❌

Reading view

↗️

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

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Xiangrui Yang Β· Cheng Peng Β· Yunfeng Zhao Β· Liang Zeng Β· Ao Hu Β· Jiawei Yang Β· Shengzhe Wang Β· Jingshan Lv Β· Xiao Liang Β· Chen Yang Β· Jiaqiang Liu Β· Yiming Qiu
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.
  •  
  • ↗️
↗️

Who Judges the Judges? A Chinese Safety QA Benchmark for Evaluating LLM Responses and Safety Judges

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Rui Yang Β· Shuang Huang Β· Junhua Liu Β· Ziqi Zhao Β· Qingzhong Yan Β· Yuhang Sun Β· Cong Liu Β· Guoping Hu Β· Rui Mei Β· Jing Shao
arXiv:2609.01210v2 Announce Type: replace-cross Abstract: Safety benchmarks for large language models often assess the risk of a user query, although the outcome of question answering depends on whether the response violates a policy. This distinction is critical in Chinese harmful-content evaluation, where linguistic variation and adversarial transformations can obscure risky intent. We introduce C-SafeQA, a policy-grounded benchmark for response-level Chinese safety evaluation. It comprises 538 base queries and 8,877 adversarial queries answered by four full-model LLM deployments, yielding 37,660 query-response records labeled safe, unsafe, or disputed. Reference labels are generated through agreement-aware multi-model adjudication and blind audits of stratified subsets by three safety experts. C-SafeQA supports both evaluation of target-model safety and auditing of seven automated safety judges against shared reference labels. Unsafe-response rates range from 0.93% to 3.35% on base queries and from 11.68% to 30.05% on adversarial queries. On the adversarial subset, judges show substantial trade-offs between unsafe-response recall and risk-query-conditioned safe-response false positive rate, and no judge dominates all metrics. Both acrostic transformations reduce unsafe recall for all seven judges, revealing mechanism-specific evaluator weaknesses. Dataset records, metadata, verification code, and judge scripts are publicly released to support recomputation, while benchmark construction, target-response generation, and private adjudication remain outside the release boundary.
  •  
  • ↗️
↗️

Integrated multi-omics analysis of metabolomics and proteomics uncovers dysregulated amino acid metabolism in HCC metastasis

βœ‡Omics in Hepatocellular
By: Jing Wu Β· Jin Wang Β· Zhen Yu Β· Rui Yang

Front Immunol. 2026 Aug 19;17:1856643. doi: 10.3389/fimmu.2026.1856643. eCollection 2026.

ABSTRACT

BACKGROUND: Metastasis is the primary cause of treatment failure and adverse prognosis in hepatocellular carcinoma (HCC), and the molecular basis of HCC metastasis remains poorly defined. This work investigated the potential mechanisms underlying HCC metastasis through integrated multi-omics analysis of metabolomics and proteomics.

METHOD: This retrospective study included 105 individuals with HCC, with comparative analysis between metastatic and non-metastatic cases. We further evaluated the effects of metastasis on serum metabolomics and proteomics in HCC patients.

RESULT: Widespread disturbances in amino acid metabolism were identified via untargeted metabolomics in HCC patients with metastasis, closely governing inflammation-related metabolic remodeling and oxidative stress responses. Specifically, we identified 91 and 59 distinct differential metabolites capable of indicating HCC metastasis, with the screening criteria set as log2 fold change > 1.5, adjusted P value < 0.05, and VIP > 1.5 in positive and negative modes, respectively. The alanine, aspartate and glutamate metabolism pathway correlated with HCC-associated lung metastasis, while the gluconeogenesis pathway was linked to HCC-associated bone metastasis. Compared with HCC (non-metastatic hepatocellular carcinoma), the key molecular alterations in the multi-omics network of HCC_M (HCC with metastasis) are implicated in inflammatory metabolic reprogramming, oxidative stress response, gluconeogenesis, glycolysis, and the tricarboxylic acid (TCA) cycle. Twenty-five proteins, including PKM2, PERCK, ALDH2, CPS1, GLS1, GLUD1, GOT1, and SLC38A2, were identified as potential biomarkers for HCC metastasis.

CONCLUSION: By integrating untargeted metabolomic and proteomic profiling, we identified distinct metabolic and proteomic changes linked to HCC metastasis. This work also characterized the pathological characteristics and core pathways underlying HCC metastasis, while identifying potential therapeutic candidates.

