❌

Reading view

Minimal residual disease and relapse surveillance in osteosarcoma: an action-linked framework integrating liquid biopsy and imaging biomarkers

J Bone Oncol. 2026 Sep 16;61:100803. doi: 10.1016/j.jbo.2026.100803. eCollection 2026 Dec.

ABSTRACT

Osteosarcoma relapse surveillance remains dominated by scheduled imaging because salvage treatment depends on anatomical confirmation of pulmonary, local or extrapulmonary recurrence. However, radiological recurrence may occur after a biologically active phase in which residual viable disease or micrometastatic progression is already present but not yet localizable. This clinical-translational review reframes postoperative osteosarcoma surveillance as an action-linked decision workflow rather than a comparison of isolated biomarker technologies. Current evidence suggests that tumor-informed circulating tumor DNA (ctDNA) sequencing provides the strongest osteosarcoma-specific minimal residual disease (MRD) signal, with postoperative positivity associated with inferior event-free survival and, in selected patients, molecular detection preceding imaging-confirmed relapse or progression. Cell-free DNA methylation may offer a mutation-independent adjunct, whereas circulating tumor cells, extracellular vesicles and circulating microRNAs remain exploratory signals without validated postoperative surveillance actions. Chest computed tomography (CT) and local magnetic resonance imaging (MRI) remain indispensable for disease localization and treatment planning, while diffusion-weighted imaging, dynamic contrast-enhanced MRI and radiomics currently provide mainly local viability or risk-enrichment information rather than proven surveillance-intervention evidence. The near-term role of integrated biomarkers is therefore not to replace guideline-based imaging, but to define protocolized pathways for molecular-positive/imaging-negative, imaging-positive/molecular-negative, concordant high-risk and concordant low-risk states. Future studies should test whether biomarker-triggered reassessment improves clinically meaningful outcomes, including resectability, second complete remission, clinical trial access, patient burden and survival, rather than simply documenting recurrence earlier.

PMID:42824543 | PMC:PMC13628598 | DOI:10.1016/j.jbo.2026.100803

  •  

Transketolase-like 1 potentiates PD-1 blockade in hepatocellular carcinoma by glycolysis to prime dendritic cell lactylation

Signal Transduct Target Ther. 2026 Sep 28;11(1):418. doi: 10.1038/s41392-026-02875-2.

ABSTRACT

Hepatocellular carcinoma (HCC) exhibits a suboptimal response to immune checkpoint blockade (ICB) therapy; to overcome this resistance, we aimed to delineate key immune resistance factors via multi-omics analysis, develop strategies to block their immunosuppressive axes, and engineer a targeted nanosystem to enhance immunotherapy efficacy against PD-1 resistance in HCC. Using transcriptomic and proteomic data from anti-PD-1-treated HCC patients, along with functional validation in murine models and mechanistic molecular and cell biology studies, we identified transketolase-like 1 (TKTL1) as a dual-nature biomarker where overexpression predicted poor baseline prognosis yet enhanced response to ICB. Mechanistically, TKTL1 diverts glucose flux into glycolysis rather than pentose phosphate pathway (PPP), recruiting USP9X to deubiquitinate and stabilize HIF-1α, which upregulates HK2 to amplify glycolytic output and lactate accumulation. This metabolic rewiring orchestrates dual immunosuppressive circuits through HIF-1α-driven CCL4 secretion recruiting PD-L1high dendritic cells (DCs), coupled with lactate-induced TRIM28K408 lactylation that stabilizes PD-L1 by blocking ubiquitin-mediated degradation. We engineered a hepatoma-membrane-coated MnO₂ nanosystem (CQLH) co-delivering a TKTL1 inhibitor and lactate oxidase, which disrupted the TKTL1-HIF-1α-HK2 axis, depleted lactate, and reprogrammed the tumor microenvironment, thereby enhanced anti-PD-1 therapy to suppress tumor growth, especially in TKTL1high tumors. These findings define a critical "TKTL1-glycolysis-lactate-DC" axis driving anti-PD-1 sensitivity in HCC, position TKTL1 as both a potential biomarker for ICB response and a tractable therapeutic target, and demonstrate that the targeted CQLH nanosystem overcomes resistance and enhances anti-PD-1 efficacy, offering a precision immunotherapeutic strategy for TKTL1high HCC.

