Brief

Researchers have developed a novel method to imbue large language models with specific personality traits without requiring model retraining. This approach involves identifying key neurons within the LLM that correlate with personality dimensions, based on the Big Five personality traits framework. By manipulating these identified neurons, the system can induce desired personality characteristics in the model's output, demonstrating comparable effectiveness to fine-tuned models but with greater efficiency and flexibility. AI

Researchers have developed SpAArSIST, an optimized version of the AASIST model for anti-spoofing in audio. This new configuration reduces computational requirements by over 20% and model size by 4%, while significantly improving out-of-domain robustness. The system also introduces a composite score to aid in selecting models for deployment based on accuracy, calibration, and compute efficiency. AI

A new study published on arXiv reveals that multilingual large language models exhibit biases in mental health evaluations based on prompt language. Researchers found that prompts in Chinese elicited higher stigma scores and more conservative depression severity judgments compared to equivalent prompts in English when using models like GPT-4o and Qwen3-32B. This suggests that LLMs do not apply consistent evaluative standards across languages in sensitive domains, potentially leading to under-estimation errors in mental health assessments. AI

Researchers have developed Fanar-Sadiq, a multi-agent system designed for accurate and grounded Islamic question answering. This bilingual Arabic-English platform addresses the limitations of standard LLMs in religious contexts by incorporating specialized modules for diverse query types. It supports retrieval-augmented generation for jurisprudential answers, exact scripture lookup, and precise calculations for zakat and inheritance, with a focus on verification and canonical grounding. AI

A new research paper explores how concept drift affects machine learning models used for detecting phishing emails. The study aims to evaluate the performance degradation of these systems as phishing tactics evolve and to propose mitigation strategies. The paper highlights the increasing sophistication of phishing attacks and the critical role of email spam filters in protecting users. AI

Researchers have developed a new metric called "target alignment" to theoretically understand why predictive coding (PC) is more sample-efficient than backpropagation (BP) in neural networks. Their analysis, particularly in deep linear networks, shows that PC learning is more efficient, especially in deep, narrow, and pre-trained models. The study provides analytical expressions and experimental validation, offering insights into optimizing PC for effective learning. AI

Researchers have developed a new reinforcement learning framework called OMatG-IRL for generating crystalline materials. This method allows for the incorporation of target properties into the generative process without needing to compute the score, a limitation of previous approaches. OMatG-IRL operates directly on learned velocity fields, enabling efficient exploration and policy-gradient estimation at inference time. The framework has demonstrated competitive performance in crystal structure prediction, achieving significant improvements in sampling efficiency and generation time. AI

Researchers have developed a Lie-algebraic framework to analyze the expressivity and error bounds of parallelizable sequence models like Transformers. Their theory establishes a direct link between a model's depth and its expressivity, showing that increasing depth exponentially reduces approximation error. This theoretical insight was validated through experiments on symbolic and continuous-valued state-tracking tasks, confirming the empirical performance of deep sequence models. AI

Researchers have developed a new method for detecting information leakage in machine learning models without requiring access to training data or code. The technique analyzes only the model's predictions and outcomes to identify contamination. This approach categorizes leakage into three types: miscalibrated, broad-calibrated, and deterministic, with specific tests designed for each, offering a way to assess reproducibility in ML-based science. AI

Researchers have developed LakeFM, a new foundation model designed to understand and forecast aquatic ecosystem dynamics. Unlike previous models, LakeFM can handle irregular time series data and generalize across lakes with varying characteristics. It was pre-trained on a large dataset of simulated and observed lakes, demonstrating strong forecasting performance and the ability to produce physically plausible predictions. AI

Researchers have developed a new method for density estimation, extending the minimum-distance estimator approach to Hellinger distance. This technique allows for the creation of near-linear time algorithms with near-optimal sample complexity for learning classes of densities, including univariate mixtures of log-concave densities and mixtures of Gaussians. AI

