Purpose: The purpose of this study was to determine if an ensemble of multiple LLM agents could be used collectively to provide a more reliable assessment of a pixel-based AI triage tool than a single LLM. Methods: 29,766 non-contrast CT head exams from fourteen hospitals were processed by a commercial intracranial hemorrhage (ICH) AI detection tool. Radiology reports were analyzed by an ensemble of eight open-source LLM models and a HIPAA compliant internal version of GPT-4o using a single multi-shot prompt that assessed for presence of ICH. 1,726 examples were manually reviewed. Performance characteristics of the eight open-source models and consensus were compared to GPT-4o. Three ideal consensus LLM ensembles were tested for rating the performance of the triage tool. Results: The cohort consisted of 29,766 head CTs exam-report pairs. The highest AUC performance was achieved with llama3.3:70b and GPT-4o (AUC= 0.78). The average precision was highest for Llama3.3:70b and GPT-4o (AP=0.75 & 0.76). Llama3.3:70b had the highest F1 score (0.81) and recall (0.85), greater precision (0.78), specificity (0.72), and MCC (0.57). Using MCC (95% CI) the ideal combination of LLMs were: Full-9 Ensemble 0.571 (0.552-0.591), Top-3 Ensemble 0.558 (0.537-0.579), Consensus 0.556 (0.539-0.574), and GPT4o 0.522 (0.500-0.543). No statistically significant differences were observed between Top-3, Full-9, and Consensus (p > 0.05). Conclusion: An ensemble of medium to large sized open-source LLMs provides a more consistent and reliable method to derive a ground truth retrospective evaluation of a clinical AI triage tool over a single LLM alone.

Online social networks have transformed the ways in which political mobilization messages are disseminated, raising new questions about how peer influence operates at scale. Building on the landmark 61-million-person Facebook experiment \citep{bond201261}, we develop an agent-based simulation framework that integrates real U.S. Census demographic distributions, authentic Twitter network topology, and heterogeneous large language model (LLM) agents to examine the effect of mobilization messages on voter turnout. Each simulated agent is assigned demographic attributes, a personal political stance, and an LLM variant (\texttt{GPT-4.1}, \texttt{GPT-4.1-Mini}, or \texttt{GPT-4.1-Nano}) reflecting its political sophistication. Agents interact over realistic social network structures, receiving personalized feeds and dynamically updating their engagement behaviors and voting intentions. Experimental conditions replicate the informational and social mobilization treatments of the original Facebook study. Across scenarios, the simulator reproduces qualitative patterns observed in field experiments, including stronger mobilization effects under social message treatments and measurable peer spillovers. Our framework provides a controlled, reproducible environment for testing counterfactual designs and sensitivity analyses in political mobilization research, offering a bridge between high-validity field experiments and flexible computational modeling.\footnote{Code and data available at https://github.com/CausalMP/LLM-SocioPol}

Karl Marx once wrote that ``the human essence is the ensemble of social relations'', suggesting that individuals are not isolated entities but are fundamentally shaped by their interactions with other entities, within which contexts play a constitutive and essential role. With the advent of computers and artificial intelligence, these contexts are no longer limited to purely human--human interactions: human--machine interactions are included as well. Then a central question emerges: How can machines better understand our situations and purposes? To address this challenge, researchers have recently introduced the concept of context engineering. Although it is often regarded as a recent innovation of the agent era, we argue that related practices can be traced back more than twenty years. Since the early 1990s, the field has evolved through distinct historical phases, each shaped by the intelligence level of machines: from early human--computer interaction frameworks built around primitive computers, to today's human--agent interaction paradigms driven by intelligent agents, and potentially to human--level or superhuman intelligence in the future. In this paper, we situate context engineering, provide a systematic definition, outline its historical and conceptual landscape, and examine key design considerations for practice. By addressing these questions, we aim to offer a conceptual foundation for context engineering and sketch its promising future. This paper is a stepping stone for a broader community effort toward systematic context engineering in AI systems.

