❌

Reading view

Inhalable carrier-free self-assembled leonurine-ursolic acid nanoaggregates ameliorate acute lung injury by suppressing TLR4/MyD88-NET axis

Mater Today Bio. 2026 Aug 18;40:103583. doi: 10.1016/j.mtbio.2026.103583. eCollection 2026 Oct.

ABSTRACT

TLR4 activation and the cascade of neutrophil extracellular trap (NET) formation exacerbate excessive inflammation and organ damage in the pathogenesis of acute lung injury (ALI), yet effective pharmacological interventions remain unavailable. Nanoaggregates derived from natural products offer promising avenue by leveraging synergistic anti-inflammatory effects. In this study, we surprisingly discovered that leonurine and ursolic acid spontaneously self-assemble into nanoparticles (LUNP) through non-covalent interactions, achieving a drug loading capacity of 100%. The LUNP platform exhibits superior biophysical properties, including enhanced mucus penetration, pH-responsive drug release, improved cellular uptake, and prolonged retention within inflamed lung tissue. Mechanistically, LUNP ameliorates ALI by dampening TLR4/MyD88/NF-κB-driven inflammatory activation, thereby remodeling the microenvironment to limit NOX4-PAD4-mediated NET formation. Notably, inhalational LUNP exhibits outstanding biosafety with minimal off-target distribution. Overall, this work introduces a synergistic self-assembled nanoplatform for precise pulmonary intervention in ALI, showcasing its ability to safely and effectively orchestrate the coordinated modulation of multiple pathological pathways. In summary, by inhibiting both TLR4 activation and NET formation, the synergistic LUNP platform offers an efficient, safe, and easily accessible therapeutic strategy for ALI, providing a promising solution for clinical translation.

PMID:42750707 | PMC:PMC13577835 | DOI:10.1016/j.mtbio.2026.103583

  •  

Integrated multi-omic profiling enables recurrence risk stratification beyond pathological stage in resected EGFR-mutant lung adenocarcinoma

J Thorac Oncol. 2026 Sep 16:104204. doi: 10.1016/j.jtho.2026.104204. Online ahead of print.

ABSTRACT

BACKGROUND: Early-stage EGFR-mutant lung adenocarcinoma (LUAD) demonstrates heterogeneous outcomes after curative surgery, yet adjuvant treatment decisions are guided by pathological stage alone. Following the ADAURA trial, adjuvant osimertinib is the standard of care for resected stage IB-IIIA EGFR-mutant LUAD; however, real-world data demonstrate that up to 40% of patients remain disease-free at five years without adjuvant osimertinib, underscoring the need for improved risk stratification.

PATIENTS AND METHODS: We performed integrated clinical, genomic and transcriptomic profiling of 400 patients with resected stage IA-IIIA EGFR-mutant LUAD. EGFR-mutant recurrence risk models integrating clinical, genomic and transcriptomic data were developed and validated across one internal and three external cohorts.

RESULTS: Genomic instability, including TP53 co-mutations, copy number alterations and APOBEC-associated mutational signatures, increased with pathological stage. RBM10 co-mutations were enriched in tumours with L858R mutations and correlated with upregulation of WNT signalling and epithelial-mesenchymal transition. Transcriptomic features outperformed clinical or genomic variables alone in predicting recurrence risk, and a multi-omic model demonstrated superior and reproducible performance, achieving a median concordance index of 75.4% across four independent validation cohorts. The multi-omic model stratified recurrence risk within individual pathological stages, including stage I disease, and identified patients most likely to benefit from adjuvant EGFR TKI.

CONCLUSIONS: These findings define the molecular heterogeneity of early-stage EGFR-mutant LUAD and support multi-omic risk stratification to inform adjuvant EGFR TKI decisions beyond pathological stage. Prospective validation in larger cohorts will be required to confirm these findings.

PMID:42749051 | DOI:10.1016/j.jtho.2026.104204

  •  

Type 1 interferon perturbates clonal competition by reshaping human blood development

Nat Genet. 2026 Sep 15. doi: 10.1038/s41588-026-02751-3. Online ahead of print.

ABSTRACT

Inflammation accelerates evolutionary dynamics of hematopoietic stem cells (HSCs) in clonal hematopoiesis and myeloid neoplasms. We studied HSCs, progenitors and immune cells from patients with myeloproliferative neoplasms at baseline and following interferon-α (IFNα) treatment, the only therapy to deplete mutated stem cells. We deployed single-cell multiomics methods that distinguish the IFNα effects on mutated stem cells from the admixed wild-type HSCs, with respect to their differentiation, transcriptomes, immunophenotypes and chromatin accessibility. IFNα simultaneously activated HSCs into two polarized states: a lymphoid progenitor expansion associated with an anti-inflammatory state and an inflammatory myeloid progenitor state derived from HSCs. The augmented lymphoid differentiation balanced the typical myeloproliferative-neoplasm-induced myeloid bias, associated with normalized blood counts. Somatic mutations modified the effects of IFNα on HSC differentiation and cell cycle entry rates. Clonal fitness upon IFNα exposure was due to resistance of CALR- or JAK2-mutated stem cells to differentiate into inflammatory myeloid progenitors.

