Weekly Recap

A structured Retrieval-Augmented Generation (RAG) framework that leverages annotated mammography reports as in-context demonstrations, enabling joint named entity and relation extraction without task-specific fine-tuning is proposed and achieves performance comparable to fine-tuned BETO models in low-resource settings, while significantly reducing training and deployment overhead.

Inspired by Mixture-of-Experts (MoE) principles, Evolvable-Substrate HyperNEAT is partitioned into non-overlapping spatial segments, each assigned to a separately evolved specialist network, which forces evolution to discover features across the entire image.

Realtime-Venus, a proactive full-duplex interaction system with two separately trained 9B models: Realtime-Venus-Omni for audio-visual interaction and Realtime-Venus-Audio for spoken interaction, which leads the compared models on MMAU, Llama Questions, and Speech CMMLU.

The study provides new insights into molecular aspects of γ-herpesvirus replication, as well as antiviral drug development and targeting strategies, and comparative antiviral effects spanning the three respective herpesviruses.

It is demonstrated that proteolipid protein 1 (PLP1), a protein expressed on the surface of oligodendrocytes and Schwann cells, is co-expressed with PSMA in the CNS and PNS, and suggested a specific solution for PSMA CAR-Ts.

Realtime-Venus, a proactive full-duplex interaction system with two separately trained 9B models: Realtime-Venus-Omni for audio-visual interaction and Realtime-Venus-Audio for spoken interaction, which leads the compared models on MMAU, Llama Questions, and Speech CMMLU.

The study provides new insights into molecular aspects of γ-herpesvirus replication, as well as antiviral drug development and targeting strategies, and comparative antiviral effects spanning the three respective herpesviruses.

It is demonstrated that proteolipid protein 1 (PLP1), a protein expressed on the surface of oligodendrocytes and Schwann cells, is co-expressed with PSMA in the CNS and PNS, and suggested a specific solution for PSMA CAR-Ts.

Challenges such as data quality, algorithmic bias, lack of transparency, privacy, cybersecurity, regulatory uncertainty and the need for human oversight remain important.

StepAudio 3 Realtime, an audio-language foundation model organized around a continuous listen-converse-think-act loop, achieves dialogue and reasoning performance comparable to dedicated reasoning models while speaking in real time, and resolves the tension between deep deliberation and latency via Think-While-Speaking.

This work shows that specialist optimization implicitly selects from this latent trajectory space, and establishes a new view of specialist training: when gold reasoning is absent, tuning choices directly control the latent supervision passed to downstream models.

Machine learning and deep learning approaches may improve efficiency, reproducibility, and pattern recognition in several forensic workflows, particularly postmortem imaging, skeletal age estimation, mass spectrometry-based toxicology, biometric identification, digital pathology, and natural language processing of medicolegal reports.

The results suggest that fine-tuning is not just about how much a model changes, but how that change is spent, and that changing the accessible directions can qualitatively alter the outcome of fine-tuning.

A controlled cross-modality framework that instantiates the same task suite in a variety of modalities to test the Convergent Emergence Hypothesis and shows that paired-mapping ICL emerges across six modalities, surpasses controlled baselines, and has correlated per-task effects across five of them.

This study synthesizes the literature using a PRISMA-guided systematic review and bibliometric analysis of 135 Scopus-indexed publications published between 2007 and 2026 to provide directions for developing transparent, scalable, and consumer-centered AI personalization strategies in digital commerce.

This work presents RelateAnything, a 53M-parameter model taking an image and regions from any source and returning scored relations over a predicate vocabulary supplied at inference as strings, a 53M-parameter model taking an image and regions from any source and returning scored relations over a predicate vocabulary supplied at inference as strings.

This work introduces StepAudio 3 Music, a large-scale, long-form music generation model that supports explicit musical planning and open-domain text-controlled generation, and achieves the highest AudioBox Content Enjoyment, Content Usefulness, and Production Quality scores and the highest MuQ-MuLan similarity among the evaluated systems.

