探索全球前沿学术脉络

01.

medRxiv (Medicine) 2026-06-22 DOI: HASH:0925ebd1900fbd35ba66390c6d44ce5b

Body composition subphenotypes, cardiometabolic risk and incident outcomes: validation in the population-based NAKO and UK Biobank imaging cohorts

作者:

Grune↗Haueise↗von Itter↗M.-N↗Jung↗Bamberg↗Bibi↗Friedrich↗C. M↗Fromherz↗Kauczor↗H.-U↗…

Background Anthropometric measures do not adequately capture heterogeneity in body fat distribution and corresponding cardiometabolic risk, whereas magnetic resonance imaging (MRI) enables precise differentiation and quantification of adipose tissue compartments and ectopic fat. We aimed to validate previously derived MRI-based body composition subphenotypes and their cardiometabolic risk profiles in two independent European cohorts. Methods Using deep learning-based image analysis, we quantified bone marrow, visceral, subcutaneous, cardiac, renal sinus, hepatic, skeletal muscle, and pancreatic fat in the imaging substudies of two population-based cohorts: the German National Cohort (NAKO, N=29,314, age range 19-74 years) and the UK Biobank (N=36,109, age range 40-69 years). Body composition subphenotypes, previously identified by k-means clustering, were evaluated using a rigorous statistical cluster validation framework with method-based and results-based approaches. In NAKO, cross-sectional associations between subphenotypes and estimated cardiovascular disease risk scores were examined using linear regression. In UK Biobank, longitudinal associations between subphenotypes and incident cardiometabolic outcomes, ascertained through hospital record linkage, were analysed using Cox regression. Findings All five body composition subphenotypes were robustly validated across both cohorts, and showed distinct fat distribution patterns and cardiometabolic risk profiles: I "lean", II "average adiposity", III "bone and muscle adiposity", IV "hepato-abdominal adiposity", and V "general and pancreatic adiposity". Subphenotypes I-III showed progressive adipose tissue remodelling patterns likely reflecting ageing trajectories. The "hepato-abdominal adiposity" subphenotype showed highest risk of incident diabetes, whereas the "general and pancreatic adiposity" subphenotype showed highest overall cardiovascular disease burden and metabolic impairment. Interpretation MRI-derived body composition subphenotypes represent distinct fat distribution patterns that reflect ageing- and disease-related processes, which supports the potential of body composition phenotyping for improved cardiometabolic risk stratification and targeted prevention.

阅读与讨论 → 访问原文 →

02.

arXiv (CS.CV) 2026-06-24 DOI: arXiv:2606.24888

DiffusionBench: On Holistic Evaluation of Diffusion Transformers

作者:

Xingjian Leng↗Jaskirat Singh↗Zhanhao Liang↗Ethan Smith↗Martin Bell↗Aninda Saha↗Yuhui Yuan↗Liang Zheng↗

Diffusion transformer (DiT) research on image generation has converged to a single evaluation setup: class-conditional generation on ImageNet. While methods improve the FID and related metrics, it is increasingly unclear whether they reflect real progress in generative modeling. The natural alternative, i.e., text-to-image (T2I) generation, is perceived as too costly or inconvenient to train and evaluate and is often skipped. We argue that this perception no longer holds. We introduce NanoGen, a unified DiT training and evaluation framework. NanoGen matches state-of-the-art DiT baselines on ImageNet and, with 12 lines of configuration change, also trains competitive text-to-image models. It currently supports RAE, VAE, pixel-space, and MeanFlow diffusion methods under both ImageNet and T2I setups. Under NanoGen, training T2I requires comparable compute to ImageNet. After training 21 latent diffusion models with NanoGen, we observe that method ranking shows no strong correlation between ImageNet and T2I generation: Pearson correlation is between -0.377 and -0.580 across three metrics. This suggests that a method which improves class-conditional ImageNet FID may show no corresponding improvement on T2I, clearly indicating the necessity of evaluating DiTs on both tasks. To this end, we summarize ImageNet and text-to-image results, which yields DiffusionBench, a holistic benchmark for DiT research. We recommend reporting DiffusionBench in place of ImageNet alone: methods that improve DiffusionBench are more likely to reflect broader progress.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2605.16865

MixSD: Mixed Contextual Self-Distillation for Knowledge Injection

作者:

Jiarui Liu↗Lechen Zhang↗Yongjin Yang↗Yinghui He↗Yingheng Wang↗Weihao Xuan↗Zhijing Jin↗Mona Diab↗