PMID:42802226 | PMC:PMC13616917 | DOI:10.1038/s41392-026-02875-2

  •  

ESAM reduces sensitivity to anti-HER2 therapy in HER2-positive breast cancer by activating the mTOR pathway

Cell Death Discovery, Published online: 16 September 2026; doi:10.1038/s41420-026-03349-8

ESAM reduces sensitivity to anti-HER2 therapy in HER2-positive breast cancer by activating the mTOR pathway
  •  

Advances in single-cell and spatial multi-omics for deciphering the mechanisms of pan-organ metastasis in breast cancer

Biochim Biophys Acta Rev Cancer. 2026 Sep 15:189717. doi: 10.1016/j.bbcan.2026.189717. Online ahead of print.

ABSTRACT

Breast cancer deaths are mainly caused by metastasis to distant organs, not by the primary tumor. Bone, lung, liver, and brain are the most common metastatic sites, each showing different clinical behaviors and treatment responses-a pattern often called metastatic organotropism. Bulk omics can provide tissue-level information, but they fall short in identifying rare metastasis-initiating clones or capturing how tumor cells adapt to distinct organ microenvironments. With recent progress in single-cell sequencing, multi-omics integration, and spatial profiling, it is now possible to study metastasis at much finer cellular and spatial resolution. In this review, we synthesize current evidence from two complementary perspectives. First, we summarize pan-organ programs associated with metastatic competence, including partial epithelial-mesenchymal transition, lineage plasticity, stem-like states, stress tolerance, metabolic flexibility, immune evasion, and stromal-vascular remodeling. Second, we discuss how these programs are reshaped by organ-specific microenvironments: osteolytic and mixed bone remodeling and marrow dormancy in bone, inflammatory vascular niches in lung, tolerogenic antigen presentation and hepatic metabolism in liver, and blood-brain/blood-tumor barrier constraints, glial crosstalk, neuronal interactions, and lipid-metabolic adaptation in brain. We also highlight how CTC/CTM profiling, spatial mapping, and longitudinal integration refine the understanding of dissemination, dormancy, colonization, outgrowth, and treatment resistance. Although these approaches hold translational promise, most remain at the discovery or early validation stage and require assay simplification, prospective testing, and cross-center standardization. Overall, single-cell and spatial multi-omics are reframing breast cancer metastasis as a dynamic, multi-stage, and tissue-shaped process, providing a foundation for future biomarker development and mechanism-guided therapeutic strategies.

PMID:42744123 | DOI:10.1016/j.bbcan.2026.189717

  •  

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.
  •  

BlueLM-GUI Technical Report: A Real-Device-Centric Flywheel for Self-Improving Mobile GUI Agents

arXiv:2609.12394v1 Announce Type: new Abstract: Mobile GUI agents are shifting from multi-module frameworks to native models trained end-to-end, yet industrial deployment faces three persistent gaps. Sandbox training produces a distribution mismatch with production environments; expensive real-device failures remain underutilized; and fixed benchmarks saturate, losing the power to guide iteration. We present BlueLM-GUI, a 35B-A3B mobile GUI agent built as a real-device-centric flywheel that closes these gaps through three principles. Every Sample Matters: a dual-track pipeline with Heterogeneous Triple-System Consensus evaluation and an Error Correction \& Derivation Module salvages every trajectory into usable supervision. Every Rollout Is Real: a three-stage recipe---continual pre-training, supervised fine-tuning, and agentic reinforcement learning on hundreds of real phones---grounds every rollout in real production environments, so the capability the model learns transfers directly to deployment. Every Query Evolves: a quota-driven benchmark methodology with three orthogonal axes enables precise attribution and allows the benchmark to be systematically upgraded as the model improves. BlueLM-GUI achieves 87.4 on MobileGUI-VBench, surpassing the best closed-source model by 5.1 points, and 84.9 on AndroidWorld, the best result among open-source models and competitive with closed-source models. These results demonstrate that grounding model training and iterative improvement in both real devices and the three Every principles yields strong, robust, and transferable mobile GUI capability.
  •  