Data selection is a critical aspect of Reinforcement Learning with Verifiable Rewards (RLVR) for enhancing the reasoning capabilities of large language models (LLMs). Current data selection methods are largely heuristic-based, lacking theoretical guarantees and generalizability. This work proposes a theoretically-grounded approach using influence functions to estimate the contribution of each data point to the learning objective. To overcome the prohibitive computational cost of policy rollouts required for online influence estimation, we introduce an off-policy influence estimation method that efficiently approximates data influence using pre-collected offline trajectories. Furthermore, to manage the high-dimensional gradients of LLMs, we employ sparse random projection to reduce dimensionality and improve storage and computation efficiency. Leveraging these techniques, we develop \textbf{C}urriculum \textbf{R}L with \textbf{O}ff-\textbf{P}olicy \text{I}nfluence guidance (\textbf{CROPI}), a multi-stage RL framework that iteratively selects the most influential data for the current policy. Experiments on models up to 7B parameters demonstrate that CROPI significantly accelerates training. On a 1.5B model, it achieves a 2.66x step-level acceleration while using only 10\% of the data per stage compared to full-dataset training. Our results highlight the substantial potential of influence-based data selection for efficient RLVR.

Abstract--Human smuggling networks are complex and constantly evolving, making them difficult to analyze comprehensively. Legal case documents offer rich factual and procedural insights into these networks but are often long, unstructured, and filled with ambiguous or shifting references, posing significant challenges for automated knowledge graph (KG) construction. Existing methods either overlook coreference resolution or fail to scale beyond short text spans, leading to fragmented graphs and inconsistent entity linking. We propose LINK-KG, a modular framework that integrates a three-stage, LLM-guided coreference resolution pipeline with downstream KG extraction. At the core of our approach is a type-specific Prompt Cache, which consistently tracks and resolves references across document chunks, enabling clean and disambiguated narratives for structured knowledge graph construction from both short and long legal texts. LINK-KG reduces average node duplication by 45.21% and noisy nodes by 32.22% compared to baseline methods, resulting in cleaner and more coherent graph structures. Human smuggling networks represent highly adaptive and organized systems involving a web of actors, routes, vehicles, and intermediaries, often operating under the radar of restrictive immigration policies [1]. These networks exploit legal loopholes, adjust swiftly to enforcement changes, and frequently intersect with transnational criminal organizations. Effectively analyzing their structure and behavior is critical for informing policy, enhancing security, and preventing exploitation. However, much of the actionable insight remains embedded in lengthy, unstructured legal documents, such as court rulings, field reports, and case transcripts, making automated analysis both essential and challenging.

Large language models (LLMs) continue to advance, with an increasing number of domain-specific variants tailored for specialised tasks. However, these models often lack transparency and explainability, can be costly to fine-tune, require substantial prompt engineering, yield inconsistent results across domains, and impose significant adverse environmental impact due to their high computational demands. To address these challenges, we propose the Bayesian network LLM fusion (BNLF) framework, which integrates predictions from three LLMs, including FinBERT, RoBERTa, and BERTweet, through a probabilistic mechanism for sentiment analysis. BNLF performs late fusion by modelling the sentiment predictions from multiple LLMs as probabilistic nodes within a Bayesian network. Evaluated across three human-annotated financial corpora with distinct linguistic and contextual characteristics, BNLF demonstrates consistent gains of about six percent in accuracy over the baseline LLMs, underscoring its robustness to dataset variability and the effectiveness of probabilistic fusion for interpretable sentiment classification.

Large language models (LLMs) such as ChatGPT are increasingly integrated into high-stakes decision-making, yet little is known about their susceptibility to social influence. We conducted three preregistered conformity experiments with GPT-4o in a hiring context. In a baseline study, GPT consistently favored the same candidate (Profile C), reported moderate expertise (M = 3.01) and high certainty (M = 3.89), and rarely changed its choice. In Study 1 (GPT + 8), GPT faced unanimous opposition from eight simulated partners and almost always conformed (99.9%), reporting lower certainty and significantly elevated self-reported informational and normative conformity (p < .001). In Study 2 (GPT + 1), GPT interacted with a single partner and still conformed in 40.2% of disagreement trials, reporting less certainty and more normative conformity. Across studies, results demonstrate that GPT does not act as an independent observer but adapts to perceived social consensus. These findings highlight risks of treating LLMs as neutral decision aids and underline the need to elicit AI judgments prior to exposing them to human opinions.