Researchers have developed a new method to probe and influence the cultural values embedded within large language models. This approach uses scenario-based dilemmas, translating survey questions into behavioral choices to reveal implicit model preferences rather than relying on direct, often safety-aligned, responses. The study found that interventions to steer cultural values can lead to shifts along multiple dimensions simultaneously, similar to human behavior, and that this entanglement persists across different steering techniques without significantly degrading general task performance. AI

Researchers have developed FlexiBrain, a novel framework for processing fMRI data that is agnostic to spatial and temporal resolution variations. This approach utilizes a Mamba-JEPA backbone and dynamic patch resizing to avoid destructive standardization, preserving subject-specific anatomical information. FlexiBrain has demonstrated superior performance across five neuroscience tasks, outperforming existing methods by up to 12 percentage points and significantly reducing preprocessing computational costs. AI

Researchers have developed a new method for optimizing Mamba-2 inference, focusing on compiler-first state space duality. This approach enables portable autoregressive caching with $O(1)$ complexity, eliminating the need for custom CUDA or Triton kernels. The resulting single-source inference path, implemented in JAX, demonstrates significant speedups on Google Cloud TPUs and NVIDIA GPUs, achieving high hardware utilization and matching reference perplexity scores. AI

Researchers have developed LatticeBridge, a novel method for structured sequence generation that addresses the challenge of satisfying multiple input-derived constraints within a single output. This approach frames the problem as a rare-event sequential inference task, combining a prefix language model with instance-compiled surface automata and a specialized Monte Carlo decoder. LatticeBridge aims to improve the faithfulness of generated sequences by ensuring all required anchors are jointly realized, outperforming baseline methods on benchmarks like CommonGen and WikiBio. AI

Researchers have identified a specific failure mode in semantic segmentation models, termed 'semantic label flips,' where models correctly identify object boundaries but assign incorrect semantic labels to foreground pixels. This issue is exacerbated by correlation shifts between training and testing data, particularly when non-causal features are strongly tied to labels. The study proposes a new metric, 'Flip,' to quantify these within-object label swaps and an entropy-based 'flip-risk' score to detect such cases during inference. AI

Researchers have developed a new unsupervised framework to adapt vision-language models (VLMs) for more comprehensive multi-label image recognition. The method addresses the tendency of VLMs to focus on a single iconic object, thereby missing other relevant labels in an image. By employing "cutting" and "sewing" stages, the framework enhances the model's ability to identify multiple objects and adjust label distributions without requiring manual annotations. Experiments show this approach significantly outperforms existing unsupervised methods and even some weakly supervised baselines. AI

Researchers have introduced WildCross, a new benchmark designed to evaluate robotic perception systems in natural environments. The benchmark includes over 476,000 RGB frames with depth and surface normal annotations, along with pose and lidar data. This work expands on previous results, focusing on metric depth estimation to highlight the limitations of current vision models trained primarily on urban data. AI

Researchers have developed VL-DINO, a new object detection model that effectively integrates knowledge from CLIP, a vision-language model. The model uses novel modules to construct better training samples and fuse visual and textual information. In zero-shot tests on the LVIS benchmark, VL-DINO achieved state-of-the-art results, outperforming previous methods. AI

Researchers have developed a framework called 3D-CBM to enhance interpretability in 3D generative models by integrating Concept Bottleneck Models. This approach aims to bridge the semantic gap in deep geometric learning by aligning latent representations with human-defined concepts. The framework has demonstrated effectiveness in a proof-of-concept experiment, achieving high accuracy in concept prediction and enabling precise interventions for error correction in 3D models. AI

Researchers have introduced PCS-UQ, a new framework for uncertainty quantification in machine learning, designed to enhance trustworthiness in high-stakes applications. The framework integrates principles of predictability, computability, and stability to screen models and capture variability. PCS-UQ has demonstrated strong performance on various benchmarks, outperforming existing conformal methods in interval width and subgroup coverage, with efficient variants proposed for deep learning applications. AI