Supervised fine-tuning (SFT) is widely used to inject new knowledge into language models, but it often degrades pretrained capabilities such as reasoning and general-domain performance. We argue this forgetting arises because fine-tuning targets from humans or external systems diverge from the model's autoregressive distribution, forcing the optimizer to imitate low-probability token sequences. To address this problem, we propose MixSD, a simple external-teacher-free method for distribution-aligned knowledge injection. Instead of training on fixed targets, MixSD constructs supervision dynamically by mixing tokens from two conditionals of the base model itself: an expert conditional that observes the injected fact in context, and a naive conditional that reflects the model's original prior. The resulting supervision sequences preserve the factual learning signal while remaining substantially closer to the base model's distribution. We evaluate MixSD on two synthetic corpora that we construct to study factual recall and arithmetic function acquisition in a controlled setting, together with established benchmarks for open-domain factual question answering and knowledge editing. Across multiple model scales and settings, MixSD consistently achieves a better memorization-retention trade-off compared to SFT and on-policy self distillation baselines, retaining up to 100% of the base model's held-out capability while maintaining near-perfect training accuracy, whereas standard SFT retains as little as 1%. We further show that MixSD produces substantially lower-NLL supervision targets under the base model and reduces harmful movement along Fisher-sensitive parameter directions. These results suggest that aligning supervision with the model's native generation distribution is a simple and effective principle for knowledge injection that mitigates catastrophic forgetting.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.15017

Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents

作者:

Sina Hajimiri↗Masih Aminbeidokhti↗Jose Dolz↗Ismail Ben Ayed↗Issam H. Laradji↗Spandana Gella↗Nicolas Gontier↗

Online web agents often augment a base actor with memory, workflow, or skill modules. These modules can improve performance, but they also consume test-time tokens, a cost rarely reported alongside the actor's inference cost. We study online augmentation, where this overhead is paid on every task, and re-evaluate its benefits under a fixed total inference budget. We compare AWM, ASI, and ReasoningBank with a token-matched vanilla baseline that uses the same budget for additional actor steps. Across three WebArena domains and three models, Gemini 3 Flash, GPT-5.4-mini, and Qwen 3.6-27B, the vanilla baseline matches or surpasses all three augmentation methods in aggregate success rate while often using fewer total tokens. We observe a similar trend on WorkArena-L1 with Qwen 3.6-27B, indicating that the effect extends to enterprise knowledge-work tasks. Our results suggest that skills and workflow memory can be useful in specific domains, but their apparent gains often vanish against a budget-matched actor. We further show that run-to-run variance materially affects outcomes and should be reported as a core evaluation criterion for online web agents.

阅读与讨论 → 访问原文 →

05.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.19489

Concept Flow Models: Anchoring Concept-Based Reasoning with Hierarchical Bottlenecks

作者:

Ya Wang↗Adrian Paschke↗

arXiv:2606.19489v1 Announce Type: cross Abstract: Concept Bottleneck Models (CBMs) enhance interpretability by projecting learned features into a human-understandable concept space. Recent approaches leverage vision-language models to generate concept embeddings, reducing the need for manual concept annotations. However, these models suffer from a critical limitation: as the number of concepts approaches the embedding dimension, information leakage increases, enabling the model to exploit spurious or semantically irrelevant correlations and undermining interpretability. In this work, we propose Concept Flow Models (CFMs), which replace the flat bottleneck with a hierarchical, concept-driven decision tree. Each internal node in the hierarchy focuses on a localized subset of discriminative concepts, progressively narrowing the prediction scope. Our framework constructs decision hierarchies from visual embeddings, distributes semantic concepts at each hierarchy level, and trains differentiable concept weights through probabilistic tree traversal. Extensive experiments on diverse benchmarks demonstrate that CFMs match the predictive performance of flat CBMs, while substantially mitigating information leakage by reducing effective concept usage. Furthermore, CFMs yield stepwise decision flows that enable transparent and auditable model reasoning with hierarchical class structures.

阅读与讨论 → 访问原文 →

06.

arXiv (CS.CL) 2026-06-18 DOI: arXiv:2505.23851

ASyMOB: Algebraic Symbolic Mathematical Operations Benchmark

作者:

Michael Shalyt↗Rotem Elimelech↗Ido Kaminer↗

Large language models (LLMs) are increasingly applied to symbolic mathematics, yet existing evaluations often conflate pattern memorization with genuine reasoning. To address this gap, we present ASyMOB, a high-resolution dataset of 35,368 validated symbolic math problems spanning integration, limits, differential equations, series, and hypergeometrics. Unlike prior benchmarks, ASyMOB systematically perturbs each seed problem using symbolic, numeric, and equivalence-preserving transformations, enabling a fine-grained assessment of generalization. Our evaluation reveals three key findings: (1) most models' performance collapses under minor perturbations, while top systems exhibit an apparent regime shift in robustness; (2) integrated code tools stabilize performance, particularly for weaker models; and (3) we identify examples where Computer Algebra Systems (CAS) fail while LLMs succeed, as well as problems solved only via a hybrid LLM-CAS approach, highlighting a promising integration frontier. ASyMOB serves as a principled diagnostic tool for measuring and accelerating progress toward building verifiable, trustworthy AI for scientific discovery.

阅读与讨论 → 访问原文 →

07.