Continual Search is introduced, an iterative framework that nudges the judge, over successive turns, to keep searching for unresolved diagnostic evidence within multiple benchmark suites and model families, demonstrating that effective search supersedes raw model scale.

This paper proposes Generalized Agent Iteration (GAI), a formal framework that describes iterative policy improvement and RSI as two cases of a single learning paradigm that rests on the classical account, makes existing systems comparable, and provides a principled basis for analyzing and designing new ones.

This work proposes a physics-guided framework for improving single-image part-aware 3D generation with physically compatible geometry and stable connections, and introduces parameterized connectors at their contact surfaces.

Latent Interface Training (LIT), a framework-agnostic two-stage strategy that first establishes a spatial-goal-conditioned action prior without images, then constrains visual conditioning through a pose-supervised latent interface, improves overall LIBERO-Plus success while preserving or improving average LIBERO success.

A comparison of two state-of-the-art agentic AI systems, Claude Code (Anthropic) and Codex (OpenAI), tasked with autonomously executing a simple end-to-end gravitational wave data analysis pipeline on a shared computing infrastructure without human intervention finds that both systems achieved comparable detection efficiency and template bank size.

NaB–cannabinoid combinations elicit distinct biological outcomes in colorectal cancer cells that are strongly determined by cellular context and provide a mechanistic framework for the further development of boron–cannabinoid-based therapeutic approaches in colorectal cancer.

UFO is proposed, the first unified framework for omni-condition alignment simultaneous evaluation, and UFO-Bench is presented, a dedicated benchmark designed to holistically evaluate the performance of existing customization models under the diverse mutual interactions of textual and visual conditions.

Across reasoning, long-context understanding, and agentic tasks, SAS consistently outperforms trainable sparse attention baselines across attention budgets, with especially large gains under tight budgets, demonstrating more effective context ranking for downstream tasks.

ESRL is introduced, an architecture-aware framework that explicitly explores the expert-routing space of MoE models, and preserves high-confidence experts as anchors, and restricts stochastic routing to a plausible candidate pool, thereby retaining reliable computation paths.

This study introduces StepAudio 3 Gen, a general-purpose audio generation model that supports zero-shot text-to-speech (TTS), voice design, vocal generation, sound effects, music, vibe speech, and mixtures of multiple audio types within a unified framework.

The current frameworks surrounding device monitoring are discussed and what challenges are presented when incorporating model monitoring in the healthcare setting are highlighted.

This review synthesizes recent advances in AI-based AD diagnosis, tracing the evolution from traditional ML to DL and LLMs and particular emphasis is placed on the emerging role of LLMs in extracting disease-related information from speech and clinical narratives, integrating heterogeneous biomedical data sources, and enabling multimodal frameworks for AD assessment.

Time-dependent mGPS 2 status during first-line therapy was associated with shorter OS, but not PFS, and time-dependent modified EPCRC-defined cachexia status was not significantly associated with PFS or OS.

This Review compares land surface, Earth system, statistical, and machine learning model approaches that are used to project future fire activity across global and regional scales under multiple climate scenarios to show that different modelling approaches provide complementary strengths for distinct applications.

This work presents ZGCM-1, a fully open 7B dense foundation model trained from scratch with extreme data, system, and algorithmic efficiency, and establishes an AI-native R&D workflow where agent swarms autonomously manage cluster operations, data curation, and rapid diagnostic evaluation.

This work introduces Minimal Intervention Reinforcement Learning (MInTRL), which expands the exploration frontier through sparse, local interventions in otherwise on-policy rollouts, and establishes minimal intervention as an effective paradigm for enhancing on-policy RL.

NGU, a simple adaptive sampling method that keeps generating samples for a problem until one is correct, is introduced, a simple adaptive sampling method that naturally uses fewer samples to filter out easy problems and allocates more compute to solving harder problems.

Dynin-Robotics achieves competitive performance on LIBERO and zero-shot LIBERO-Plus, together with a 78.4% average success rate across four manipulation conditions on a Franka Research 3 robot.

