suicidal ideation as well as anxiety and PTSD, while stress emerges as a broad, overlapping category. Comparative experiments with traditional machine learning methods, domain-specific transformers, and prompting-based large language models demonstrate that multiMentalRoBERTa achieves superior performance, with macro F1-scores of 0.839 in the six-class setup and 0.870 in the five-class setup (excluding stress), outperforming both fine-tuned MentalBERT and baseline classifiers. Beyond predictive accuracy, explainability methods, including Layer Integrated Gradients and KeyBERT, are applied to identify lexical cues that drive classification, with a particular focus on distinguishing depression from suicidal ideation. The findings emphasize the effectiveness of fine-tuned transformers for reliable and interpretable detection in sensitive contexts, while also underscoring the importance of fairness, bias mitigation, and human-in-the-loop safety protocols. Overall, multiMentalRoBERTa is presented as a lightweight, robust, and deployable solution for enhancing support in mental health platforms.

arXiv:2511.05385v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) utilizes external knowledge to augment Large Language Models' (LLMs) reliability. For flexibility, agentic RAG employs autonomous, multi-round retrieval and reasoning to resolve queries. Although recent agentic RAG has improved via reinforcement learning, they often incur substantial token overhead from search and reasoning processes. This trade-off prioritizes accuracy over efficiency. To address this issue, this work proposes TeaRAG, a token-efficient agentic RAG framework capable of compressing both retrieval content and reasoning steps. 1) First, the retrieved content is compressed by augmenting chunk-based semantic retrieval with a graph retrieval using concise triplets. A knowledge association graph is then built from semantic similarity and co-occurrence. Finally, Personalized PageRank is leveraged to highlight key knowledge within this graph, reducing the number of tokens per retrieval. 2) Besides, to reduce reasoning steps, Iterative Process-aware Direct Preference Optimization (IP-DPO) is proposed. Specifically, our reward function evaluates the knowledge sufficiency by a knowledge matching mechanism, while penalizing excessive reasoning steps. This design can produce high-quality preference-pair datasets, supporting iterative DPO to improve reasoning conciseness. Across six datasets, TeaRAG improves the average Exact Match by 4% and 2% while reducing output tokens by 61% and 59% on Llama3-8B-Instruct and Qwen2.5-14B-Instruct, respectively. Code is available at https://github.com/Applied-Machine-Learning-Lab/TeaRAG.

arXiv:2510.22780v2 Announce Type: replace Abstract: AI agents are continually optimized for tasks related to human work, such as software engineering and professional writing, signaling a pressing trend with significant impacts on the human workforce. However, these agent developments have often not been grounded in a clear understanding of how humans execute work, to reveal what expertise agents possess and the roles they can play in diverse workflows. In this work, we study how agents do human work by presenting the first direct comparison of human and agent workers across multiple essential work-related skills: data analysis, engineering, computation, writing, and design. To better understand and compare heterogeneous computer-use activities of workers, we introduce a scalable toolkit to induce interpretable, structured workflows from either human or agent computer-use activities. Using such induced workflows, we compare how humans and agents perform the same tasks and find that: (1) While agents exhibit promise in their alignment to human workflows, they take an overwhelmingly programmatic approach across all work domains, even for open-ended, visually dependent tasks like design, creating a contrast with the UI-centric methods typically used by humans. (2) Agents produce work of inferior quality, yet often mask their deficiencies via data fabrication and misuse of advanced tools. (3) Nonetheless, agents deliver results 88.3% faster and cost 90.4-96.2% less than humans, highlighting the potential for enabling efficient collaboration by delegating easily programmable tasks to agents.

arXiv:2412.05447v3 Announce Type: replace-cross Abstract: Retrieval-Augmented Generation (RAG) is one of the leading and most widely used techniques for enhancing LLM retrieval capabilities, but it still faces significant limitations in commercial use cases. RAG primarily relies on the query-chunk text-to-text similarity in the embedding space for retrieval and can fail to capture deeper semantic relationships across chunks, is highly sensitive to chunking strategies, and is prone to hallucinations. To address these challenges, we propose TOBUGraph, a graph-based retrieval framework that first constructs the knowledge graph from unstructured data dynamically and automatically. Using LLMs, TOBUGraph extracts structured knowledge and diverse relationships among data, going beyond RAG's text-to-text similarity. Retrieval is achieved through graph traversal, leveraging the extracted relationships and structures to enhance retrieval accuracy, eliminating the need for chunking configurations while reducing hallucination. We demonstrate TOBUGraph's effectiveness in TOBU, a real-world application in production for personal memory organization and retrieval. Our evaluation using real user data demonstrates that TOBUGraph outperforms multiple RAG implementations in both precision and recall, significantly improving user experience through improved retrieval accuracy.