We identify three critical gaps that hinder the na ıve integration of SD into RL systems: diminishing speedups at large batch sizes, drafter staleness under continual actor updates, and drafter-induced policy degradation. Among these stages, generation is consistently the dominant bottleneck (Zhong et al., 2025a). Classic policy optimization methods such as PPO (Schul-man et al., 2017; 2015) combine trajectory-level rewards A natural optimization to address this bottleneck is speculative decoding (SD) (Leviathan et al., 2023; Chen et al., SD has already been widely adopted in LLM serving systems (e.g., SGLang (Zheng et al., 2024), Among various SD variants, EAGLE-3 (Li et al., 2025) represents the current state of the art, achieving the Overview of SD in RL training and our proposed ReSpec system. Consequently, a single static SD configuration cannot provide reliable speedups across diverse RL workloads, as shown in Figure 3. (G2) Drafter staleness under continual actor updates. As the actor (i.e., target) model evolves with each policy update, a fixed drafter rapidly becomes misaligned with the actor Moreover, this variance increases with drafter staleness (G2), causing a higher ratio of impoverished trajectories.

Self-improvement has emerged as a mainstream paradigm for advancing the reasoning capabilities of large vision-language models (LVLMs), where models explore and learn from successful trajectories iteratively. However, we identify a critical issue during this process: the model excels at generating high-quality trajectories for simple queries (i.e., head data) but struggles with more complex ones (i.e., tail data). This leads to an imbalanced optimization that drives the model to prioritize simple reasoning skills, while hindering its ability to tackle more complex reasoning tasks. Over iterations, this imbalance becomes increasingly pronounced--a dynamic we term the "Matthew effect"--which ultimately hinders further model improvement and leads to performance bottlenecks. To counteract this challenge, we introduce four efficient strategies from two perspectives: distribution-reshaping and trajectory-resampling, to achieve head-tail re-balancing during the exploration-and-learning self-improvement process. Extensive experiments on Qwen2-VL-7B-Instruct and InternVL2.5-4B models across visual reasoning tasks demonstrate that our methods consistently improve visual reasoning capabilities, outperforming vanilla self-improvement by 3.86 points on average.

Identifying and addressing security issues during the early phase of the development lifecycle is critical for mitigating the long-term negative impacts on software systems. Code review serves as an effective practice that enables developers to check their teammates' code before integration into the codebase. To streamline the generation of review comments, various automated code review approaches have been proposed, where LLM-based methods have significantly advanced the capabilities of automated review generation. However, existing models primarily focus on general-purpose code review, their effectiveness in identifying and addressing security-related issues remains underexplored. Moreover, adapting existing code review approaches to target security issues faces substantial challenges, including data scarcity and inadequate evaluation metrics. To address these limitations, we propose SecureReviewer, a new approach designed for enhancing LLMs' ability to identify and resolve security-related issues during code review. Specifically, we first construct a dataset tailored for training and evaluating secure code review capabilities. Leveraging this dataset, we fine-tune LLMs to generate code review comments that can effectively identify security issues and provide fix suggestions with our proposed secure-aware fine-tuning strategy. To mitigate hallucination in LLMs and enhance the reliability of their outputs, we integrate the RAG technique, which grounds the generated comments in domain-specific security knowledge. Additionally, we introduce SecureBLEU, a new evaluation metric designed to assess the effectiveness of review comments in addressing security issues. Experimental results demonstrate that SecureReviewer outperforms state-of-the-art baselines in both security issue detection accuracy and the overall quality and practical utility of generated review comments.