Nature Medicine 2026-06-25 DOI: HASH:87e5707d9de9b0049768e196e4b0a55c

<b>HPV vaccination linked to dramatic reduction in cervical cancer deaths</b>

作者:

Karen O’Leary↗

A new study examines the impact of the national HPV vaccination program in England, initiated almost two decades ago — showing huge declines in cervical cancer mortality among young women. A new study examines the impact of the national HPV vaccination program in England, initiated almost two decades ago — showing huge declines in cervical cancer mortality among young women.

阅读与讨论 → 访问原文 →

08.

arXiv (CS.LG) 2026-06-17 DOI: arXiv:2603.04198

Stable and Steerable Sparse Autoencoders with Weight Regularization

作者:

Piotr Jedryszek↗Oliver M. Crook↗

arXiv:2603.04198v2 Announce Type: replace-cross Abstract: Sparse autoencoders (SAEs) are widely used to extract human-interpretable features from neural network activations, but their learned features can vary substantially across random seeds and training choices. To improve stability, we studied weight regularization by adding L1 or L2 penalties on encoder and decoder weights, and evaluate how regularization interacts with common SAE training defaults. On MNIST, we observe that L2 weight regularization produces a core of highly aligned features and, when combined with tied initialization and unit-norm decoder constraints, it dramatically increases cross-seed feature consistency. For TopK SAEs trained on language model activations (Pythia-70M-deduped), adding a small L2 weight penalty increased the fraction of features shared across three random seeds and roughly doubles steering success rates, while leaving the mean of automated interpretability scores essentially unchanged. Finally, in the regularized setting, activation steering success becomes better predicted by auto-interpretability scores, suggesting that regularization can align text-based feature explanations with functional controllability.

阅读与讨论 → 访问原文 →

09.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.15055

Bridging Geographic Bias in Urban Streetscape Inference via Lifelong Learning with Visual-Semantic Pivoting

作者:

Xinze Zhang↗

Visual perception of urban streetscapes underpins evidence-based decisions in landscape planning, public health, and place-making. Yet models trained on a few well-photographed metropolises systematically misjudge underrepresented districts, propagating geographic bias into downstream policy. We address this gap with HVSP-LL, a lifelong learning framework that couples a stratified visual-semantic pivoting module with an equity-aware rehearsal mechanism. The pivoting module organises landscape concepts along a three-tier ontology (macro structure, meso composition, micro element) and aligns image features to learnable semantic anchors at each tier, providing transferable representations that resist distributional drift. The lifelong adaptation component sequentially absorbs new urban regions while constraining inter-region perception gaps through a worst-region sample-reweighting objective and a structurally-aware exemplar buffer. We evaluate HVSP-LL on a panoramic streetscape benchmark assembled from twelve cities across four continents and seven perceptual dimensions. The framework attains 0.834 Spearman correlation on the held-out city sequence, an absolute 6.1 point improvement over the strongest continual baseline, and shrinks the inter-city perception gap to 0.094 – a 38% reduction relative to the strongest continual baseline (0.151) and a 57% reduction relative to a representative regularisation baseline (0.218). Ablations confirm that each tier of the pivoting hierarchy contributes monotonically, and the equity-aware rehearsal converts mean backward transfer from -0.038 (without retention) to +0.013, eliminating catastrophic forgetting on the held-out sequence. Our results indicate that hierarchical anchoring is a practical pathway toward geographically equitable streetscape inference at city scale.

阅读与讨论 → 访问原文 →

10.

arXiv (CS.CV) 2026-06-11 DOI: arXiv:2605.29588

Brain-IT-VQA: From Brain Signals to Answers

作者:

Roman Beliy↗Matias Cosarinsky↗Oliver Heinimann↗Navve Wasserman↗Michal Irani↗

Decoding visual content from fMRI signals recorded while a person views images, and specifically answering questions about the seen images, is a long-standing challenge. While significant progress has been made in recent years in visual question answering (VQA) from fMRI, performance remains limited. Moreover, although recent models can make increasingly accurate predictions, they have rarely been used as tools for understanding the structure of visual representations in the brain. We present Brain-IT-VQA, a framework for visual question answering from fMRI. Building on the Brain Interaction Transformer (Brain-IT), our method decodes language tokens from brain activity and integrates them with a language model to answer visual questions. Our model substantially outperforms previous fMRI-based captioning and VQA approaches. We further introduce NSD-VQA, a new dataset and benchmark for visual question answering from fMRI. Unlike existing image-fMRI VQA datasets, which typically provide only a few broad and weakly controlled questions per image, NSD-VQA provides on average 20 question-answer pairs per image across 20 controlled question categories that disentangle multiple levels of visual understanding. This enables more reliable and interpretable evaluation despite limited fMRI test data. Together, Brain-IT-VQA and NSD-VQA provide both a strong predictive framework and a tool for studying brain representations. Using this benchmark, we quantify which forms of visual and semantic information can be reliably decoded from fMRI responses to natural images. We further analyze the contributions of different brain regions across question types.