PMID:42745000 | DOI:10.1038/s41588-026-02751-3

  •  

EBV reactivation priming of the peripheral immune system in multiple sclerosis relapse

Nature Medicine, Published online: 16 September 2026; doi:10.1038/s41591-026-04665-3

Increased expression of EBV reactivation genes in B cells and MS risk genes targeted by the EBV protein EBNA-2 precedes MS attacks, linking EBV reactivation and genetic risk to the development of MS relapses.
  •  

Factor IX Padua AAV gene therapy in adolescents with hemophilia B: a phase 1 trial

Nature Medicine, Published online: 16 September 2026; doi:10.1038/s41591-026-04636-8

In this single-arm phase 1 trial, an AAV gene therapy carrying the Padua variant of factor IX was well tolerated in 11 adolescents with hemophilia B and led to reductions in annualized bleeding rate.
  •  

On-premise medical AI agents for reliable clinical decision-making

Nature Medicine, Published online: 15 September 2026; doi:10.1038/s41591-026-04609-x

An autonomous clinical AI agent enhances decision-making through on-premise deployment and reliability metrics, achieving high diagnostic accuracy and selective autonomy.
  •  

Broadly neutralizing antibodies in adult males living with HIV undergoing analytical treatment interruption: secondary and exploratory outcomes of the phase II randomized controlled RIO trial

Nature Medicine, Published online: 15 September 2026; doi:10.1038/s41591-026-04644-8

In the phase 2 RIO trial, there was delayed viral rebound and resistance to broadly neutralizing antibodies 3BNC117-LS and 10-1074-LS in adult males living with HIV undergoing analytical treatment interruption, and initial reservoir sensitivity to autologous antibodies was associated with a longer time to rebound.
  •  

Spatial atlas of the human brain vasculature reveals specialized cell ensembles

An integrative cell and spatial atlas defines the organizing logic of the adult human cerebrovasculature, providing a framework for studying how vascular cell states support brain function and shape neurological disease vulnerability.
  •  

Advancing cancer detection and treatment using longitudinal routine clinical data

Liu et al. develop Oncoformer, a multimodal transformer that reads routine laboratory tests and chest X-rays already collected in everyday care. Across more than 3.6 million individuals, it detects cancer, infers tumor stage, and stratifies treatment response and recurrence risk, pointing toward risk-adapted cancer care built on data already in hand.
  •  

Digitally Adapting LGBTQ-Affirmative Cognitive Behavioral Therapy for Chinese Men Who Have Sex With Men Living With HIV: User-Centered Design Approach

Background: Chinese men who have sex with men living with HIV (MSMLWH) experience substantial psychological distress driven by minority stress and HIV-related challenges. However, culturally tailored digital mental health interventions that address HIV-specific maladaptive cognitive schemas and culturally specific psychosocial stressors remain scarce in China. Objective: This study aimed to systematically adapt an evidence-based cognitive behavioral therapy (CBT) intervention Effective Skills to Empower Effective Men (ESTEEM) into a WeChat (Tencent) Mini-Program–based intervention (iESTEEM) specifically for Chinese MSMLWH and to evaluate its preliminary feasibility and usability. Methods: We used a three-phase user-centered design approach guided by the Assessment, Decision, Adaptation, Production, Topical Experts, Integration, Training, and Testing (ADAPT-ITT) framework. The study proceeded in three phases: (1) a qualitative needs assessment using semistructured interviews with 20 MSMLWH (mean age 23.25, SD 3.08 years); (2) systematic intervention adaptation and platform development, including theater testing (n=5); and (3) a 2-week pilot study involving 10 MSMLWH and five counselors to evaluate feasibility, usability, and acceptability through focus groups and objective platform analytics. Results: Phase 1 identified 3 major themes of psychological distress: persistent health anxiety fueled by catastrophizing, intersectional stigma internalization, the disclosure dilemma, and intimacy barriers rooted in defectiveness and shame schemas. Participants also prioritized anonymity and bite-sized learning. Guided by these findings, iESTEEM was developed as a counselor-assisted, privacy-preserving WeChat Mini-Program incorporating HIV-specific scenarios, multimodal learning modules, and a back-end risk-alert system. During the 2-week pilot, participants logged into the platform 14.1 (SD 6.7) times per person and completed 134.3 (SD 103.1) minutes of learning activities; all participants accessed module 1, and 90% (9/10) accessed modules 2‐5. Anxiety scores decreased from 8.9 (SD 2.3) to 7.2 (SD 3.0), whereas depression scores remained stable. All participants expressed a willingness to continue using the program and to recommend it to peers. Participants and counselors endorsed its contextual relevance, privacy protections, and clinical utility. Conclusions: This study provides a theory- and evidence-informed model for culturally adapting digital mental health interventions for highly stigmatized populations. By integrating lesbian, gay, bisexual, transgender, and queer (LGBTQ)-affirmative CBT principles, HIV-specific adaptations, and a privacy-preserving, counselor-assisted WeChat Mini-Program, iESTEEM demonstrated promising preliminary feasibility, acceptability, and engagement among Chinese MSMLWH. These findings support the potential of culturally tailored digital interventions to expand access to psychological support for this stigmatized population in resource-constrained settings. Ongoing randomized controlled trials will further evaluate its efficacy, implementation outcomes, and mechanism of action. Trial Registration: Chinese Clinical Trial Registry ChiCTR2400080263; https://www.chictr.org.cn/showproj.html?proj=216926
  •  