arXiv:2502.16395v3 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly used to automate data analysis through executable code generation. Yet, data science tasks often admit multiple statistically valid solutions, e.g. different modeling strategies, making it critical to understand the reasoning behind analyses, not just their outcomes. While manual review of LLM-generated code can help ensure statistical soundness, it is labor-intensive and requires expertise. A more scalable approach is to evaluate the underlying workflows-the logical plans guiding code generation. However, it remains unclear how to assess whether an LLM-generated workflow supports reproducible implementations. To address this, we present AIRepr, an Analyst-Inspector framework for automatically evaluating and improving the reproducibility of LLM-generated data analysis workflows. Our framework is grounded in statistical principles and supports scalable, automated assessment. We introduce two novel reproducibility-enhancing prompting strategies and benchmark them against standard prompting across 15 analyst-inspector LLM pairs and 1,032 tasks from three public benchmarks. Our findings show that workflows with higher reproducibility also yield more accurate analyses, and that reproducibility-enhancing prompts substantially improve both metrics. This work provides a foundation for transparent, reliable, and efficient human-AI collaboration in data science. Our code is publicly available.

arXiv:2506.09338v2 Announce Type: replace-cross Abstract: Process reward models (PRMs) play a central role in guiding inference-time scaling algorithms for large language models (LLMs). However, we observe that even state-of-the-art PRMs can be poorly calibrated. Specifically, they tend to overestimate the success probability that a partial reasoning step will lead to a correct final answer, particularly when smaller LLMs are used to complete the reasoning trajectory. To address this, we present a calibration approach -- performed via quantile regression -- that adjusts PRM outputs to better align with true success probabilities. Leveraging these calibrated success estimates and their associated confidence bounds, we introduce an \emph{instance-adaptive scaling} (IAS) framework that dynamically adjusts the compute budget based on the estimated likelihood that a partial reasoning trajectory will yield a correct final answer. Unlike conventional methods that allocate a fixed number of reasoning trajectories per query, this approach adapts to each instance and reasoning step when using our calibrated PRMs. Experiments on mathematical reasoning benchmarks show that (i) our PRM calibration method achieves small calibration error, outperforming the baseline methods, (ii) calibration is crucial for enabling effective IAS, and (iii) the proposed IAS strategy reduces inference costs while maintaining final answer accuracy, utilizing less compute on more confident problems as desired.

Nature Medicine, Published online: 10 November 2025; doi:10.1038/s41591-025-04065-z

Author Correction: Global burden of chikungunya virus infections and the potential benefit of vaccination campaigns

arXiv:2509.19088v3 Announce Type: replace-cross Abstract: Digital representations of individuals ("digital twins") promise to transform social science and decision-making. Yet it remains unclear whether such twins truly mirror the people they emulate. We conducted 19 preregistered studies with a representative U.S. panel and their digital twins, each constructed from rich individual-level data, enabling direct comparisons between human and twin behavior across a wide range of domains and stimuli (including never-seen-before ones). Twins reproduced individual responses with 75% accuracy and seemingly low correlation with human answers (approximately 0.2). However, this apparently high accuracy was no higher than that achieved by generic personas based on demographics only. In contrast, correlation improved when twins incorporated detailed personal information, even outperforming traditional machine learning benchmarks that require additional data. Twins exhibited systematic strengths and weaknesses - performing better in social and personality domains, but worse in political ones - and were more accurate for participants with higher education, higher income, and moderate political views and religious attendance. Together, these findings delineate both the promise and the current limits of digital twins: they capture some relative differences among individuals but not yet the unique judgments of specific people. All data and code are publicly available to support the further development and evaluation of digital twin pipelines.