阅读与讨论 → 访问原文 →

11.

arXiv (CS.CL) 2026-06-24 DOI: arXiv:2606.23937

When Retrieval Metrics Mislead: Measuring Policy Signal in Long-Horizon Tool-Use Agents

作者:

Tianyu Ding↗Juan Pablo De la Cruz Weinstein↗

Exact-match retrieval recall is often used as a proxy for whether a retriever supplies useful policy context to a downstream decision model. We test this proxy for pre-action policy classification in tau-bench using Qwen2.5-3B/7B classifiers. Under gold-policy conditioning, a compact structured state improves macro-F1 over raw trajectories by 0.13-0.17 after tuning. We then replace the benchmark-designated policy clause with the top-ranked clause retrieved from decision-time context. Although the exact governing clause is retrieved at rank 1 for only 7% of airline states, the primary 3B classifier obtains macro-F1 0.58 with retrieved clauses versus 0.60 with gold clauses (Delta=-0.02, task-cluster 95% CI [-0.23,+0.21]); mismatched-policy and no-policy controls score 0.32 and 0.21. We do not detect a macro-F1 difference between retrieved and gold clauses in this configuration, although the interval remains too wide to establish non-inferiority. The same qualitative pattern appears with a second retriever and at 7B, while varying across fine-tuning configurations. These results indicate that exact-match clause recall can underestimate downstream policy utility in this benchmark setting, motivating evaluation with retrieved policies in the classification loop rather than recall alone.

阅读与讨论 → 访问原文 →

12.

medRxiv (Medicine) 2026-06-12 DOI: HASH:520428f92613eef30c58a49466a58e51

Conversational Artificial Intelligence-Enabled Precision Oncology Reveals Context-Specific TGFβ and JAK/STAT Alterations in Pancreatic Cancer

作者:

Diaz↗F. C↗Waldrup↗Carranza↗F. G↗Manjarrez↗Velazquez-Villarreal↗

Background: Pancreatic ductal adenocarcinoma (PDAC) is characterized by extensive molecular complexity, profound stromal remodeling, and limited responsiveness to systemic therapies. Although gemcitabine-based regimens remain widely utilized, the molecular pathways that influence treatment-associated biological variation are incompletely understood. The TGF{beta} and JAK/STAT signaling networks are recognized regulators of tumor progression, immune modulation, and therapeutic resistance; however, their genomic architecture in clinically stratified PDAC populations remains poorly defined. Methods: We employed a conversational artificial intelligence-driven analytical framework to investigate TGF{beta} and JAK/STAT pathway alterations in a cohort of 184 PDAC patients. Clinical and molecular data were integrated to generate age- and treatment-stratified cohorts, enabling pathway-level and gene-level analyses according to gemcitabine exposure. Findings generated through AI-assisted interrogation were subsequently evaluated using conventional statistical approaches. Results: TGF{beta} pathway alterations were identified in approximately one-quarter to one-third of tumors across clinical subgroups and demonstrated relatively stable frequencies regardless of age at diagnosis or gemcitabine treatment status. Gene-level analyses revealed that pathway disruption was predominantly driven by recurrent alterations in SMAD4, with additional low-frequency events involving TGFBR1 and TGFBR2. Notably, TGFBR2 mutations were significantly more frequent among late-onset PDAC patients receiving gemcitabine compared with untreated late-onset patients (8.8% vs. 1.4%; p = 0.04), suggesting a potential treatment-associated enrichment. In contrast, JAK/STAT pathway alterations were rare throughout the cohort, with only isolated mutations observed in pathway components including JAK1, JAK2, JAK3, STAT1, STAT3, and related regulatory genes. No significant differences in JAK/STAT alteration frequencies were identified according to age or treatment exposure. Conclusions: TGF{beta} and JAK/STAT pathways exhibit distinct genomic architectures in PDAC. TGF{beta} pathway disruption represents a recurrent feature of disease biology, largely driven by SMAD4 alterations, while TGFBR2 enrichment in gemcitabine-treated late-onset tumors suggests a potential context-specific association worthy of further investigation. Conversely, genomic alterations within the JAK/STAT pathway are uncommon, indicating that pathway activity may be regulated predominantly through non-genomic mechanisms. These findings demonstrate the utility of conversational artificial intelligence agents for rapid, scalable, and clinically contextualized pathway interrogation and support future studies integrating multi-omic data to refine precision medicine strategies in PDAC.