Occamy-1.0: Open Pareto-frontier 35B Intelligence for Co-work

arXiv:2609.11977v1 Announce Type: new Abstract: Co-work agents execute complex workflows that combine information gathering, tool use, coding, and file manipulation across many model invocations. Because cost and latency accumulate over the full episode, their practical value depends not only on peak capability but also on how efficiently that capability is delivered. Yet many steps in everyday work emphasize state tracking, coordination, recovery, and follow-through rather than frontier-scale reasoning. We present Occamy-1.0, a cost-efficient co-work model obtained by further training the post-trained Qwen3.6-35B-A3B checkpoint. We construct execution-grounded data and environments, capture replayable long-horizon trajectories across multiple harnesses, and use staged post-training to develop and consolidate complementary execution capabilities. Across a broad suite of co-work benchmarks, Occamy-1.0 is consistently among the strongest comparably sized models and remains competitive with substantially larger frontier systems on several tasks. Under our stated evaluation and pricing protocol, its aggregate performance across four representative benchmarks places it at the low-cost knee of the observed cost--performance Pareto frontier. Supporting evaluations in tool calling, coding, and instruction following further show that this specialization preserves broad agentic capability. We release the model weights and a subset of the training data to support research on practical co-work agents and agentic post-training.
  •  

OneLA: Scaling Linear-Attention Decoding to Large Beams in Generative Recommendation

arXiv:2609.12399v1 Announce Type: new Abstract: Generative recommendation (GR) relies on large-beam decoding to generate hundreds of candidate items, creating a new scaling challenge for recurrent linear attention. Existing linear attention serving systems either materialize a full recurrent state for every beam or repeatedly replay shared history, incurring substantial memory and traffic overhead. To address this, we present OneLA, a linear-attention decoding framework that exploits the shared prompt and short divergent suffixes of GR workloads. Specifically, OneLA represents all beam states using a single shared prompt-derived state and compact, append-only records of their divergent transitions. Using this representation, OneLA computes only the state information required at each decoding step, without reconstructing a full recurrent state for every beam. Furthermore, OneLA uses a lightweight ancestry index to track the transition records that make up each beam's history, allowing beams to be updated without moving or copying existing records. A fused GPU kernel further reuses the shared state across beams. Our analysis shows that OneLA achieves 1.54-2.46x end-to-end decode speedups while substantially reducing recurrent-state memory use and data movement.
  •  

Beyond ID Embeddings: Process-Grounded Language Modeling for Cognitive Diagnosis

arXiv:2609.12403v1 Announce Type: new Abstract: Cognitive Diagnosis Models (CDMs) play a pivotal role in personalized online learning. Traditional CDMs rely on discrete, ID-based embeddings to represent students, exercises, and concepts. This paradigm diverges from the nature of learner cognition, where knowledge is not stored and retrieved as isolated symbols. As a result, CDMs suffer from semantic limitations when new exercises or concepts appear. In this paper, we propose a Process-aware Language Cognitive Diagnosis (PLCD) framework that uses language-derived structures as cognitive priors and response records to calibrate student posterior states. PLCD leverages large language models (LLMs) to construct concept schemas and cognitive process graphs, and uses target-conditioned semantic memory to retrieve historical responses that are relevant to each target exercise. A process-grounded Language-to-Cognition Mapper with DA-MoE experts and process-level contrastive learning then maps the textual evidence into a unified cognitive space. Experimental results show that PLCD not only outperforms traditional baselines in predicting student performance but also exhibits strong cognitive transfer capabilities. These results connect the computational power of LLMs with the psychometric goal of measuring latent knowledge states, suggesting that structured language priors calibrated by response records can improve cold-start robustness and cognitive grounding.
  •  