It is demonstrated that a general-purpose agent can directly drive a physical robot throughout task execution without any task-specific or environment-specific training and Agent as Policy (AGP) is introduced, which places task planning and execution under the agent's control.

Abemaciclib as both first- and second-line therapy was associated with sustained treatment duration in Japanese patients, and particularly, first-line abemaciclib plus NSAI followed by second-line abemaciclib plus fulvestrant may be a clinically meaningful treatment option.

Whether current evidence supports using artificial intelligence to move beyond predicting response under a fixed regimen to guiding systemic treatment tailoring is examined; the evidence does not yet support changing an individual patient’s systemic treatment.

A hybrid quantum model for recovering the damping coefficient of an underdamped harmonic oscillator from sparse noisy observations is studied, showing that both data and physics terms are needed, that circuit-depth effects are regime dependent, and that the observed ranking is robust to the tested observation-noise and training-size changes.

Using standard health measures might allow follow-ups for comparing health gender-gap across the developmental phases of individuals accounting for urban-rural divide accentuated in developing countries and diverse population.

This pilot study supports the feasibility and acceptability of the e-predictD β-version, despite lower-than-expected app usability, and warrants confirmation in a randomized controlled trial.

Screening and treatment of anxiety and depressive symptoms in patients with RH can improve the effectiveness of antihypertensive therapy and optimize BP control.

A hybrid method with VGG-16 deep learning model and a hand-crafted feature extraction technique, namely Principal Component Analysis, is proposed for the classification of diabetic retinopathy (DR).

This work used baseline survey data from Kenyan WLHIV receiving routine HIV care participating in a cluster randomized clinical trial evaluating a reproductive health counselling intervention at 10 HIV clinics to assess the relationship between depression and/or anxiety and unmet need for contraception.

It is shown that all six estimators identify correct annotations as most informative, and are able to audit the quality of mixed text corpora, choosing the annotations that result in the best downstream forecasting results without the need for model training.

This work derives the generalized score matching objective on a convex subset of $\mathbb{R}^{d}$ constructively starting from Minimum Probability Flow (MPF) learning, and shows how classical score matching as well as domain-adapted variants for non-negative data arise naturally within the proposed framework.

A structured Retrieval-Augmented Generation (RAG) framework that leverages annotated mammography reports as in-context demonstrations, enabling joint named entity and relation extraction without task-specific fine-tuning is proposed and achieves performance comparable to fine-tuned BETO models in low-resource settings, while significantly reducing training and deployment overhead.

The case is made that barriers for new accelerators can be lowered by starting with a baseline AI accelerator, and adding minimal logic to support new operators demanded by new specialized domains, which leads to a versatile chip that can be manufactured at high volume and deployed for a range of popular applications.

eConv-TasNet is proposed, an efficient variant of Conv-TasNet that improves both effectiveness and efficiency without relying on resource-intensive modules, and achieves competitive performance compared with state-of-the-art methods while requiring significantly fewer parameters and lower inference cost.

MaP-WAM is introduced, a Memory-as-Plans framework that decomposes memory-dependent world-action modeling into memory-grounded planning and plan-conditioned execution, and uses long-term multimodal episodic context as planning-time evidence rather than repeatedly conditioning the executor on the full history.

Generative Verification (GenV) is introduced, which distills an offline Z3-equivalence oracle into a reference-free, continuous reference-equivalence score by repurposing the language model's native vocabulary space and theoretically proves that structural, verdict-only verification heuristics are mathematically bounded to chance-level detection on these deceptively valid traces.

UniH3, a novel framework that Unifies Hierarchical Homogeneity and Heterogeneity and Heterogeneity for all-in-one medical image restoration, is proposed and extensive experiments on two large-scale benchmarks demonstrate that UniH3 achieves state-of-the-art performance on both all-in-one and single-task medical image restoration.

ActMap is introduced, a white-box representation that compresses the generation-time hidden- state trajectory into a fixed tensor of temporal-statistic channels that preserves structure across transformer depth and pooled hidden coordinates that supports abstention, routing, and selective verification from a single generation, making it a practical primitive for scalable oversight of deployed models.

