❌

Normal view

Received — 4 April 2026 ⏭

Unifying Group-Relative and Self-Distillation Policy Optimization via Sample Routing

✇cs.AI, q-bio.NC updates on arXiv.org
By: Gengsheng Li · Tianyu Yang · Junfeng Fang · Mingyang Song · Mao Zheng · Haiyun Guo · Dan Zhang · Jinqiao Wang · Tat-Seng Chua
3 April 2026 at 12:00
arXiv:2604.02288v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has become a standard paradigm for post-training large language models. While Group Relative Policy Optimization (GRPO) is widely adopted, its coarse credit assignment uniformly penalizes failed rollouts, lacking the token-level focus needed to efficiently address specific deviations. Self-Distillation Policy Optimization (SDPO) addresses this by providing denser, more targeted logit-level supervision that facilitates rapid early improvement, yet it frequently collapses during prolonged training. We trace this late-stage instability to two intrinsic flaws: self-distillation on already-correct samples introduces optimization ambiguity, and the self-teacher's signal reliability progressively degrades. To resolve these issues, we propose Sample-Routed Policy Optimization (SRPO), a unified on-policy framework that routes correct samples to GRPO's reward-aligned reinforcement and failed samples to SDPO's targeted logit-level correction. SRPO further incorporates an entropy-aware dynamic weighting mechanism to suppress high-entropy, unreliable distillation targets while emphasizing confident ones. Evaluated across five benchmarks and two model scales, SRPO achieves both the rapid early improvement of SDPO and the long-horizon stability of GRPO. It consistently surpasses the peak performance of both baselines, raising the five-benchmark average on Qwen3-8B by 3.4% over GRPO and 6.3% over SDPO, while simultaneously yielding moderate response lengths and lowering per-step compute cost by up to 17.2%.
  • ✇cs.AI, q-bio.NC updates on arXiv.org
  • Agentified Assessment of Logical Reasoning Agents Zhiyu Ni · Yifeng Xiao · Zheng Liang
    arXiv:2603.02788v4 Announce Type: replace Abstract: We present a framework for evaluating and benchmarking logical reasoning agents when assessment itself must be reproducible, auditable, and robust to execution failures. Building on agentified assessment, we use an assessor agent to issue tasks, enforce execution budgets, parse outputs, and record structured failure types, while the agent under test only needs to expose a standardized agent-to-agent interface. As a case study, we benchmark an
     
  • 🏷️
  • ↗️

Agentified Assessment of Logical Reasoning Agents

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zhiyu Ni · Yifeng Xiao · Zheng Liang
3 April 2026 at 12:00
arXiv:2603.02788v4 Announce Type: replace Abstract: We present a framework for evaluating and benchmarking logical reasoning agents when assessment itself must be reproducible, auditable, and robust to execution failures. Building on agentified assessment, we use an assessor agent to issue tasks, enforce execution budgets, parse outputs, and record structured failure types, while the agent under test only needs to expose a standardized agent-to-agent interface. As a case study, we benchmark an auto-formalization agent for first-order logic (FOL) reasoning on a solver-verified and repaired split of FOLIO. The agent translates natural language premises and conclusions into executable Z3Py programs and employs satisfiability modulo theories (SMT) solving to determine logical entailment. On the cleaned FOLIO validation set, the auto-formalization agent achieves 86.70% accuracy under the assessor protocol, outperforming a chain-of-thought baseline (73.89%).

Labels Matter More Than Models: Rethinking the Unsupervised Paradigm in Time Series Anomaly Detection

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zhijie Zhong · Zhiwen Yu · Kaixiang Yang · Yongheng Liu · Jun Jiang · C. L. Philip Chen
3 April 2026 at 12:00
arXiv:2511.16145v2 Announce Type: replace-cross Abstract: Time series anomaly detection (TSAD) is a critical data mining task often constrained by label scarcity. Consequently, current research predominantly focuses on Unsupervised Time-series Anomaly Detection (UTAD), relying on increasingly complex architectures to model normal data distributions. However, this algorithm-centric trend often overlooks the significant performance gains achievable from limited anomaly labels available in practical scenarios. This paper challenges the premise that algorithmic complexity is the optimal path for TSAD. Instead of proposing another intricate unsupervised model, we present a comprehensive benchmark and empirical study to rigorously compare supervised and unsupervised paradigms. To isolate the value of labels, we introduce \stand, a deliberately minimalist supervised baseline. Extensive experiments on five public datasets demonstrate that: (1) Labels matter more than models: under a limited labeling budget, simple supervised models significantly outperform complex state-of-the-art unsupervised methods; (2) Supervision yields higher returns: the performance gain from minimal supervision far exceeds the incremental gains from architectural innovations; and (3) Practicality: \stand~exhibits superior prediction consistency and anomaly localization compared to unsupervised counterparts. These findings advocate for a paradigm shift in TSAD research, urging the community to prioritize data-centric label utilization over purely algorithmic complexity. The code and benchmark are publicly available at https://github.com/EmorZz1G/STAND.