January Ventures co-founder Jennifer Neundorfer discussed this AI-driven funding market on the Equity podcast during TechCrunch Disrupt.

arXiv:2510.15418v2 Announce Type: replace-cross Abstract: Retrieval-Augmented Generation systems are essential for providing fact-based guidance from Malaysian Clinical Practice Guidelines. However, their effectiveness with image-based queries is limited, as general Vision-Language Model captions often lack clinical specificity and factual grounding. This study proposes and validates a framework to specialize the MedGemma model for generating high-fidelity captions that serve as superior queries. To overcome data scarcity, we employ a knowledge distillation pipeline to create a synthetic dataset across dermatology, fundus, and chest radiography domains, and fine-tune MedGemma using the parameter-efficient QLoRA method. Performance was rigorously assessed through a dual framework measuring both classification accuracy and, via a novel application of the RAGAS framework, caption faithfulness, relevancy, and correctness. The fine-tuned model demonstrated substantial improvements in classification performance, while RAGAS evaluation confirmed significant gains in caption faithfulness and correctness, validating the models ability to produce reliable, factually grounded descriptions. This work establishes a robust pipeline for specializing medical VLMs and validates the resulting model as a high-quality query generator, laying the groundwork for enhancing multimodal RAG systems in evidence-based clinical decision support.

Nature Biotechnology, Published online: 07 November 2025; doi:10.1038/s41587-025-02941-0

Publisher Correction: Deep-learning-based virtual screening of antibacterial compounds

Background: In low- and middle-income countries, maternal, newborn, and child health face significant challenges due to infrastructure limitations, access disparities, and service delivery inefficiencies. The Sehatmandi mobile health (mHealth) app was deployed in 2018 to address these issues across 189 health facilities in Afghanistan's Bamyan and Badakhshan provinces. This app aims to enhance service provision through real-time data monitoring, improved accountability, and performance-based health system strengthening. Objective: This study aims to explore the enablers, challenges, and lessons learned for the sustainability of the Sehatmandi mHealth intervention from the perspective of key stakeholders to inform the future scaling of digital health tools in fragile and resource-constrained settings. Methods: A qualitative study was conducted between June and July 2024 involving 24 in-depth interviews with stakeholders, including health facility managers, administrators, and high-level decision-makers. Participants were selected using stratified purposive sampling to ensure diverse facility representation. Interviews were conducted in person or virtually by using a semistructured guide, recorded, transcribed and translated into English. Thematic content analysis was performed using NVivo version 11 software. Ethics approval was obtained, and informed consent was secured from all the participants. Results: Stakeholders reported that Sehatmandi improved health system responsiveness by enhancing performance monitoring, accountability, timely reporting, and data-driven decision-making. Offline data entry was identified as a critical feature, enabling data collection in remote areas without internet access and ensuring synchronization when connectivity was resumed. However, several barriers affected the long-term sustainability: poor internet connectivity, electricity shortages,inadequate technical support, and high staff turnover, which disrupted functionality and data quality. Training gaps and insufficient supervision further hampered consistent and effective use. Participants emphasized the need for structured capacity building, regular follow-up, and sustainable funding to maintain the intervention. Integration with national health information systems and alignment with broader digital health strategies were also seen as prerequisites for scaling and institutionalization. Conclusions: The Sehatmandi mHealth intervention demonstrated enhanced performance monitoring and accountability across health care facilities in Afghanistan’s conflict-affected settings. However, for digital health interventions to be sustainable and scalable in low- and middle-income countries, foundational investments in digital infrastructure, continuous training and monitoring, system-level integration, and long-term funding are essential. These findings provide actionable insights for governments, implementers, and donors aiming to strengthen health systems through digital innovation in fragile settings.