VRL-Bench: Benchmarking agents on computer control tasks under finite trial budgets

arXiv:2609.12404v1 Announce Type: new Abstract: Learning from trial and error is a promising way to improve language agents on complex tasks such as computer control. Reflexion introduced verbal reinforcement learning, which turns failed trials into text that guides later attempts without updating model parameters. We introduce VRL-Bench, a harness for fair evaluation of trial-and-error learning under finite trial budgets. Across three models on MiniWoB and WebShop, we evaluate updates from several prominent verbal-memory methods spanning Reflexion and later work: each improves observed success over memory-free retry in some settings but reduces it in others. Replay experiments show that using reflection can reduce success rates, revealing a trade-off between exploiting experience and continued exploration. We propose VEX$^2$, a verbal exploration--exploitation scheduler that uses a language model to jointly select policies and allocate the remaining trial budget. VEX$^2$ is the only evaluated update to achieve positive observed success-rate gains over retry in all six settings.
  •  

Hierarchical Belief Modeling for Zero-Shot Opponent Adaptation in Partially Observable Multi-Agent Navigation

arXiv:2609.12422v1 Announce Type: new Abstract: Lux AI Season 3 requires agents to act under partial observability, randomized episode level dynamics, and a best of five match structure that rewards both tactical execution and fast adaptation. We present HORIZON, a hierarchical agent that combines symmetry aware spatial perception, dual memory belief tracking, relic centric graph attention, information gain driven exploration, and an opponent conditioned policy mixture. HORIZON separates short horizon control from cross match meta reasoning, while auxiliary belief and world model objectives stabilize learning. Trained with PPO in a large scale JAX simulator, the resulting agent explicitly infers hidden game parameters and opponent style. Experiments show consistent gains in match win rate, episode win rate, adaptation gain, and league rating over strong recurrent and feed forward baselines.
  •  

EvoRS: On-Policy Self-Evolution of Reward Systems for Open-Ended Reinforcement Learning

arXiv:2609.12459v1 Announce Type: new Abstract: Open-ended reinforcement learning often relies on rubric-based rewards for tasks without directly verifiable answers. Yet the policy and reward system form a dynamic feedback loop: as the policy optimizes the current reward, an initially useful reward system may become unreliable due to reward hacking or reduced response discriminability. The reward system should therefore evolve rather than remain fixed during training. Existing dynamic-rubric methods adapt evaluation criteria, but reward failures can also arise from scoring mechanisms or signal composition. We introduce EvoRS, a self-evolving RL framework that evolves the reward system from on-policy experience, representing it as an executable Reward-DAG. Specifically, an agentic designer updates this system from on-policy rollouts and reward traces to maintain train-time reliability. Across writing and roleplay, EvoRS achieves the best quality under all three judges, outperforming the policy by \(2.107\) and \(4.767\) points, respectively, while reducing reward hacking and coverage failures and preserving reward informativeness. Ablations confirm that a comprehensive fixed reward system cannot remain reliable in open-ended tasks and must evolve throughout training.
  •  

Confidence-Gated Transductive Test Generation for Code Reranking

arXiv:2609.12489v1 Announce Type: new Abstract: Test case synthesis is crucial for evaluating and ranking programs generated by large language models (LLMs). However, constructing high-quality test cases remains challenging because reliable expected outputs are often difficult to obtain. We propose Confidence-Gated Transductive Test Generation (CoTT), which first uses an efficient inductive procedure and invokes transductive generation only when inductive confidence is low. This adaptive design improves output reliability while allocating extra computation only when needed. On code reranking benchmarks, CoTT outperforms prior baselines across the reported metrics while reducing cost relative to applying transductive generation to every input. These results show that confidence-based allocation of test-time computation provides a favorable efficiency-effectiveness trade-off with a single efficient LLM.
  •  

Reproducing and Evaluating the Generalizability of Subliminal Learning in Open-Weight Models

arXiv:2609.12586v1 Announce Type: new Abstract: In this reproduction paper we investigate subliminal learning, a consequence of distillation where teacher models transmit behavioral preference traits through semantically unrelated data. The original paper explores two types of traits (animal preferences and misalignment), three data modalities (number sequences, code, and chain of thought), and several model families. We reproduce their experiments and extend the setup along three axes: new preference categories (actors and politicians), a new task (chess move generation), and an additional open-weight model (Ministral8B). We also run a controlled ablation on the numbers task's answer-space size (1-, 2-, and 3-digit sequences). We focus on open-weight models with accessible checkpoints on HuggingFace, since the original paper's GPT-4.x fine-tuning is no longer available. Our reproduction supports the original paper's claims, but our extensions show they are not universal as transmission strength varies across traits and tasks, and one model shows almost no effect at all.
  •  
❌