阅读与讨论 → 访问原文 →

13.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2606.14078

Rethinking Backdoor Adversarial Unlearning through the Lens of Catastrophic Forgetting in Continual Learning

作者:

Zhenqian Zhu↗Yamin Hu↗Yujiang Liu↗Luping Wei↗Wenbo Hou↗Bin Li↗Haodong Li↗Wenjian Luo↗

arXiv:2606.14078v1 Announce Type: cross Abstract: Existing studies reveal that current backdoor defenses exhibit limited robustness and often fail against specific types of attacks. More concerningly, prevailing safety tuning strategies tend to provide only superficial safety protection, as they fall short of completely eliminating the backdoor effects. In this work, we present a novel formulation of backdoor learning and unlearning as a sequential, three-stage process from a continual learning perspective. Within this framework, we formally define complete backdoor unlearning and further derive the necessary conditions for achieving it based on the mechanism of catastrophic forgetting. Guided by these insights, we propose Blind Inversion-Backdoor Adversarial Unlearning (BI-BAU), which formulates the generation of adversarial examples satisfying the unlearning conditions as a blind inversion problem. We solve this by integrating the bi-level optimization process of adversarial training into an Expectation-Maximization (EM) algorithm framework to optimize the maximum a posteriori (MAP) objective. Furthermore, BI-BAU is extended to untargeted adversarial scenarios with unknown target classes, as well as to multi-modal contrastive learning tasks, enhancing its applicability to real-world deployment scenarios where pre-trained models may be compromised. Extensive experiments demonstrate that our method exhibits general applicability across a wide spectrum of backdoor attacks and can effectively and thoroughly eliminate the backdoor effects from a backdoor model.

阅读与讨论 → 访问原文 →

14.

bioRxiv (Bioinfo) 2026-06-15 DOI: HASH:a2645eba07cb37eaf4a2b728c3d94ee4

SMLMFlow: Improving Structural Resolution in Single Molecule Localization Microscopy with Flow Matching

作者:

Bauer↗Panconi↗Cunha↗Latron↗Sage↗Peters↗Griffie↗

While Single Molecule Localization Microscopy (SMLM) aims to generate precise coordinates of molecular targets in cells, the resulting point clouds are inherently blurred by additive noise sources across the experimental, imaging, and processing workflow. This blurring often limits SMLM's ability to accurately quantify complex assembled structures required to address biological issues, despite reported localization precision down to a couple of nanometers. Here, we present SMLMFlow, a machine learning framework for improving structural resolution in SMLM datasets that combines a graph neural network and a hierarchical transformer with flow matching. We show that SMLMFlow improves structural resolution and downstream quantification across different structures, including filaments and protein nano-clusters, and generalizes to new unseen photophysics models.

阅读与讨论 → 访问原文 →

15.

arXiv (CS.CL) 2026-06-12 DOI: arXiv:2510.02524

Unraveling Syntax: Language Modeling and the Substructure of Grammars

作者:

Laura Ying Schulz↗Daniel Mitropolsky↗Tomaso Poggio↗

While language models achieve impressive results, their learning dynamics are far from understood. Many domains of interest – such as natural language syntax, coding languages, arithmetic – are captured by context-free grammars (CFGs). In this work, we extend prior work on neural language modeling of CFGs in a novel direction: how language modeling behaves with respect to CFG substructure, namely subgrammars. We define subgrammars, and prove a set of fundamental theorems connecting language modeling and subgrammars. We show that language modeling loss recurses linearly over its top-level subgrammars; applied recursively, the loss decomposes into losses for "irreducible" subgrammars. Under additional assumptions, and empirically, parametrized models learn subgrammars in parallel, unlike children who first master simple substructures. We find that subgrammar pretraining can improve final performance, but only for tiny models relative to the grammar, while alignment analyses show that pretraining consistently leads to internal representations that better reflect the grammar's substructure.

阅读与讨论 → 访问原文 →

16.

arXiv (CS.LG) 2026-06-11 DOI: arXiv:2509.20241

Energy Use of AI Inference, Efficiency Pathways, and Test-Time Scaling

作者:

Felipe Oviedo↗Fiodar Kazhamiaka↗Esha Choukse↗Allen Kim↗Amy Luers↗Melanie Nakagawa↗Ricardo Bianchini↗Juan M. Lavista Ferres↗

arXiv:2509.20241v2 Announce Type: replace Abstract: As AI inference scales to billions of queries, estimates of per-query energy use are increasingly important for capacity planning, efficiency interventions, and policy. Yet many public estimates assume non-production settings, leading to systematic overestimation. We introduce a bottom-up framework estimating inference energy from token throughput, node power, and overhead under large-scale deployment assumptions. For frontier-scale models (>200B parameters) on H100 nodes, we estimate a median energy of 0.31 Wh/query (IQR 0.16-0.60), indicating widely cited estimates are overstated by 4-20x. In test-time scaling scenarios 15x longer than typical queries, the median energy rises 13x to 3.91 Wh (IQR 2.15-7.05). Across models, serving systems, and hardware, we estimate 8-20x line-of-sight energy reductions. At datacenter scale, serving 1 billion queries/day requires 0.7 GWh; if 10% are long queries, demand rises to 1.7 GWh/day. With efficiency interventions, it falls to 0.8 GWh/day, mitigating the energy impact of test-time scaling.