MAIA: Multi-Agent Intent Articulation for Requirement Discovery in Art Commissions

arXiv:2609.12097v1 Announce Type: cross Abstract: In bespoke art commissions, laypeople know what they feel but lack the words to specify it: one participant wanted a laid-off truck driver depicted as "a ghost in his own machine" but left the medium, scale, and palette unsaid. We frame this as an articulation bottleneck at an under-served upstream stage: requirement discovery, which precedes any artist or image generator and forces the commissioner to constitute intent in the first place. We present MAIA (Multi-Agent Intent Articulation), a multi-agent system that scaffolds this stage through Socratic inquiry under a "Verification over Invention" rule, turning vague affect into a text-only brief of visual terms the user verifies. In a within-subjects study (N = 16), the full configuration produced a large, significant gain in Cognitive Support over a minimal baseline (r = 0.96, p_FDR = 0.015; LMM p_FDR
  •  

SeqMoE: Toward Full-Load Performance via Predictive and Graph-Compatible MoE Offloading

arXiv:2609.12978v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) creates a structural advantage for offloading: only a small fraction of activated experts need to reside in device memory, and if they can be loaded in time for computation, offloading can in principle approach full-load performance, where all model weights reside in device memory. Yet translating MoE's structural advantage into practical offloading gains remains challenging. We propose SeqMoE to bridge this gap. To maximize expert hits, we build predictive memory management: (i) Sequence-to-sequence prediction. We are the first to recast expert activation prediction as sequence modeling, enabling accurate multi-step, multi-layer forecasts that provide a long and reliable window for downstream decisions. (ii) Joint prefetch scheduling. We formulate prefetch scheduling as Job Sequencing with Deadlines to maximize expected expert hits and improve bandwidth efficiency. (iii) Forecast-driven caching. Leveraging the recursive nature of sequence modeling, we introduce a probabilistic Belady policy for future-aware eviction. To eliminate execution bottleneck, we develop (iv) Graph-compatible offloading runtime. We derive general runtime principles encompassing compute-transparent expert placement and synchronization-free orchestration disciplines for end-to-end graph capture. With 45% expert residency, SeqMoE averages a 96.97% hit rate and 80.22% of full-load performance, advancing the state of the art in MoE offloading.
  •  

MMSDH facilitates ACSL4 propionylation to counteract ferroptosis upon hypoxia and impairs PDAC chemotherapy efficacy

Nature Cancer, Published online: 11 September 2026; doi:10.1038/s43018-026-01236-w

Zheng et al. describe how hypoxia-induced methylmalonate semialdehyde dehydrogenase lactylation promotes acyl-CoA synthetase long-chain family member 4 propionylation and degradation, thereby suppressing ferroptosis induced by chemotherapy, and develop a blocking peptide that increased chemotherapy efficacy in pancreatic ductal adenocarcinoma.
  •  

Switchable single-atom catalysts for highly selective C–C coupling in direct methane oxidation

Nature Nanotechnology, Published online: 07 September 2026; doi:10.1038/s41565-026-02271-5

Single copper atoms on boron nanosheets dynamically and reversibly switch to clusters, enabling the direct conversion of methane to acetic acid with 97% selectivity and high activity without the requirement for carbon monoxide.
  •  

Why Sample What You Can Enumerate? Exact Policy Optimization for Genomic Tool Selection

arXiv:2609.10221v1 Announce Type: new Abstract: Reinforcement learning over a frozen reasoner has become a common recipe for teaching a policy which external tools to invoke. We show that this recipe becomes structurally mismatched in specialist scientific settings where the complete tool-subset space is enumerable. There, a small set of recurring computational capabilities covers the domain, so the space of tool subsets is combinatorial yet small enough to enumerate, and GRPO still estimates an action expectation from a handful of sampled rollouts. Worse, the approximation degrades as training succeeds: as the policy concentrates on preferred subsets it resamples them, sampled rewards collide, and the group-normalized advantage vanishes. On genomic reasoning the fraction of questions yielding no reward signal rises from 0.2% under a uniform reference policy to 20.8% after GRPO training. As a remedy, we introduce FGPO (Full-Group Policy Optimization), which (1) scores every tool subset and optimizes the exact action expectation, so each update sees the complete action space, and (2) precomputes the reward of each question--subset pair into an exhaustive table, removing frozen-reasoner calls from the training loop entirely. Across five frozen reasoners and three genomic benchmarks, FGPO outperforms GRPO in all 15 settings by 6.75 points on average and up to 14.20, while a standard on-demand GRPO schedule would require 2.4 times as many frozen-reasoner reward evaluations and, on GenomeQA, FGPO cuts invoked tools per question from 2.36 to 1.40.
  •  