Background: Current evidence dissemination methods fall short of meeting clinical nurses’ needs, hindering the implementation of evidence-based nursing practice. Large language models (LLMs), with their advanced natural language processing capabilities, offer potential as innovative tools to facilitate evidence dissemination. However, general-purpose LLMs typically lack domain-specific knowledge, are insufficient to support effective evidence dissemination in clinical contexts. It is essential to develop artificial intelligence tools tailored to nurses’ needs and preferences to enhance evidence dissemination. Objective: The aim of this study is to identify the challenges and barriers clinical nurses face in disseminating evidence, examine their perspectives on the use of existing LLMs to support evidence dissemination, and explore their needs and preferences regarding an LLM-based nursing evidence question-answering system. Methods: This qualitative study used a combined method of focus group discussions and the nominal group technique (NGT). Using purposive sampling, nurses with diverse specialties, professional titles, and years of experience were recruited, resulting in a total of 22 clinical nurses who completed the entire study. A total of 2 focus group discussions were conducted online via Tencent Meeting between November and December 2024 to explore the challenges and barriers nurses face in disseminating evidence, as well as their perspectives on using existing LLMs to support evidence dissemination. The data were analyzed using qualitative content analysis following the approach of Graneheim and Lundman. Subsequently, the NGT was used between March and April 2025 to identify nurses’ needs and preferences for the system to be developed. To overcome geographical constraints and participants’ busy schedules, the NGT was conducted entirely online, using online questionnaires and WeChat groups. Overall, 2 rounds of voting were conducted to determine the priority ranking of the functionalities. Results: The focus group yielded 3 main themes and 7 subthemes. Three main themes were identified as (1) pathways for evidence dissemination among nurses, (2) barriers that hinder the effective dissemination of evidence, and (3) advantages and limitations of using LLMs to support evidence dissemination. The limitations of current LLMs served as the foundation for nurses’ subsequent reflections in the nominal group discussions on the desired functions of a newly developed LLM. The NGT sessions ultimately identified 9 desired functions. After prioritization, the top 3 ranked functions were evidence-based, high-quality question-answering, evidence source provision, and personalized evidence recommendation. Conclusions: The current evidence dissemination process faces multiple barriers. LLMs hold promise as innovative tools to support evidence dissemination, but require further refinement. Clinical nurses have identified key functional needs, guiding the development of LLMs specifically tailored to clinical nursing practice.

arXiv:2511.02997v1 Announce Type: new Abstract: As AI systems become more capable and widely deployed as agents, ensuring their safe operation becomes critical. AI control offers one approach to mitigating the risk from untrusted AI agents by monitoring their actions and intervening or auditing when necessary. Evaluating the safety of these protocols requires understanding both their effectiveness against current attacks and their robustness to adaptive adversaries. In this work, we systematically evaluate a range of control protocols in SHADE-Arena, a dataset of diverse agentic environments. First, we evaluate blue team protocols, including deferral to trusted models, resampling, and deferring on critical actions, against a default attack policy. We find that resampling for incrimination and deferring on critical actions perform best, increasing safety from 50% to 96%. We then iterate on red team strategies against these protocols and find that attack policies with additional affordances, such as knowledge of when resampling occurs or the ability to simulate monitors, can substantially improve attack success rates against our resampling strategy, decreasing safety to 17%. However, deferring on critical actions is highly robust to even our strongest red team strategies, demonstrating the importance of denying attack policies access to protocol internals.

arXiv:2511.03051v1 Announce Type: new Abstract: Evaluating large language models (LLMs) as judges is increasingly critical for building scalable and trustworthy evaluation pipelines. We present ScalingEval, a large-scale benchmarking study that systematically compares 36 LLMs, including GPT, Gemini, Claude, and Llama, across multiple product categories using a consensus-driven evaluation protocol. Our multi-agent framework aggregates pattern audits and issue codes into ground-truth labels via scalable majority voting, enabling reproducible comparison of LLM evaluators without human annotation. Applied to large-scale complementary-item recommendation, the benchmark reports four key findings: (i) Anthropic Claude 3.5 Sonnet achieves the highest decision confidence; (ii) Gemini 1.5 Pro offers the best overall performance across categories; (iii) GPT-4o provides the most favorable latency-accuracy-cost tradeoff; and (iv) GPT-OSS 20B leads among open-source models. Category-level analysis shows strong consensus in structured domains (Electronics, Sports) but persistent disagreement in lifestyle categories (Clothing, Food). These results establish ScalingEval as a reproducible benchmark and evaluation protocol for LLMs as judges, with actionable guidance on scaling, reliability, and model family tradeoffs.

arXiv:2511.03545v1 Announce Type: new Abstract: This paper presents a comprehensive theoretical investigation into the parameterized complexity of explanation problems in various machine learning (ML) models. Contrary to the prevalent black-box perception, our study focuses on models with transparent internal mechanisms. We address two principal types of explanation problems: abductive and contrastive, both in their local and global variants. Our analysis encompasses diverse ML models, including Decision Trees, Decision Sets, Decision Lists, Boolean Circuits, and ensembles thereof, each offering unique explanatory challenges. This research fills a significant gap in explainable AI (XAI) by providing a foundational understanding of the complexities of generating explanations for these models. This work provides insights vital for further research in the domain of XAI, contributing to the broader discourse on the necessity of transparency and accountability in AI systems.