Semantic Refinement with LLMs for Graph Representations

✇cs.AI, q-bio.NC updates on arXiv.org
By: Safal Thapaliya · Zehong Wang · Jiazheng Li · Ziming Li · Yanfang Ye · Chuxu Zhang
3 April 2026 at 12:00
arXiv:2512.21106v2 Announce Type: replace-cross Abstract: Graph-structured data exhibit substantial heterogeneity in where their predictive signals originate: in some domains, node-level semantics dominate, while in others, structural patterns play a central role. This structure-semantics heterogeneity implies that no graph learning model with a fixed inductive bias can generalize optimally across diverse graph domains. However, most existing methods address this challenge from the model side by incrementally injecting new inductive biases, which remains fundamentally limited given the open-ended diversity of real-world graphs. In this work, we take a data-centric perspective and treat node semantics as a task-adaptive variable. We propose a Graph-Exemplar-guided Semantic Refinement (GES) framework for graph representation learning which -- unlike existing LLM-enhanced methods that generate node descriptions without graph context -- leverages structurally and semantically similar nodes from the graph itself to guide semantic refinement. Specifically, a GNN is first trained to produce predictive states, which along with structural and semantic similarity are used to retrieve in-graph exemplars that inform an LLM in refining node descriptions. We evaluate our approach on both text-rich and text-free graphs. Results show consistent improvements on semantics-rich and structure-dominated graphs, demonstrating the effectiveness of data-centric semantic refinement under structure-semantics heterogeneity.

MemFactory: Unified Inference & Training Framework for Agent Memory

✇cs.AI, q-bio.NC updates on arXiv.org
By: Ziliang Guo · Ziheng Li · Bo Tang · Feiyu Xiong · Zhiyu Li
3 April 2026 at 12:00
arXiv:2603.29493v3 Announce Type: replace-cross Abstract: Memory-augmented Large Language Models (LLMs) are essential for developing capable, long-term AI agents. Recently, applying Reinforcement Learning (RL) to optimize memory operations, such as extraction, updating, and retrieval, has emerged as a highly promising research direction. However, existing implementations remain highly fragmented and task-specific, lacking a unified infrastructure to streamline the integration, training, and evaluation of these complex pipelines. To address this gap, we present MemFactory, the first unified, highly modular training and inference framework specifically designed for memory-augmented agents. Inspired by the success of unified fine-tuning frameworks like LLaMA-Factory, MemFactory abstracts the memory lifecycle into atomic, plug-and-play components, enabling researchers to seamlessly construct custom memory agents via a "Lego-like" architecture. Furthermore, the framework natively integrates Group Relative Policy Optimization (GRPO) to fine-tune internal memory management policies driven by multi-dimensional environmental rewards. MemFactory provides out-of-the-box support for recent cutting-edge paradigms, including Memory-R1, RMM, and MemAgent. We empirically validate MemFactory on the open-source MemAgent architecture using its publicly available training and evaluation data. Across the evaluation sets, MemFactory improves performance over the corresponding base models on average, with relative gains of up to 14.8%. By providing a standardized, extensible, and easy-to-use infrastructure, MemFactory significantly lowers the barrier to entry, paving the way for future innovations in memory-driven AI agents.