阅读与讨论 → 访问原文 →

17.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2606.18847

WorldLines: Benchmarking and Modeling Long-Horizon Stateful Embodied Agents

作者:

Yehang Zhang↗Jianchong Su↗Haojian Huang↗Yifan Chang↗Tianhao Zhou↗Xinli Xu↗Yingjie Xu↗Yinchuan Li↗Zexi Li↗Ying-Cong Chen↗

arXiv:2606.18847v1 Announce Type: new Abstract: To assist humans over extended periods in real homes, embodied agents must remember user routines, world states, and past interactions. Existing long-term memory benchmarks mainly evaluate language-centric retrieval and question answering, while embodied benchmarks often focus on short-horizon task execution without testing long-term memory use in dynamic environments. We introduce WorldLines, a project-driven benchmark for long-horizon embodied household assistance. It constructs temporally extended household traces with dialogues, actions, execution feedback, object and device state changes, and converts them into evidence-linked samples for Memory QA and Embodied Task Planning. We further propose ObsMem, an observer-grounded memory framework that maintains visibility-aware memories and action-native state trails for state-aware decisions. Experiments reveal persistent challenges in partial observability, overwritten world states, and translating long-term memory into embodied plans, while ObsMem offers a stronger reference architecture for this setting.

阅读与讨论 → 访问原文 →

18.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2602.09591

On the Optimal Reasoning Length for RL-Trained Language Models

作者:

Daisuke Nohara↗Taishi Nakamura↗Rio Yokota↗

Reinforcement learning substantially improves reasoning in large language models, but it also tends to lengthen chain-of-thought outputs and increase computational cost. Although length-control methods have been proposed, the length-accuracy relationship they induce remains unclear. We train policies with several length-control methods on multiple base models in a controlled setup and find that, across both mathematical reasoning and code generation, accuracy is non-monotonic in output length, peaking at an intermediate value. Mode accuracy, however, continues to improve with length even in settings where sample accuracy plateaus or declines, indicating that the non-monotonic length-accuracy relationship is driven by dispersion around an increasingly correct center.

阅读与讨论 → 访问原文 →

19.

arXiv (CS.CV) 2026-06-19 DOI: arXiv:2606.20303

GEN-Guard: Correcting Generalization Failures for Deployable Federated Surgical AI

作者:

Julia Alekseenko↗Pietro Mascagni↗AISafeChole Consortium↗Nicolas Padoy↗

Federated Learning (FL) in surgical video AI enables collaborative model training without sharing sensitive data. However, standard evaluation practices - selecting the "best" global model based only on validation data from participating hospitals - can lead to suboptimal deployment choices. We identify this critical failure mode as performance leakage, where the selected model overfits internal federation data and fails to generalize to unseen institutions. We propose GEN-Guard, a practical post-hoc framework to detect and correct generalization failures in federated surgical AI. It integrates Generalization Detection via Client-Blocked Evaluation (CBE), which validates performance on isolated client distributions to prevent performance leakage, and Generalization Correction through Disagreement-Aware Distillation (DAD), which learns adaptive feature-level corrections for cross-institutional robustness. Both components operate after standard FL convergence while providing robust support for zero-shot adaptation to unseen environments. We first quantify the severity of performance leakage, observing Model Selection Failures (MSFs) exceeding 80% under standard evaluation. GEN-Guard is evaluated on two multi-center clinical challenges: surgical phase recognition in laparoscopic cholecystectomy and polyp segmentation in colonoscopy. Across both datasets, GEN-Guard consistently corrects these failures, improving in-federation F1 scores by up to 2 points, unseen-institution performance by up to 3 points, and worst-case institutional performance by 3-9 points. Performance leakage represents a systematic and previously under-recognized risk in federated surgical AI. GEN-Guard provides a practical solution for detecting and correcting such failures. By improving cross-institutional robustness and zero-shot generalization, it strengthens the reliability of FL for real-world surgical deployment.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.AI) 2026-06-25 DOI: arXiv:2606.25366

Reliability-Asymmetric Spacecraft Autonomy: Co-Designing a Capable Learned GNC Stack with a Verified, Adaptation-Aware Runtime Shield

作者:

Alireza Shojaei↗

arXiv:2606.25366v1 Announce Type: cross Abstract: Deep-space missions need onboard autonomy that is both capable and certifiable. Rule-based autonomy is certifiable but brittle, while learned autonomy is capable but hard to verify. We present AMPLE-GNC, a three-tier guidance, navigation, and control stack. Its capability path combines a small foundation-model commander that maps natural language to PDDL+, a constraint-screening verifier, and a fault-adaptive controller. All three are bounded by a runtime shield with nine linear-temporal-logic invariants whose predictor soundness is machine-checked by the Kind 2 model checker. On a 6-DOF Basilisk testbed, we make three contributions. First, we deploy an edge commander. Fine-tuning a pretrained 360M model with grammar-constrained decoding gives a hard output-validity guarantee and 84% planner-executable actions. On a de-leaked test, novel-phrasing generalization is 38% exact and 51% action, rising to 48% exact after phrasing-diversity re-finetuning; we separate syntactic validity from semantic accuracy. Second, we introduce a fault-adaptive controller. Rapid Motor Adaptation infers latent actuator faults online and recovers 97.8% of actuator-sign faults and 94.4% of continuous-gain faults within the training randomization envelope. Fault-unaware PD and from-scratch end-to-end RL both score 0%, while the strongest classical-adaptive baseline reaches 55% on continuous gain. Beyond the envelope, a split-conformant retrain scores 57-67%, and adding 4x more in-regime data worsens performance, showing that randomization breadth, not data volume, drives generalization. Robustness is flat under star-tracker noise to 0.005. Third, we show that a latching safe-hold shield can suppress even a capable controller. A split-conformal recovery-deadline certificate with adaptation-aware engagement reconciles safety and recovery, keeping the controller 94.5% autonomous while still catching non-recovery.

阅读与讨论 → 访问原文 →

21.

medRxiv (Medicine) 2026-06-25 DOI: HASH:88ea8687521e41cecbf2639940d0e67e

Sources and transmission dynamics of Campylobacter and Shigella using culture-independent molecular methods in Tanzania (SCAT): a protocol for a prospective observational cohort study

作者:

Mbelele↗P. M↗Katengu↗Wettstone↗E. G↗Pholwat↗Elwood↗Ahmed↗Guga↗G. W↗Temu↗Habiye↗…

Introduction: Enteric infections caused by Shigella and Campylobacter species are associated with acute diarrhea as well as enteric dysfunction leading to chronic malnutrition and poor child development. However, interventional studies to reduce exposure to these pathogens have been ineffective, and evidence that identifies putative sources and species transmission dynamics between humans, animals and the environment is scarce. Because isolation of these organisms from both human and environmental samples is challenging, studies using molecular detection may help reduce this substantial knowledge gap and lead to targeted interventions to reduce the burden of disease. Methods and analysis: This longitudinal cohort study will enroll 100 index infants as well as their families in Haydom, Tanzania. Families will be followed for one year, with collection of stool samples monthly and during incident diarrhea from all household members. Additionally, we will sample from the index child's environment, including domestic animal stools, drinking water, milk, porridge, flies, and soil. All samples will undergo nucleic acid extraction and quantitative PCR for Campylobacter and Shigella. Serial serologic tests will also be performed to identify seroconversion to these pathogens. Associations between pathogen detection from the environment and household members and detection in index children will be estimated, and the temporal patterning of infection cases will be analyzed using transmission models. Ethics and dissemination: This trial has been approved by the Tanzanian National Institute for Medical Research and the University of Virginia Institutional Review Board.

阅读与讨论 → 访问原文 →

22.

medRxiv (Medicine) 2026-06-17 DOI: HASH:31f58ac2e5c13c7a9a5c82dfd8f36530

Deep learning for interactive and automated inner retinal layer segmentation in OCT images of patients with retinitis pigmentosa using limited training data

作者:

Laurence↗D. S↗Schilling↗Grimm↗N.-A↗Mace↗Bemme↗Pape↗

Purpose: New therapeutic strategies such as optogenetics have created a need for accurate tracking of inner retina degeneration in Retinitis pigmentosa (RP) patients. We introduce two tailored deep learning models to segment the RNFL (retinal nerve fibre layer), GCIPL (ganglion cell inner plexiform layer), INL (inner nuclear layer), CFT (central foveal thickness) and RPE (retinal pigment epithelium) in RP: The first is based on a Segment Anything Model (SAM), the second on nnU-Net. To our knowledge, SAM has not yet been applied to retinal layers in OCT data. Methods: SD-OCT images of a retrospective cohort of 37 RP patients were included. Data for four training cycles were prepared semi-automatically in MATLAB, then assessed and corrected by three expert graders. 1,700 segmented B-Scans from two open datasets were used for pretraining. For post-processing, semantic retinal boundary detection was developed. The final models, OCT-SAM and nnU-Net, were trained on 228 annotated RP scans. Detected layer thicknesses were validated against manual segmentation at 90 random points in 30 OCT B-Scans. Finally, OCT-SAM was tested on three RP cases with retrospective, longitudinal OCT data. Results: nnU-Net achieved a precision, recall and F-1 score of 0.96 while OCT-SAM performance resulted in slightly lower values of 0.93, 0.8 and 0.85, respectively. OCT-SAM measurements had low bias and good agreement with manual annotations, confirming reliability. Conclusions: OCT-SAM enabled fast data annotation and tool integration, whereas nnU-Net provided the best segmentation performance. OCT-SAM demonstrated longitudinal reproducibility and detected RP-characteristic pathologies and degenerative changes. Future work will extend OCT-SAM to 3D OCT segmentation.