EvolveScaler: Synthesizing Information-Evolution Contexts via Executable State Machines and Natural-Language Rendering

arXiv:2609.08435v2 Announce Type: replace Abstract: In persistent interactions, long contexts may encode an evolving process rather than a fixed record: later events can revise or revoke earlier information, changing what remains valid and what conclusions follow. We call this setting information evolution (IE). Solving IE requires identifying valid records, applying updates in order, and reconstructing the query-relevant state from the event history. Existing text-first synthesis pipelines make such data difficult to verify because state transitions and answer logic remain implicit. We introduce EvolveScaler, a code-driven framework that defines information evolution before rendering it as natural language. Human-authored operational specifications define state transitions, record validity, difficulty controls, and executable answer logic; a strong LLM then synthesizes a self-contained simulator from each specification. Executing validated simulators produces natural-language multi-turn event histories, while deterministic replay computes reference answers and atomic checklists. We instantiate EvolveScaler with 117 task prototypes and 159 final-question operators across five difficulty levels spanning approximately 7 to 1,200 events per instance, yielding about 35,100 training examples and 585 validated evaluation instances. On the very_long tier, the strongest model reaches 59.3% avg@5, while six models score below 10%. Training an internal A3B model on 6,000 EvolveScaler examples improves performance over its base checkpoint on all eight independently constructed out-of-distribution benchmarks, with a 5.25-point average gain. These results show that code-driven IE synthesis provides both challenging evaluation and transferable training supervision.
  •  

Neural Scalable Symbolic Search Framework for Complex Logical Queries with Multiple Free Variables

arXiv:2605.25985v1 Announce Type: new Abstract: Complex Query Answering (CQA) is a fundamental knowledge representation and reasoning task over incomplete knowledge graphs (KGs). Answering existential first-order queries with $k$ free variables (i.e., $\text{EFO}_k$ queries) is a crucial yet challenging problem, as it requires ranking answer tuples in $\mathcal{E}^k$, where $\mathcal{E}$ denotes the entity set of a KG. This quickly becomes intractable as $k$ grows. Consequently, existing benchmarks and methods rely on marginal rankings over individual variables; however, marginal rankings are a poor proxy for the true joint ranking of tuples. Building on neural symbolic search for $\text{EFO}_1$ queries, we propose Neural Scalable Symbolic Search (NS3), a budgeted framework that approximates joint ranking without enumerating $\mathcal{E}^k$. NS3 (i) answers marginalized sub-queries to obtain necessary candidate sets, (ii) merges multiple free variables into hypernodes whose domains are pruned and controlled by a dynamic budget $B$, and (iii) progressively reduces an $\text{EFO}_k$ query to an $\text{EFO}_{k-1}$ query over a budgeted reduced domain. Across three standard KG datasets, NS3 substantially improves joint ranking performance while retaining strong marginal accuracy. We further release a joint-ranking benchmark that extends existing $\text{EFO}_1$ datasets to $k=3$, enabling systematic evaluation of multi-variable queries. Our code is provided in https://github.com/HKUST-KnowComp/NS3_KDD2026.
  •  

Towards Evaluation Engineering: An Empirical Study of ML Evaluation Harnesses in the Wild

arXiv:2605.24213v1 Announce Type: cross Abstract: Evaluation harnesses are software systems that orchestrate model evaluation by managing model invocation, data loading, metric computation, and result reporting. Despite their critical role in machine learning infrastructure, their operational challenges and engineering concerns have received limited attention so far. We present an empirical study of 57 evaluation harnesses, deriving a five-stage harness model and classifying 16,560 issues by workflow stage and root cause. Most harness operational challenges concentrate in the Specification stage (41.4% of issues), where harnesses integrate external models, datasets, and scoring judges. The three most frequent root causes of operational challenges are unimplemented features (24.3%), documentation gaps (20.3%), and missing input validation (17.2%), which together account for 61.7% of classified issues, spanning both defects in existing functionality and capability gaps that block intended workflows. Root causes also vary by workflow stage: environment incompatibility and external dependency breakage account for 36.2% of provisioning issues, whereas algorithmic error (25.9%) and validation gap (22.5%) dominate assessment issues. Together, these contributions establish an empirical foundation for treating evaluation engineering as a distinct software engineering concern.
  •  