PMID:42688489 | PMC:PMC13534100 | DOI:10.3389/fimmu.2026.1856643

  •  
  • ↗️
↗️

MoBiQuant: Mixture-of-Bits Quantization for Token-Adaptive Any-Precision LLM

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Dongwei Wang Β· Jinhee Kim Β· Seokho Han Β· Denis Gudovskiy Β· Yohei Nakata Β· Tomoyuki Okuno Β· KhayTze Peong Β· Kang Eun Jeon Β· Jong Hwan Ko Β· Yiran Chen Β· Huanrui Yang
arXiv:2602.20191v2 Announce Type: replace-cross Abstract: Dynamic runtime latency and memory constraints necessitate flexible large language model (LLM) deployment, where an LLM can be inferred with various quantization precisions based on available computational resources. Recent work on such any-precision quantization either relies on hardware-inefficient vector quantization or induces additional scaling factors when switching between bit-widths. Meanwhile, existing post-training quantization (PTQ) methods calibrated for a fixed low precision show poor generalizability under runtime precision change. In this work, we attribute the source of poor generalization across bit-widths to a precision-dependent \textit{outlier migration} phenomenon where the distribution of PTQ-sensitive tokens changes across precisions. Motivated by this observation, we propose \texttt{MoBiQuant}, a novel any-precision Mixture-of-Bits quantization framework that adjusts weight precision for flexible LLM inference based on token sensitivity. Specifically, we propose a many-in-one recursive residual quantization that can iteratively reconstruct higher-precision weights at runtime and mitigates \textit{outlier migration} with a token-aware router to dynamically select the optimal inference precision of each token.Extensive experiments show that \texttt{MoBiQuant} matches or surpasses frontier single-precision PTQ while exhibiting strong elasticity, achieving significant memory savings and throughput gains of up to $1.34\times$ over state-of-the-art any-precision methods.
  •  
  • ↗️
↗️

A SAUR gene enhances maize drought resilience by promoting silk elongation

βœ‡Nature - Issue - nature.com science feeds
By: Chaohui Zhu Β· Zhirui Yang Β· Shiping Yang Β· Xueyan Zhou Β· Boxin Liu Β· Tian Tian Β· Bochen Zhao Β· Yingying Xie Β· Yujun Liu Β· Jinkui Cheng Β· Huaijun Tang Β· Yanjun Zhang Β· Xiaoqing Xie Β· Lei Zhang Β· Cheng Liu Β· Xingrong Wang Β· Shengxue Liu Β· Feng Qin

Nature, Published online: 20 May 2026; doi:10.1038/s41586-026-10566-9

The Small Auxin Up RNA (SAUR) protein ZmSAUR72 in maize (Zea mays) promotes silk growth via regulation of H+-ATPase activity, and is a key determinant of the anthesis-silking interval and thus resilience to drought.
  •  
  • ↗️
↗️

SemLoc: Structured Grounding of Free-Form LLM Reasoning for Fault Localization

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Zhaorui Yang Β· Haichao Zhu Β· Qian Zhang Β· Rajiv Gupta Β· Ashish Kundu
arXiv:2603.29109v1 Announce Type: cross Abstract: Fault localization identifies program locations responsible for observed failures. Existing techniques rank suspicious code using syntactic spectra--signals derived from execution structure such as statement coverage, control-flow divergence, or dependency reachability. These signals collapse for semantic bugs, where failing and passing executions follow identical code paths and differ only in whether semantic intent is satisfied. Recent LLM-based approaches introduce semantic reasoning but produce stochastic, unverifiable outputs that cannot be systematically cross-referenced across tests or distinguish root causes from cascading effects. We present SemLoc, a fault localization framework based on structured semantic grounding. SemLoc converts free-form LLM reasoning into a closed intermediate representation that binds each inferred property to a typed program anchor, enabling runtime checking and attribution to program structure. It executes instrumented programs to construct a semantic violation spectrum--a constraint-by-test matrix--from which suspiciousness scores are derived analogously to coverage-based methods. A counterfactual verification step further prunes over-approximate constraints and isolates primary causal violations. We evaluate SemLoc on SemFault-250, a corpus of 250 Python programs with single semantic faults. SemLoc outperforms five coverage-, reduction-, and LLM-based baselines, achieving Top-1 accuracy of 42.8% and Top-3 of 68%, while reducing inspection to 7.6% of executable lines. Counterfactual verification provides an additional 12% accuracy gain and identifies primary causal semantic constraints.
  •  
  • ↗️
↗️

Unannotated noncoding transcripts as a source of intratumor heterogeneity in malignant cell states

βœ‡(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Xiaolan Zhou Β· Limin Lin Β· Yu Chen Β· Guangrui Yang Β· Ting Ni Β· Zhaozhao Zhao

Sci China Life Sci. 2026 Mar 16. doi: 10.1007/s11427-025-3273-6. Online ahead of print.

ABSTRACT

Phenotypic diversity of malignant cells within a tumor underlies intratumor heterogeneity (ITH), a key determinant of cancer metastasis and treatment failure. However, the molecular mechanisms driving this heterogeneity are poorly understood. Here, we curated and analyzed a cohort of 3' tag-based single-cell RNA-seq covering 12 common cancer types. We identified thousands of poly(A) site (PAS) peaks representing the 3' ends of previously unannotated transcripts, whose expression is widely associated with diverse malignant cellular states. By integrating multi-omics data, we characterized the expression patterns and epigenetic landscape of these unannotated PAS peak-associated transcripts (UPTs). The expression heterogeneity of UPTs was supported by multi-region sampling bulk RNA-seq data and recapitulated within cancer cell lines. As proof of principle validation, functional experiments confirmed that two noncoding UPTs promoted the proliferation and migration of lung cancer cells. Our results suggest that epigenetic activation of unannotated noncoding transcripts might represent a previously unrecognized mechanism contributing to transcriptomic ITH.