Leptomeningeal metastatic cancer cells induce a permissive choroid plexus vasculature through extracellular-vesicle-derived 5-HIAA signaling

✇Nature Cancer
By: Yang Huang · Yifan Hou · Jiahui Yang · Wanqi Yang · Hanxin Liu · Haibao Peng · Xiaohui Li · Jiaxu Zhao · Ye Zhang · Rui Zeng · Youming Zeng · Cheng Li · Qiangqiang Zhang · Yudan Chi
3 April 2026 at 08:00

Nature Cancer, Published online: 03 April 2026; doi:10.1038/s43018-026-01145-y

Huang, Hou, Yang et al. demonstrate that leptomeningeal metastatic cells favor the formation of a premetastatic niche by remodeling the choroid plexus vasculature through the serotonin metabolite 5-hydroxyindoleacetic acid, which signals into endothelial cells through the aryl hydrocarbon receptor.
Received — 2 April 2026 ⏭

Correction: Integrative multi-omics and machine learning reveals the spatial niche distribution and role of CYP27A1+TAMs in immunotherapy response in non-small cell lung cancer

✇Omics In Lung
By: Qingsheng Liu · Xufeng Liu · Han Zhang · Yuhang Jiang · Ying Shi · Qiuqiao Mu · Yuhao Jing · Daqiang Sun
1 April 2026 at 18:00

Front Immunol. 2026 Mar 16;17:1822612. doi: 10.3389/fimmu.2026.1822612. eCollection 2026.

ABSTRACT

[This corrects the article DOI: 10.3389/fimmu.2026.1782545.].

PMID:41918731 | PMC:PMC13033988 | DOI:10.3389/fimmu.2026.1822612

Correction: Integrative multi-omics and machine learning reveals the spatial niche distribution and role of CYP27A1+TAMs in immunotherapy response in non-small cell lung cancer

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Qingsheng Liu · Xufeng Liu · Han Zhang · Yuhang Jiang · Ying Shi · Qiuqiao Mu · Yuhao Jing · Daqiang Sun
1 April 2026 at 18:00

Front Immunol. 2026 Mar 16;17:1822612. doi: 10.3389/fimmu.2026.1822612. eCollection 2026.

ABSTRACT

[This corrects the article DOI: 10.3389/fimmu.2026.1782545.].

PMID:41918731 | PMC:PMC13033988 | DOI:10.3389/fimmu.2026.1822612

Predicting Neuromodulation Outcome for Parkinson's Disease with Generative Virtual Brain Model

✇cs.AI, q-bio.NC updates on arXiv.org
By: Siyuan Du · Siyi Li · Shuwei Bai · Ang Li · Haolin Li · Mingqing Xiao · Yang Pan · Dongsheng Li · Weidi Xie · Yanfeng Wang · Ya Zhang · Chencheng Zhang · Jiangchao Yao
1 April 2026 at 12:00
arXiv:2603.29176v1 Announce Type: new Abstract: Parkinson's disease (PD) affects over ten million people worldwide. Although temporal interference (TI) and deep brain stimulation (DBS) are promising therapies, inter-individual variability limits empirical treatment selection, increasing non-negligible surgical risk and cost. Previous explorations either resort to limited statistical biomarkers that are insufficient to characterize variability, or employ AI-driven methods which is prone to overfitting and opacity. We bridge this gap with a pretraining-finetuning framework to predict outcomes directly from resting-state fMRI. Critically, a generative virtual brain foundation model, pretrained on a collective dataset (2707 subjects, 5621 sessions) to capture universal disorder patterns, was finetuned on PD cohorts receiving TI (n=51) or DBS (n=55) to yield individualized virtual brains with high fidelity to empirical functional connectivity (r=0.935). By constructing counterfactual estimations between pathological and healthy neural states within these personalized models, we predicted clinical responses (TI: AUPR=0.853; DBS: AUPR=0.915), substantially outperforming baselines. External and prospective validations (n=14, n=11) highlight the feasibility of clinical translation. Moreover, our framework provides state-dependent regional patterns linked to response, offering hypothesis-generating mechanistic insights.

Route-Induced Density and Stability (RIDE): Controlled Intervention and Mechanism Analysis of Routing-Style Meta Prompts on LLM Internal States

✇cs.AI, q-bio.NC updates on arXiv.org
By: Dianxing Zhang · Gang Li · Sheng Li
1 April 2026 at 12:00
arXiv:2603.29206v1 Announce Type: new Abstract: Routing is widely used to scale large language models, from Mixture-of-Experts gating to multi-model/tool selection. A common belief is that routing to a task ``expert'' activates sparser internal computation and thus yields more certain and stable outputs (the Sparsity--Certainty Hypothesis). We test this belief by injecting routing-style meta prompts as a textual proxy for routing signals in front of frozen instruction-tuned LLMs. We quantify (C1) internal density via activation sparsity, (C2) domain-keyword attention, and (C3) output stability via predictive entropy and semantic variation. On a RouterEval subset with three instruction-tuned models (Qwen3-8B, Llama-3.1-8B-Instruct, and Mistral-7B-Instruct-v0.2), meta prompts consistently densify early/middle-layer representations rather than increasing sparsity; natural-language expert instructions are often stronger than structured tags. Attention responses are heterogeneous: Qwen/Llama reduce keyword attention, while Mistral reinforces it. Finally, the densification--stability link is weak and appears only in Qwen, with near-zero correlations in Llama and Mistral. We present RIDE as a diagnostic probe for calibrating routing design and uncertainty estimation.