阅读与讨论 → 访问原文 →

23.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.15370

MNet++: Extended 2D/3D Networks for Anisotropic Medical Image Segmentation

作者:

Kirsten Odendaal↗Rade Bajic↗

This work demonstrates a full reproduction and extension of MNet, a hybrid 2D/3D convolutional network designed for anisotropic medical image segmentation. The original architecture was re-implemented within the nnU-Net framework to verify its reported performance and robustness to variable voxel spacing, known as anisotropy. Experiments were conducted on PROMISE prostate MRI and a controlled subset of LiTS liver CT under matched preprocessing and compute constraints. The reproduced MNet achieved a Dice similarity coefficient (DSC) of 89.0 +/- 0.9% on PROMISE, within 0.8% of the published result, and 94.3 +/- 1.9% / 54.6 +/- 3.1% for liver and tumor segmentation on LiTS, respectively. Two lightweight extensions were further introduced: (1) a learned Fusion Gating mechanism enabling adaptive 2D-3D feature blending, and (2) a VMamba state-space module for efficient long-range depth modelling. The Spatial Gating variant improved DSC by +0.8% with less than 3% inference overhead, while VMamba improved performance consistency, reducing PROMISE Dice variation to +/- 0.7% and achieving the strongest LiTS liver performance at 95.8% Dice. Both extensions preserved MNet robustness to anisotropy, with delta Dice = 1.5% across 1-4 mm voxel spacing. Overall, the study confirms MNet reproducibility and demonstrates that adaptive fusion and state-space modelling have the potential to further strengthen segmentation reliability under anisotropic conditions. However, further tests are required to provide definitive conclusions.

阅读与讨论 → 访问原文 →

24.

arXiv (quant-ph) 2026-06-25 DOI: arXiv:2606.25933

From spectral structure to sensing limits in quantum thermometry

作者:

Youssef Aiache↗Simone Cavazzoni↗Abderrahim El Allati↗Paolo Bordone↗Matteo G. A. Paris↗

arXiv:2606.25933v1 Announce Type: new Abstract: The precision of a quantum thermometer is fundamentally constrained by the spectral structure of the probe itself, and a systematic mapping between the configurations of energy levels and thermometric performance provides relevant information to design optimized devices. In this work, we establish such a mapping by analyzing a broad class of quantum systems, ranging from finite spin ensembles and degenerate atoms to confining potentials, quantum walks, and continuous-spectrum models. We derive exact scaling laws for the quantum Fisher information, revealing two distinct high-temperature universality classes: finite-spectrum probes exhibit a $T^{-4}$ decay, while unbounded or continuous spectra yield a slower $T^{-2}$ decay. At low temperatures, we show that sensitivity, though universally exponentially suppressed, can be enhanced arbitrarily by engineering degenerate excited states or a quantum walk on a fully connected topology. By contrast, specific quantum walk topologies provide a distinct enhancement mechanism based on gap engineering, whereby an optimal network size yields an optimized $T^{-2}$ low-temperature scaling. Furthermore, power-law spectra enable tunable scaling of thermometric performance with system size, offering a design principle for optimal probes in specific temperature windows. Our results contribute to transform spectral information into a resource for quantum thermometry, providing both fundamental bounds and practical guidelines to tailored temperature sensing.

阅读与讨论 → 访问原文 →

25.

arXiv (CS.LG) 2026-06-25 DOI: arXiv:2606.25335

Stagnant Neuron: Towards Understanding the Plasticity Loss in Multi-Agent Reinforcement Learning Value Factorization Methods

作者:

Zhengzhu Liu↗Zeming Gao↗Haoyuan Qin↗Jiawei Hu↗Junhao Wu↗Miao Zhu↗Haipeng Zhang↗Chennan Ma↗Siqi Shen↗Cheng Wang↗

arXiv:2606.25335v1 Announce Type: new Abstract: Multi-Agent Reinforcement Learning (MARL) value factorization methods can suffer from a loss of plasticity, gradually failing to adapt when transferring to new task instances. We trace this issue to stagnant neurons, units whose gradient updates become negligibly small relative to their weights, thereby hindering learning. While existing plasticity injection methods exist, they prove ineffective for such neurons. To address this, we propose Knowledge-retentive Neuron-level PlastIcity Focusing InjEction (KNIFE), a novel method that directly targets stagnant neurons. KNIFE replaces each stagnant neuron with a composite unit comprising three specialized components: a frozen knowledge neuron to preserve acquired knowledge, a re-initialized active neuron to restore learning capacity, and a compensation neuron to ensure the combined output matches the original, thus maintaining previous learned cooperation knowledge. Extensive experiments on SMACv2, predator-prey, and matrix games demonstrate that KNIFE significantly outperforms state-of-the-art plasticity injection methods.

阅读与讨论 → 访问原文 →