PMID:41870780 | DOI:10.1007/s11427-025-3273-6

  •  
  • ↗️
↗️

Unannotated noncoding transcripts as a source of intratumor heterogeneity in malignant cell states

βœ‡Omics In Lung
By: Xiaolan Zhou Β· Limin Lin Β· Yu Chen Β· Guangrui Yang Β· Ting Ni Β· Zhaozhao Zhao

Sci China Life Sci. 2026 Mar 16. doi: 10.1007/s11427-025-3273-6. Online ahead of print.

ABSTRACT

Phenotypic diversity of malignant cells within a tumor underlies intratumor heterogeneity (ITH), a key determinant of cancer metastasis and treatment failure. However, the molecular mechanisms driving this heterogeneity are poorly understood. Here, we curated and analyzed a cohort of 3' tag-based single-cell RNA-seq covering 12 common cancer types. We identified thousands of poly(A) site (PAS) peaks representing the 3' ends of previously unannotated transcripts, whose expression is widely associated with diverse malignant cellular states. By integrating multi-omics data, we characterized the expression patterns and epigenetic landscape of these unannotated PAS peak-associated transcripts (UPTs). The expression heterogeneity of UPTs was supported by multi-region sampling bulk RNA-seq data and recapitulated within cancer cell lines. As proof of principle validation, functional experiments confirmed that two noncoding UPTs promoted the proliferation and migration of lung cancer cells. Our results suggest that epigenetic activation of unannotated noncoding transcripts might represent a previously unrecognized mechanism contributing to transcriptomic ITH.

PMID:41870780 | DOI:10.1007/s11427-025-3273-6

  •  
  • ↗️
↗️

FAPE-IR: Frequency-Aware Planning and Execution Framework for All-in-One Image Restoration

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Jingren Liu Β· Shuning Xu Β· Qirui Yang Β· Yun Wang Β· Xiangyu Chen Β· Zhong Ji
arXiv:2511.14099v3 Announce Type: replace-cross Abstract: All-in-One Image Restoration (AIO-IR) aims to develop a unified model that can handle multiple degradations under complex conditions. However, existing methods often rely on task-specific designs or latent routing strategies, making it hard to adapt to real-world scenarios with various degradations. We propose FAPE-IR, a Frequency-Aware Planning and Execution framework for image restoration. It uses a frozen Multimodal Large Language Model (MLLM) as a planner to analyze degraded images and generate concise, frequency-aware restoration plans. These plans guide a LoRA-based Mixture-of-Experts (LoRA-MoE) module within a diffusion-based executor, which dynamically selects high- or low-frequency experts, complemented by frequency features of the input image. To further improve restoration quality and reduce artifacts, we introduce adversarial training and a frequency regularization loss. By coupling semantic planning with frequency-based restoration, FAPE-IR offers a unified and interpretable solution for all-in-one image restoration. Extensive experiments show that FAPE-IR achieves state-of-the-art performance across seven restoration tasks and exhibits strong zero-shot generalization under mixed degradations.
  •  
  • ↗️
↗️

BEAT: Visual Backdoor Attacks on VLM-based Embodied Agents via Contrastive Trigger Learning

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Qiusi Zhan Β· Hyeonjeong Ha Β· Rui Yang Β· Sirui Xu Β· Hanyang Chen Β· Liang-Yan Gui Β· Yu-Xiong Wang Β· Huan Zhang Β· Heng Ji Β· Daniel Kang
arXiv:2510.27623v3 Announce Type: replace Abstract: Recent advances in Vision-Language Models (VLMs) have propelled embodied agents by enabling direct perception, reasoning, and planning task-oriented actions from visual inputs. However, such vision-driven embodied agents open a new attack surface: visual backdoor attacks, where the agent behaves normally until a visual trigger appears in the scene, then persistently executes an attacker-specified multi-step policy. We introduce BEAT, the first framework to inject such visual backdoors into VLM-based embodied agents using objects in the environments as triggers. Unlike textual triggers, object triggers exhibit wide variation across viewpoints and lighting, making them difficult to implant reliably. BEAT addresses this challenge by (1) constructing a training set that spans diverse scenes, tasks, and trigger placements to expose agents to trigger variability, and (2) introducing a two-stage training scheme that first applies supervised fine-tuning (SFT) and then our novel Contrastive Trigger Learning (CTL). CTL formulates trigger discrimination as preference learning between trigger-present and trigger-free inputs, explicitly sharpening the decision boundaries to ensure precise backdoor activation. Across various embodied agent benchmarks and VLMs, BEAT achieves attack success rates up to 80%, while maintaining strong benign task performance, and generalizes reliably to out-of-distribution trigger placements. Notably, compared to naive SFT, CTL boosts backdoor activation accuracy up to 39% under limited backdoor data. These findings expose a critical yet unexplored security risk in VLM-based embodied agents, underscoring the need for robust defenses before real-world deployment.
  •  
  • ↗️
There are no more articles
❌