iPoster: Content-Aware Layout Generation for Interactive Poster Design via Graph-Enhanced Diffusion Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Xudong Zhou · Jinyuan Liang · Qiuyi Guo · Guozheng Li
1 April 2026 at 12:00
arXiv:2603.29469v1 Announce Type: cross Abstract: We present iPoster, an interactive layout generation framework that empowers users to guide content-aware poster layout design by specifying flexible constraints. iPoster enables users to specify partial intentions within the intention module, such as element categories, sizes, positions, or coarse initial drafts. Then, the generation module instantly generates refined, context-sensitive layouts that faithfully respect these constraints. iPoster employs a unified graph-enhanced diffusion architecture that supports various design tasks under user-specified constraints. These constraints are enforced through masking strategies that precisely preserve user input at every denoising step. A cross content-aware attention module aligns generated elements with salient regions of the canvas, ensuring visual coherence. Extensive experiments show that iPoster not only achieves state-of-the-art layout quality, but offers a responsive and controllable framework for poster layout design with constraints.

MemFactory: Unified Inference & Training Framework for Agent Memory

✇cs.AI, q-bio.NC updates on arXiv.org
By: Ziliang Guo · Ziheng Li · Zhiyu Li
1 April 2026 at 12:00
arXiv:2603.29493v1 Announce Type: cross Abstract: Memory-augmented Large Language Models (LLMs) are essential for developing capable, long-term AI agents. Recently, applying Reinforcement Learning (RL) to optimize memory operations, such as extraction, updating, and retrieval, has emerged as a highly promising research direction. However, existing implementations remain highly fragmented and task-specific, lacking a unified infrastructure to streamline the integration, training, and evaluation of these complex pipelines. To address this gap, we present MemFactory, the first unified, highly modular training and inference framework specifically designed for memory-augmented agents. Inspired by the success of unified fine-tuning frameworks like LLaMA-Factory, MemFactory abstracts the memory lifecycle into atomic, plug-and-play components, enabling researchers to seamlessly construct custom memory agents via a "Lego-like" architecture. Furthermore, the framework natively integrates Group Relative Policy Optimization (GRPO) to fine-tune internal memory management policies driven by multi-dimensional environmental rewards. MemFactory provides out-of-the-box support for recent cutting-edge paradigms, including Memory-R1, RMM, and MemAgent. We empirically validate MemFactory on the open-source MemAgent architecture using its publicly available training and evaluation data. Across both in-domain and out-of-distribution evaluation sets, MemFactory consistently improves performance over the corresponding base models, with relative gains of up to 14.8%. By providing a standardized, extensible, and easy-to-use infrastructure, MemFactory significantly lowers the barrier to entry, paving the way for future innovations in memory-driven AI agents.

From Efficiency to Adaptivity: A Deeper Look at Adaptive Reasoning in Large Language Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Chao Wu · Baoheng Li · Mingchen Gao · Yu Tian · Zhenyi Wang
1 April 2026 at 12:00
arXiv:2511.10788v3 Announce Type: replace Abstract: Recent advances in large language models (LLMs) have made reasoning a central benchmark for evaluating intelligence. While prior surveys focus on efficiency by examining how to shorten reasoning chains or reduce computation, this view overlooks a fundamental challenge: current LLMs apply uniform reasoning strategies regardless of task complexity, generating long traces for trivial problems while failing to extend reasoning for difficult tasks. This survey reframes reasoning through the lens of {adaptivity}: the capability to allocate reasoning effort based on input characteristics such as difficulty and uncertainty. We make three contributions. First, we formalize deductive, inductive, and abductive reasoning within the LLM context, connecting these classical cognitive paradigms with their algorithmic realizations. Second, we formalize adaptive reasoning as a control-augmented policy optimization problem balancing task performance with computational cost, distinguishing learned policies from inference-time control mechanisms. Third, we propose a systematic taxonomy organizing existing methods into training-based approaches that internalize adaptivity through reinforcement learning, supervised fine-tuning, and learned controllers, and training-free approaches that achieve adaptivity through prompt conditioning, feedback-driven halting, and modular composition. This framework clarifies how different mechanisms realize adaptive reasoning in practice and enables systematic comparison across diverse strategies. We conclude by identifying open challenges in self-evaluation, meta-reasoning, and human-aligned reasoning control.