CollectionLoRA: Collecting 50 Effects in 1 LoRA via Multi-Teacher On-Policy Distillation

arXiv:2605.25378v1 Announce Type: cross Abstract: Customized image editing aims to equip pre-trained diffusion models with specific visual effects using limited paired data, typically via Low-Rank Adaptation (LoRA). As the number of desired effects grows, storing and dynamically loading numerous these effect LoRAs significantly increases deployment overhead. Furthermore, current pipelines typically cascade these effect LoRAs with acceleration modules for fast generation, which triggers severe parameter interference and results in concept bleeding and style degradation. We propose CollectionLoRA, a multi-teacher on-policy distillation framework capable of distilling the concepts of up to 50 different effect LoRAs along with few-step generation capabilities into a single LoRA. This fundamentally resolves the feature interference issue and significantly reduces deployment costs. Specifically, the method introduces (i) a Probabilistic Dual-Stream Routing mechanism that enables the model to randomly switch between data sources during training, effectively enhancing its generalization in unseen scenarios; (ii) an Asymmetric Orthogonal Prompting strategy to achieve concept isolation within the prompt space; (iii) a Coarse-to-Fine Distillation Objective to mitigate the distribution gap between the teacher and student models. Extensive evaluations show that CollectionLoRA distills all customized effects and few-step generation into a single LoRA, reducing deployment overhead while achieving concept fidelity comparable to or better than independently trained teacher models.
  •  

Efficient and Scalable Neural Symbolic Search for Knowledge Graph Complex Query Answering

arXiv:2505.08155v4 Announce Type: replace Abstract: Complex Query Answering (CQA) is a crucial reasoning task over Knowledge Graphs (KGs), which aims to answer first-order logical queries from incomplete KGs. While existing neural-symbolic methods achieve strong performance, they face significant complexity bottlenecks: quadratic data complexity scaling with the number of entities, and NP-hard query complexity for cyclic queries. Consequently, these approaches struggle to scale effectively to large knowledge graphs and complex queries. To address these limitations, we propose an efficient and scalable symbolic search method comprising two key components: (1) constraint strategies that drastically reduce the variable search domain, lowering data complexity; and (2) a local search algorithm that approximately solves NP-hard cyclic queries. Experiments on various CQA benchmarks demonstrate that, for tree-form queries, our method achieves 97% relative MRR with a 10$\times$ speedup using only 10% of the search space. Furthermore, it demonstrates robust performance on complex cyclic queries and large-scale KGs, effectively alleviating efficiency and scalability challenges. Our code is provided in https://github.com/HKUST-KnowComp/NLISA_KDD2026.
  •  

Kolmogorov-Arnold Fourier Networks

arXiv:2502.06018v3 Announce Type: replace-cross Abstract: Although Kolmogorov-Arnold-based interpretable networks (KANs) possess strong theoretical expressiveness, they suffer from severe parameter explosion and limited ability to capture high-frequency features in high-dimensional tasks. To address these issues, we propose the Kolmogorov-Arnold Fourier Network (KAF), which fundamentally redefines the KAN paradigm through spectral reparameterization. Our key contributions include: (1) proposing a fundamental basis transformation from the local, grid-based B-spline representation to a global, adaptive spectral representation. This shift changes the network's inductive bias, reducing parameter complexity from $O(G)$ to $O(1)$ while preserving expressiveness; (2) introducing trainable Random Fourier Features (RFF) initialized via a spectral alignment strategy, which allows the model to break the smoothness limitation of fixed kernels and accurately capture high-frequency components; and (3) implementing an adaptive hybrid GELU-Fourier activation mechanism that progressively enhances frequency representation during training. Comprehensive experiments demonstrate the superiority of KAF across computer vision (CV), natural language processing (NLP), audio, and partial differential equation (PDE) solving tasks, achieving state-of-the-art performance with improved efficiency. The code is available at https://github.com/kolmogorovArnoldFourierNetwork/KAF.
  •  