The Feature Recovery Module (FRM), a plug-and-play module that maps degraded encoder features to pristine-aligned representations while keeping the host frozen, is proposed, which improves scene-level detection, CLIP/SigLIP2 feature recovery, and all four object-level VLM tasks, with larger gains under more severe degradation.

This work proposes RDDMPI, a conditional residual diffusion framework that operates directly in residual space, and reformulates probabilistic imputation as a baseline-residual decomposition, where a pretrained model captures the dominant signal and a diffusion process models the residual uncertainty.

Results show that structured schema context and lightweight prompting can substantially reduce reliance on fine-tuning for scalable conversational access to KGs, though closing the remaining gap to fully fine-tuned approaches will likely require reducing dependence on curated question-query exemplars.

The transition from file storage to database management systems transformed stored data into managed resources into AI model management systems, and AI now faces an analogous transition from AI model storage to AI model management.

Modern VLMs already encode much of what is needed for specialized domains; the challenge is learning how to ask, suggests modern VLMs already encode much of what is needed for specialized domains; the challenge is learning how to ask.

This paper explicitly construct several admissible methods, including methods based on well-separated clusters and a non-binary version of single linkage, and shows that, in contrast to the flat clustering setting, the hierarchical analog of these axioms are jointly satisfiable.

This work proposes a fully analog readout architecture based on voltage-to-time conversion of the VMM output based on sensing the column voltage, which avoids current-mode summing and scaling circuitry, improving area and energy efficiency.

The regulatory sandboxes can be viewed as pedagogical environments for AI: dynamic spaces where alignment develops as a formative process, progressively shaping autonomous behaviors through interaction and cooperation in scenarios of increasing complexity.

This work proposes MomentUm SpEctral Clipping (Musec), which replaces Muon's spectral flattening with spectral clipping: rather than setting all singular values of the momentum matrix to approximately one, Musec clips singular values that exceed a threshold while preserving the underlying spectral structure of the momentum.

Non-language data can serve as a partial substitute for language data for the training objective of next-token prediction but does not reliably support broader linguistic generalization, and transfer from non-language data is less efficient than additional language data.

This work presents FreeFlow, a hierarchical transformer built without any flow-specific components, using instead a single feed-forward encoder--decoder, achieving state-of-the-art results on major benchmarks, including Sintel, KITTI-2015, and Spring.

Experimental results demonstrate that DRG-MAPPO achieves a state-of-the-art win rate of 87%, suggesting that the framework effectively balances relational modeling, interpretability, and optimization stability for cooperative air combat.

Generative Late-Interaction Embeddings (GLIE): k<<N vectors per page learned from the normalized centroids to serve as both a lightweight index and a basis for regenerating the page's full embedding set, with the decoder as its main design surface.

T1, a Mixture-of-Experts model of 122B total trained with reinforcement learning, operating a real shell in a cloud sandbox for up to 300+ tool-call turns per task, rewarded by executing each task's own verifier by executing each task's own verifier.

Native unified modelling is position as a promising path towards systems that perceive, reason and create within a fully end-to-end framework through SenseNova-U1.5, an 8B-MoT native unified multimodal model that understands, reasons about, and generates visual content within an encoder-free and VAE-free architecture.

Textual descriptions of reaction conditions are encoded by a fine-tuned language model trained jointly with Gaussian process surrogates, yielding task-adaptive representations within a multi-objective Bayesian optimisation loop.

This work addresses the adverse interactions between sparsity and data repetition, and presents evidence for the core mechanisms of overfitting and its potential remediation, and suggests promising avenues for future methods to reduce over-specialization in model parameters by disrupting memorization patterns.

TART is the first framework to generate guitar tablature with both fingering and expressive technique annotations directly from guitar audio, and is the first framework to generate guitar tablature with both fingering and expressive technique annotations directly from guitar audio.

The Logit Refiner is introduced, a lightweight autoregressive module that restores intra-scale dependencies by sequentially sampling tokens conditioned on frozen backbone features and generalizes to text-to-image generation, confirming that the mean-field bottleneck persists across VAR variants and is effectively alleviated by the method.