QuestA: Expanding Reasoning Capacity in LLMs via Question Augmentation

✇cs.AI, q-bio.NC updates on arXiv.org
By: Jiazheng Li · Hongzhou Lin · Hong Lu · Kaiyue Wen · Zaiwen Yang · Jiaxuan Gao · Yi Wu · Jingzhao Zhang
1 April 2026 at 12:00
arXiv:2507.13266v4 Announce Type: replace-cross Abstract: Reinforcement learning (RL) has emerged as a central paradigm for training large language models (LLMs) in reasoning tasks. Yet recent studies question RL's ability to incentivize reasoning capacity beyond the base model. This raises a key challenge: how can RL be adapted to solve harder reasoning problems more effectively? To address this challenge, we propose a simple yet effective strategy via Question Augmentation: introduce partial solutions during training to reduce problem difficulty and provide more informative learning signals. Our method, QuestA, when applied during RL training on math reasoning tasks, not only improves pass@1 but also pass@k-particularly on problems where standard RL struggles to make progress. This enables continual improvement over strong open-source models such as DeepScaleR and OpenMath Nemotron, further enhancing their reasoning capabilities. We achieve new state-of-the-art results on math benchmarks using 1.5B-parameter models: 72.50% (+10.73%) on AIME24, 62.29% (+12.79%) on AIME25, and 41.67% (+10.11%) on HMMT25. Code, data and model are available at https://github.com/foreverlasting1202/QuestA.

InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input Vision-Language Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Hongyuan Tao · Bencheng Liao · Shaoyu Chen · Haoran Yin · Qian Zhang · Wenyu Liu · Xinggang Wang
1 April 2026 at 12:00
arXiv:2512.08829v2 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) are increasingly tasked with ultra-long multimodal understanding. While linear architectures offer constant computation and memory footprints, they often struggle with high-frequency visual perception compared to standard Transformers. To bridge this gap, we introduce \textbf{InfiniteVL}. We first develop a hybrid base model called \textbf{InfiniteVL-Base} that interleaves a small fraction of Full Attention layers with Gated DeltaNet. Empowered by a tailored distillation and fine-tuning strategy, InfiniteVL-Base matches the fundamental multimodal performance of equivalent Transformers while achieving a \textbf{1.7$\times$} decoding speedup. However, the quadratic complexity of the retained Full Attention inevitably becomes an efficiency bottleneck when scaling to ultra long context. To break this barrier, we propose a novel Long-Sequence Architectural Fine-Tuning strategy that seamlessly transforms the dense attention into vision-specific sparse mechanisms. This yields two specialized variants: \textbf{InfiniteVL-Offline} for offline retrieval and \textbf{InfiniteVL-Online} for online streaming. By eliminating the computation explosion of global attention without sacrificing high-frequency visual recall, InfiniteVL-Offline achieves Transformer-level length generalization with a \textbf{5x} prefill acceleration at 256K context. Concurrently, InfiniteVL-Online delivers robust streaming perception with a constant memory footprint and a real-time throughput of \textbf{25} FPS. Code and models are available at https://github.com/hustvl/InfiniteVL.

Proposed Role of Circadian Clock Genes in Pathogenesis of HCC: Molecular Subtyping and Characterization

✇Omics in Hepatocellular
By: Zhikui Lu · Yi Zhou · Jian Luo · Zhicheng Liu · Zhenyu Xiao
28 March 2026 at 18:00

Biomedicines. 2026 Mar 12;14(3):645. doi: 10.3390/biomedicines14030645.