PiXTime: A Model for Federated Time Series Forecasting with Heterogeneous Data across Nodes

arXiv:2601.05613v2 Announce Type: replace-cross Abstract: While collaborative forecasting on distributed time series is highly desirable, directly pooling localized datasets is often impractical due to data sharing constraints. Federated learning offers a promising alternative, yet conventional federated learning algorithms require homogeneous model architectures, which are incompatible with the structural discrepancies, such as unaligned temporal resolutions and mismatched variable channels, commonly observed across decentralized nodes. To bridge this gap, we introduce PiXTime, a novel Transformer-based framework designed to natively accommodate and leverage structurally heterogeneous temporal data. At its core, PiXTime adopts a parameter-decoupling architecture, strategically partitioning the model into localized personalized modules and a globally aggregated shared backbone. Specifically, node-specific local modules act as dimensional adapters, projecting raw sequences of diverse lengths into a unified representation space. Concurrently, a globally synchronized VE Table injects consistent categorical identities into the feature space, allowing the shared backbone to collaboratively learn and generalize representations across inconsistent variable distributions. Comprehensive evaluations on multiple benchmarks demonstrate that PiXTime achieves state-of-the-art performance in heterogeneous federated environments, while maintaining robust superiority in standard homogeneous and centralized forecasting settings.
  •  

BEAR: Towards Beam-Search-Aware Optimization for Recommendation with Large Language Models

arXiv:2601.22925v3 Announce Type: replace-cross Abstract: Recent years have seen a rapid surge in research leveraging Large Language Models (LLMs) for recommendation. These methods typically employ supervised fine-tuning (SFT) to adapt LLMs to recommendation scenarios, and utilize beam search during inference to efficiently retrieve $B$ top-ranked recommended items. However, we identify a critical training-inference inconsistency: while SFT optimizes the overall probability of positive items, it does not guarantee that such items will be retrieved by beam search even if they possess high overall probabilities. Due to the greedy pruning mechanism, beam search can prematurely discard a positive item once its prefix probability is insufficient. To address this inconsistency, we propose BEAR (Beam-SEarch-Aware Regularization), a novel fine-tuning objective that explicitly accounts for beam search behavior during training. Rather than directly simulating beam search for each instance during training, which is computationally prohibitive, BEAR enforces a relaxed necessary condition: each token in a positive item must rank within the top-$B$ candidate tokens at each decoding step. This objective effectively mitigates the risk of incorrect pruning while incurring negligible computational overhead compared to standard SFT. Extensive experiments across four real-world datasets demonstrate that BEAR significantly outperforms strong baselines. Code is available at https://github.com/Tiny-Snow/BEAR-SIGIR-2026 .
  •  

Prism: Spectral-Aware Block-Sparse Attention

arXiv:2602.08426v2 Announce Type: replace-cross Abstract: Block-sparse attention is promising for accelerating long-context LLM pre-filling, yet identifying relevant blocks efficiently remains a bottleneck. Existing methods typically employ coarse-grained attention as a proxy for block importance estimation, but often resort to expensive token-level searching or scoring, resulting in significant selection overhead. In this work, we trace the inaccuracy of standard coarse-grained attention via mean pooling to a theoretical root cause: the interaction between mean pooling and Rotary Positional Embeddings (RoPE). We prove that mean pooling acts as a low-pass filter that induces destructive interference in high-frequency dimensions, effectively creating a "blind spot" for local positional information (e.g., slash patterns). To address this, we introduce Prism, a training-free spectral-aware approach that decomposes block selection into high-frequency and low-frequency branches. By applying energy-based temperature calibration, Prism restores the attenuated positional signals directly from pooled representations, enabling block importance estimation using purely block-level operations, thereby improving efficiency. Extensive evaluations confirm that Prism maintains accuracy parity with full attention while delivering up to $\mathbf{5.1\times}$ speedup.
  •  
❌