arXiv:2511.02842v1 Announce Type: cross Abstract: Many organisations pursue digital transformation to enhance operational efficiency, reduce manual efforts, and optimise processes by automation and digital tools. To achieve this, a comprehensive understanding of their unique needs is required. However, traditional methods, such as expert interviews, while effective, face several challenges, including scheduling conflicts, resource constraints, inconsistency, etc. To tackle these issues, we investigate the use of a Large Language Model (LLM)-powered chatbot to acquire organisations' digital transformation needs. Specifically, the chatbot integrates workflow-based instruction with LLM's planning and reasoning capabilities, enabling it to function as a virtual expert and conduct interviews. We detail the chatbot's features and its implementation. Our preliminary evaluation indicates that the chatbot performs as designed, effectively following predefined workflows and supporting user interactions with areas for improvement. We conclude by discussing the implications of employing chatbots to elicit user information, emphasizing their potential and limitations.

arXiv:2511.02864v1 Announce Type: cross Abstract: AlphaEvolve is a generic evolutionary coding agent that combines the generative capabilities of LLMs with automated evaluation in an iterative evolutionary framework that proposes, tests, and refines algorithmic solutions to challenging scientific and practical problems. In this paper we showcase AlphaEvolve as a tool for autonomously discovering novel mathematical constructions and advancing our understanding of long-standing open problems. To demonstrate its breadth, we considered a list of 67 problems spanning mathematical analysis, combinatorics, geometry, and number theory. The system rediscovered the best known solutions in most of the cases and discovered improved solutions in several. In some instances, AlphaEvolve is also able to generalize results for a finite number of input values into a formula valid for all input values. Furthermore, we are able to combine this methodology with Deep Think and AlphaProof in a broader framework where the additional proof-assistants and reasoning systems provide automated proof generation and further mathematical insights. These results demonstrate that large language model-guided evolutionary search can autonomously discover mathematical constructions that complement human intuition, at times matching or even improving the best known results, highlighting the potential for significant new ways of interaction between mathematicians and AI systems. We present AlphaEvolve as a powerful new tool for mathematical discovery, capable of exploring vast search spaces to solve complex optimization problems at scale, often with significantly reduced requirements on preparation and computation time.

arXiv:2511.03113v1 Announce Type: cross Abstract: Computational antibody design holds immense promise for therapeutic discovery, yet existing generative models are fundamentally limited by two core challenges: (i) a lack of dynamical consistency, which yields physically implausible structures, and (ii) poor generalization due to data scarcity and structural bias. We introduce FP-AbDiff, the first antibody generator to enforce Fokker-Planck Equation (FPE) physics along the entire generative trajectory. Our method minimizes a novel FPE residual loss over the mixed manifold of CDR geometries (R^3 x SO(3)), compelling locally-learned denoising scores to assemble into a globally coherent probability flow. This physics-informed regularizer is synergistically integrated with deep biological priors within a state-of-the-art SE(3)-equivariant diffusion framework. Rigorous evaluation on the RAbD benchmark confirms that FP-AbDiff establishes a new state-of-the-art. In de novo CDR-H3 design, it achieves a mean Root Mean Square Deviation of 0.99 {\AA} when superposing on the variable region, a 25% improvement over the previous state-of-the-art model, AbX, and the highest reported Contact Amino Acid Recovery of 39.91%. This superiority is underscored in the more challenging six-CDR co-design task, where our model delivers consistently superior geometric precision, cutting the average full-chain Root Mean Square Deviation by ~15%, and crucially, achieves the highest full-chain Amino Acid Recovery on the functionally dominant CDR-H3 loop (45.67%). By aligning generative dynamics with physical laws, FP-AbDiff enhances robustness and generalizability, establishing a principled approach for physically faithful and functionally viable antibody design.

arXiv:2511.03214v1 Announce Type: cross Abstract: Large language models (LLMs) exhibit strong semantic understanding, yet struggle when user instructions involve ambiguous or conceptually misaligned terms. We propose the Language Graph Model (LGM) to enhance conceptual clarity by extracting meta-relations-inheritance, alias, and composition-from natural language. The model further employs a reflection mechanism to validate these meta-relations. Leveraging a Concept Iterative Retrieval Algorithm, these relations and related descriptions are dynamically supplied to the LLM, improving its ability to interpret concepts and generate accurate responses. Unlike conventional Retrieval-Augmented Generation (RAG) approaches that rely on extended context windows, our method enables large language models to process texts of any length without the need for truncation. Experiments on standard benchmarks demonstrate that the LGM consistently outperforms existing RAG baselines.