Researchers have developed a new knowledge distillation framework that uses Large Language Models (LLMs) to extract physics principles from scientific literature. This framework creates a 'teacher' model that imbues a 'student' model with predictive capabilities for manufacturing processes, even with limited data. The resulting student model is lightweight, capable of high-frequency inference for real-time deployment, and shows robustness even when the LLM-derived physics knowledge is imperfect. AI

Researchers have introduced Afrispeech Semantics, a new benchmark designed to evaluate the audio semantic reasoning capabilities of spoken language models. The benchmark focuses on five distinct tasks: entailment, consistency, plausibility, accent drift, and accent restraint. This evaluation aims to uncover critical limitations in current audio reasoning assessments and guide the development of more robust and equitable audio language models, particularly concerning accent variation and domain shifts. AI

A new paper from the AIED community highlights the significant environmental costs associated with large language models (LLMs). Researchers found that while many AIED projects utilize LLMs, very few report the computational resources consumed or discuss the environmental impacts as an ethical concern. To address this, the paper proposes an open-source method and software tools for systematically measuring and reporting the carbon footprint of LLMs in AIED systems, aiming to encourage more transparent reporting of these hidden costs. AI

Researchers have developed a theoretical framework to unify the understanding of learning dynamics and generalization in transformer models. This work formalizes transformer training as an ordinary differential equation system, approximating it to kernel behaviors. The analysis reveals a two-stage scaling law for generalization error, with an initial exponential decay followed by a power-law decay after a resource threshold is met, proving this two-stage law to be tight. AI

Researchers have developed a new benchmark called the Moral Trolley Arena to evaluate how large language models compose moral judgments. This benchmark assesses models' ability to combine multiple moral signals within a single scenario, moving beyond simple preference rankings of isolated acts. Across ten frontier models, the study found that composite moral judgments are largely predictable by the strength of individual acts but are consistently compressed rather than simply additive, indicating complex moral reasoning processes in LLMs. AI

Researchers have developed MindHier, a novel framework for reconstructing images from fMRI data that moves beyond diffusion models. This new approach utilizes a scale-wise autoregressive method, incorporating a hierarchical fMRI encoder and a layer-wise alignment scheme with CLIP features. MindHier aims to mimic human visual perception by synthesizing global semantics before refining local details, resulting in faster inference times and more deterministic outputs compared to existing diffusion-based methods. AI

Researchers have developed T2MM, a novel architecture that integrates Large Language Models with multimodal capabilities to assist in science learning and model construction. Unlike static image generation, T2MM creates interactive models that respond to user adjustments within the Virtual Experimental Research Assistant (VERA) software. Technical feasibility was demonstrated using a custom dataset, where T2MM outperformed a baseline code-generation approach across all metrics. AI

A new research paper introduces the concept of a "structural attention tax" in retrieval-augmented generation (RAG) systems. The study found that the format of retrieved information, particularly knowledge graph triples, can disproportionately capture the model's attention compared to semantically equivalent natural language text. This phenomenon can reduce the effectiveness of in-context learning by up to 42%, regardless of the content's relevance. The research proposes a framework to decouple semantic and structural components of attention, suggesting strategies to mitigate this tax by optimizing retrieval quality and reducing format-driven attention capture. AI

Researchers have developed Lung-R1, a novel large language model designed for pulmonary disease diagnosis. This model is guided by LungKG, a comprehensive knowledge graph containing over 59,000 nodes and 164,000 edges related to pulmonary medicine. Lung-R1 demonstrated state-of-the-art performance in a 20-system evaluation, particularly in EMR diagnosis, outperforming previous baselines. AI

Researchers have introduced TouchThinker, a new framework designed to enhance tactile commonsense reasoning for embodied agents. This system addresses limitations in existing datasets and representation methods by introducing a million-scale dataset, TouchThinker-1M, covering 415 objects and various scenarios. Additionally, it incorporates an action-aware modeling mechanism to improve the efficiency and semantic expressiveness of tactile representations, enabling better open-world generalization. AI