This work develops Generative Marketing Mix Modeling (GMMM) to estimate the causal effects of Generative Engine Optimization and Generative Engine Marketing and investigates the empirical performance of the proposed method using simulated answers to product recommendation in English and Japanese.

3D Point Splatting (3DPS), the first differentiable point renderer for radar, derived directly from the standard solid-angle form of the radar equation, is proposed.

The Explainability Assistant is introduced, an open-source conversational XAI system that leverages the function-calling capabilities of modern Large Language Models (LLMs) to overcome limitations and achieves 94% intent-parsing accuracy, supports flexible natural language interaction, and adapts to different ML problem types without task-specific fine-tuning.

This work shows that for every fixed rational discount factor ($\gamma\in(0,1)$), exact planning remains NP-hard, and introduces a randomized polynomial-time approximation scheme for every fixed look-ahead depth.

An amortized neural sampler that combines operator learning with flow methods for sampling, enabling efficient sampling across families of stochastic differential equations, and theoretically establishes the expressivity and resolution invariance of the framework.

Findings show that foundation models require CGM-specific adaptation for reliable forecasting and that dietary context provides clinically meaningful signals beyond CGM alone, especially during postprandial periods.

SIRF (Spec-Internalized Risk Foundation Model), which internalizes a platform's complex policies, synthesized without additional human annotation via EntiGraph, MAGA rewriting and account-level chain-of-thought (CoT), into the weights via continued pretraining (CPT), so rules are applied at high precision under an ultra-low-latency, verdict-only deployment.

The results indicate that neural network spectra may contain information about privacy leakage that is not fully captured by conventional measures of overfitting, motivating spectral analysis as a promising direction for scalable privacy auditing.

A branch-and-bound algorithm tailored to the resulting robust sparse portfolio problems is developed, together with a new pruning rule that can discard exponentially many candidate portfolios in a single step.

It is found that cache performance depends on transfer granularity, intermediate memory use, and when transfers enter the request schedule, not only on device bandwidth, not only on device bandwidth.

Comparing full-precision and quantized forward passes, and identifying two mechanisms that characterize pretrained quantization robustness, these mechanisms explain why quantization error that passes through numerous layers can still produce only small output changes, and are verified across models and quantization settings.

This work generates synthetic plankton imagery conditioned on taxonomy: a CLIP encoder is adapted on a large plankton corpus with a ranked contrastive objective extended to deep, ragged taxonomies, then frozen to condition a parameter-efficient diffusion transformer.

A quantum approach to spectral feature extraction from the density of states (DOS) of a problem-dependent Hamiltonian is developed, and applied to machine learning on signed graphs, and opens routes towards quantum applications in social network balance analysis, spin-glass studies, correlation clustering, and protein-interaction networks.

LoaDiff is introduced, a diffusion-based generative model for year-long, sub-hourly smart-meter load curves that generates realistic and diverse load profiles, achieves a favorable trade-off between generation quality and limited evidence of memorization, and preserves information useful for downstream energy applications.

A certification layer is developed that wraps any severity model unmodified, with distribution-free guarantees using this structure, and attaches identical validity and certifies, on the vulnerable road users, a model-independent floor on set width that no base model beats.

This study examines how synthetic and real historical training data should be combined for low-resource OCR, and finds that joint and sequential training yield broadly similar archival accuracy under the tested practical pipelines.

Equal parameter count evaluations show that modulating the primitive's wavefront is an effective and lightweight enhancement for hologram representations, and frequency domain analysis illustrates that CVQPG has successfully preserved the mid-to-high frequency band of natural images.

Inspired by Mixture-of-Experts (MoE) principles, Evolvable-Substrate HyperNEAT is partitioned into non-overlapping spatial segments, each assigned to a separately evolved specialist network, which forces evolution to discover features across the entire image.

The results show that inconsistency under uncertainty can be effectively learned, analyzed, and repaired through response-surface modeling, providing a scalable foundation for uncertainty-aware consistency management in CPS development.