ABSTRACT

Background: Hepatocellular carcinoma (HCC) stands as a prevalent global health issue with increasing incidence and mortality rates. Hepatocellular carcinoma (HCC) exhibits profound molecular and clinical heterogeneity, which limits the effectiveness of current therapeutic strategies. Circadian rhythm disruption has been implicated in metabolic reprogramming, proliferation, and immune modulation in cancer, but its role in shaping HCC heterogeneity remains poorly defined. Methods: Four public HCC transcriptomic cohorts (TCGA-LIHC, CHCC, LIRI, LICA) were integrated using RMA normalization and ComBat for batch correction. Consensus clustering based on 31 core circadian clock genes (CCGs) identified robust molecular subtypes. Multi-omics characterization-including genomic alterations, pathway activity (GSEA/GSVA), immune microenvironment profiling (CIBERSORT, EPIC, MCP-counter, xCell), and drug-sensitivity prediction (pRRophetic/oncoPredict)-was performed to delineate subtype-specific biological properties. A nine-gene CCG-based RiskScore model was constructed using LASSO Cox regression to internally validate subtype robustness and intra-subtype risk stratification. Results: Using consensus clustering of 31 core CCGs in TCGA-LIHC and three independent validation cohorts (CHCC, LIRI, LICA), we identified three reproducible subtypes-Cluster-1 (metabolic-quiescent), Cluster-2 (transition-intermediate), and Cluster-3 (proliferation-inflammatory)-which were recapitulated across cohorts and showed distinct overall survival (Cluster-3 worst; log-rank p values significant across datasets). Multi-omic characterization revealed that Cluster-3 exhibits the highest tumor mutational burden and CNV burden with enrichment of TP53/AXIN1/TERT alterations, strong activation of cell-cycle, E2F, and G2M programs, and an immune-hot yet immunosuppressed microenvironment enriched for TAMs, Tregs and MDSCs. By contrast, Cluster-1 shows relative genomic stability, dominant hepatic metabolic signatures (fatty-acid oxidation, bile-acid and xenobiotic metabolism) and an immune-cold phenotype. Single-cell mapping linked ALAS1 expression to malignant hepatocytes predominating in Cluster-1, whereas NONO and CSNK1D localized to stromal (CAFs/TECs) and both malignant/immune compartments respectively in Cluster-3, providing a cellular mechanism for subtype-specific metabolism, angiogenesis and immune modulation. Finally, a nine-gene CCG-based RiskScore validated prognostic stratification and drug-sensitivity predictions indicated subtype-specific therapeutic vulnerabilities (notably increased predicted TKI sensitivity in Cluster-3). Conclusion: In conclusion, this study proposes a robust circadian rhythm-based molecular classification of hepatocellular carcinoma, revealing three biologically and clinically distinct subtypes characterized by divergent genomic alterations, metabolic programs, immune microenvironment states, and prognostic patterns. By integrating bulk and single-cell transcriptomic data, we identify subtype-specific roles of key circadian regulators-including ALAS1, NONO, and CSNK1D-in shaping tumor metabolism, proliferation, stromal remodeling, and immune suppression. These findings highlight circadian dysregulation as a potential upstream factor associated with HCC heterogeneity and provide a conceptual framework for developing subtype-tailored mechanistic studies and circadian-informed therapeutic strategies.

PMID:41898292 | PMC:PMC13024568 | DOI:10.3390/biomedicines14030645

Received — 26 March 2026 ⏭

Not All Tokens Are Created Equal: Query-Efficient Jailbreak Fuzzing for LLMs

✇cs.AI, q-bio.NC updates on arXiv.org
By: Wenyu Chen · Xiangtao Meng · Chuanchao Zang · Li Wang · Xinyu Gao · Jianing Wang · Peng Zhan · Zheng Li · Shanqing Guo
25 March 2026 at 12:00
arXiv:2603.23269v1 Announce Type: cross Abstract: Large Language Models(LLMs) are widely deployed, yet are vulnerable to jailbreak prompts that elicit policy-violating outputs. Although prior studies have uncovered these risks, they typically treat all tokens as equally important during prompt mutation, overlooking the varying contributions of individual tokens to triggering model refusals. Consequently, these attacks introduce substantial redundant searching under query-constrained scenarios, reducing attack efficiency and hindering comprehensive vulnerability assessment. In this work, we conduct a token-level analysis of refusal behavior and observe that token contributions are highly skewed rather than uniform. Moreover, we find strong cross-model consistency in refusal tendencies, enabling the use of a surrogate model to estimate token-level contributions to the target model's refusals. Motivated by these findings, we propose TriageFuzz, a token-aware jailbreak fuzzing framework that adapts the fuzz testing approach with a series of customized designs. TriageFuzz leverages a surrogate model to estimate the contribution of individual tokens to refusal behaviors, enabling the identification of sensitive regions within the prompt. Furthermore, it incorporates a refusal-guided evolutionary strategy that adaptively weights candidate prompts with a lightweight scorer to steer the evolution toward bypassing safety constraints. Extensive experiments on six open-source LLMs and three commercial APIs demonstrate that TriageFuzz achieves comparable attack success rates (ASR) with significantly reduced query costs. Notably, it attains a 90% ASR with over 70% fewer queries compared to baselines. Even under an extremely restrictive budget of 25 queries, TriageFuzz outperforms existing methods, improving ASR by 20-40%.