Researchers have introduced HERO, a novel framework for reinforcement learning agents designed to improve multi-turn decision-making. Unlike traditional methods that rely on terminal outcomes, HERO uses hindsight-enhanced self-distillation with next environment observations as localized feedback. This approach converts each observation into a compact turn-level diagnosis, providing actionable insights into the agent's actions. HERO has demonstrated improved task success and reduced unnecessary turns on benchmarks like TauBench and WebShop, particularly under limited training budgets where successful rollouts are infrequent. AI

Researchers have developed MoCA-Agent, a novel code agent designed for robust financial and numerical reasoning. This system breaks down questions into atomic claims, uses specialist agents to trade these claims, and synthesizes an executable Python program from verified evidence. MoCA-Agent demonstrates strong performance on various benchmarks, including financial, tabular, and multimodal chart reasoning, by aggregating evidence at the claim level for improved accuracy. AI

Researchers have developed a new framework using large language models (LLMs) to map scientific literature and identify cross-topic connections. This method was tested on a corpus of engineering articles from the Proceedings of the National Academy of Sciences, demonstrating its ability to produce semantically interpretable topics with strong quantitative performance. The LLM-based approach outperformed traditional topic modeling techniques in terms of topic diversity and overlap, achieving 75.9% accuracy in manual validation. AI

Researchers have introduced SemantiClean, a novel framework designed to extract structured semantic signals from e-commerce session data. This system prioritizes auditability and reproducibility over marginal predictive gains, organizing behavioral elements into a four-layer architecture. The framework utilizes an LLM-Integrated Semantic Inference Engine to ensure deterministic and reproducible outputs, with a focus on transparency and defensible decision trails. AI

Researchers have developed SoftMatcha 2, a novel algorithm designed for rapid and semantically flexible pattern matching across massive text datasets. This system can search through trillions of tokens in under a second, accommodating variations like substitutions, insertions, and deletions in queries. Its efficiency is achieved through dynamic corpus-aware pruning and a disk-aware design, outperforming existing methods on large corpora and demonstrating utility in identifying benchmark contamination and enhancing information retrieval. AI

Researchers have developed Trap$^2$, a new framework designed to prevent unauthorized model merging in AI. This architecture-agnostic system encodes protection directly into fine-tuned weights, degrading them when they are recomposed into unauthorized mixtures. Trap$^2$ aims to address a governance gap created by model hubs, ensuring that released weights remain effective for standalone use while undermining attempts to bypass safety alignments or licensing terms through merging. AI

A new survey paper examines Explainable AI (XAI) methods within Answer Set Programming (ASP), a symbolic AI approach. The paper categorizes different types of ASP explanations and maps them to user queries, evaluating the coverage provided by existing theories and tools. It also identifies current limitations and suggests future research directions in this area. AI

Researchers have developed MARIC, a novel multi-agent framework for image classification that enhances performance by treating the task as a collaborative reasoning process. This system employs an Outliner Agent to grasp the image's theme and generate prompts, followed by three Aspect Agents that extract detailed descriptions from different visual perspectives. A final Reasoning Agent then synthesizes these insights with a reflection step to produce a unified classification, outperforming traditional methods and monolithic vision-language models on diverse benchmarks. AI

Researchers have introduced RelayFormer, a novel framework designed to improve the localization of manipulated regions in images and videos. This unified approach addresses challenges related to resolution diversity and the separate handling of image and video data by existing methods. RelayFormer utilizes Global Local Relay (GLR) tokens and a relay-based attention mechanism to efficiently exchange contextual information while preserving fine-grained manipulation artifacts. AI

Researchers have developed MLaGA, a novel model designed to enhance Large Language Models' (LLMs) ability to process and reason over multimodal graphs. This system addresses the challenge of graphs containing diverse attribute types, such as text and images, which have been underexplored by existing LLM-based graph methods. MLaGA employs a structure-aware multimodal encoder and a multimodal instruction-tuning approach to integrate these varied attributes and graph structures into LLMs. AI

Researchers have developed a novel framework for tactile-only blind grasping using a dexterous robotic hand. Their approach utilizes a Real2Sim tactile calibration pipeline to create a digital-twin simulator that accurately reproduces real-world tactile signals. This is combined with a layout-aware tactile encoder that incorporates sensor-geometry priors and a Diffusion Policy trained on object-specific reinforcement learning experts in the simulator. The deployed policy achieved a 27% success rate on a physical robotic hand across 20 objects, without visual input. AI

Researchers have developed a new optimizer called Velocity-Regularized Adam (VRAdam) that uses physics-inspired principles to improve deep neural network training. Unlike existing methods like Adam, VRAdam incorporates a higher-order penalty on learning rates based on velocity, which helps to dampen oscillations and slow convergence when weight updates are large. This approach aims to achieve more stable and efficient training, with theoretical analysis supporting its operation at the edge of stability and derived convergence bounds. Benchmarks across image classification, language modeling, and generative modeling tasks show VRAdam outperforming standard optimizers like AdamW. AI

Researchers have introduced a novel framework for estimating precision and recall curves in generative models, moving beyond single scalar metrics. This approach frames the estimation as a binary classification problem, offering a more detailed analysis of model performance. The framework also provides a minimax upper bound on estimation risk and unifies several existing precision-recall metrics. AI

Researchers have introduced MobilityBench, a new benchmark designed to evaluate the performance of large language model (LLM) based route-planning agents in real-world mobility scenarios. The benchmark utilizes a large dataset of anonymized user queries from Amap, covering diverse routing needs across multiple cities. To ensure reproducibility, MobilityBench includes a deterministic API-replay sandbox and a multi-dimensional evaluation protocol that assesses outcome validity, instruction understanding, planning, tool use, and efficiency. Initial evaluations show current LLM agents are competent in basic information retrieval and route planning but struggle with preference-constrained planning, indicating a need for improvement in personalized mobility applications. AI

Researchers have introduced OMAD, a novel framework for online multi-agent reinforcement learning (MARL) that utilizes diffusion policies to enhance agent coordination. This approach addresses the challenge of intractable likelihoods in diffusion models, which typically hinder exploration in online MARL settings. OMAD employs a relaxed policy objective that maximizes scaled joint entropy and a joint distributional value function for decentralized policy optimization, leading to significant improvements in sample efficiency. AI

Researchers have developed a new technique called Neural FOXP2 to improve the performance of large language models in non-English languages. This method works by identifying and steering "language neurons" within the model, which are responsible for controlling language defaultness. The process involves localizing these neurons, defining steering directions, and then applying targeted activation shifts to make languages like Hindi or Spanish primary, thereby reducing English dominance. AI

Researchers have developed a new method for recognizing emotions in conversations by analyzing conversational context and discourse markers. The study found that conversational history, particularly the preceding 10-30 turns, is the most significant factor in emotion recognition, with performance plateauing quickly. Hierarchical sentence representations were beneficial in utterance-only settings but less so when conversational history was available. The research also identified a correlation between specific emotions and the position of discourse markers, suggesting that emotions like sadness are more context-dependent. AI

Researchers have developed a new pipeline for detecting AI-generated content on social media, utilizing a compact vision-language model. This approach addresses limitations of existing methods by improving generalization to new AI models, incorporating multi-modal data, and providing interpretable explanations. The model achieves state-of-the-art performance on public benchmarks and has shown positive impacts on user engagement when deployed for post recommendation on social media platforms. AI

Researchers have introduced BioDivergence, a new framework designed to evaluate how well AI models can distinguish between contextual contradictions and genuine disagreements in biomedical research abstracts. This framework moves beyond simple entailment or contradiction classifications to capture the nuanced reasons behind conflicting findings, such as differences in study populations or methodologies. BioDivergence includes a six-class conflict taxonomy and a 13-axis divergence ontology, along with a silver benchmark dataset of over 11,000 claim pairs to test model performance. AI