SortedRL: Accelerating RL Training for LLMs through Online Length-Aware Scheduling

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yiqi Zhang · Huiqiang Jiang · Xufang Luo · Zhihe Yang · Chengruidong Zhang · Yifei Shen · Dongsheng Li · Yuqing Yang · Lili Qiu · Yang You
25 March 2026 at 12:00
arXiv:2603.23414v1 Announce Type: cross Abstract: Scaling reinforcement learning (RL) has shown strong promise for enhancing the reasoning abilities of large language models (LLMs), particularly in tasks requiring long chain-of-thought generation. However, RL training efficiency is often bottlenecked by the rollout phase, which can account for up to 70% of total training time when generating long trajectories (e.g., 16k tokens), due to slow autoregressive generation and synchronization overhead between rollout and policy updates. We propose SortedRL, an online length-aware scheduling strategy designed to address this bottleneck by improving rollout efficiency and maintaining training stability. SortedRL reorders rollout samples based on output lengths, prioritizing short samples forming groups for early updates. This enables large rollout batches, flexible update batches, and near on-policy micro-curriculum construction simultaneously. To further accelerate the pipeline, SortedRL incorporates a mechanism to control the degree of off-policy training through a cache-based mechanism, and is supported by a dedicated RL infrastructure that manages rollout and update via a stateful controller and rollout buffer. Experiments using LLaMA-3.1-8B and Qwen-2.5-32B on diverse tasks, including logical puzzles, and math challenges like AIME 24, Math 500, and Minerval, show that SortedRL reduces RL training bubble ratios by over 50%, while attaining 3.9% to 18.4% superior performance over baseline given same amount of data.

Cerebra: A Multidisciplinary AI Board for Multimodal Dementia Characterization and Risk Assessment

✇cs.AI, q-bio.NC updates on arXiv.org
By: Sheng Liu · Long Chen · Zeyun Zhao · Qinglin Gou · Qingyue Wei · Arjun Masurkar · Kevin M. Spiegler · Philip Kuball · Stefania C. Bray · Megan Bernath · Deanna R. Willis · Jiang Bian · Lei Xing · Eric Topol · Kyunghyun Cho · Yu Huang · Ruogu Fang · Narges Razavian · James Zou
25 March 2026 at 12:00
arXiv:2603.21597v2 Announce Type: replace Abstract: Modern clinical practice increasingly depends on reasoning over heterogeneous, evolving, and incomplete patient data. Although recent advances in multimodal foundation models have improved performance on various clinical tasks, most existing models remain static, opaque, and poorly aligned with real-world clinical workflows. We present Cerebra, an interactive multi-agent AI team that coordinates specialized agents for EHR, clinical notes, and medical imaging analysis. These outputs are synthesized into a clinician-facing dashboard that combines visual analytics with a conversational interface, enabling clinicians to interrogate predictions and contextualize risk at the point of care. Cerebra supports privacy-preserving deployment by operating on structured representations and remains robust when modalities are incomplete. We evaluated Cerebra using a massive multi-institutional dataset spanning 3 million patients from four independent healthcare systems. Cerebra consistently outperformed both state-of-the-art single-modality models and large multimodal language model baselines. In dementia risk prediction, it achieved AUROCs up to 0.80, compared with 0.74 for the strongest single-modality model and 0.68 for language model baselines. For dementia diagnosis, it achieved an AUROC of 0.86, and for survival prediction, a C-index of 0.81. In a reader study with experienced physicians, Cerebra significantly improved expert performance, increasing accuracy by 17.5 percentage points in prospective dementia risk estimation. These results demonstrate Cerebra's potential for interpretable, robust decision support in clinical care.
❌