❌

Normal view

Received — 8 April 2026 ⏭

ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zhuowen Yuan · Zhaorun Chen · Zhen Xiang · Nathaniel D. Bastian · Seyyed Hadi Hashemi · Chaowei Xiao · Wenbo Guo · Bo Li
7 April 2026 at 12:00
arXiv:2604.04426v1 Announce Type: new Abstract: Existing research on LLM agent security mainly focuses on prompt injection and unsafe input/output behaviors. However, as agents increasingly rely on third-party tools and MCP servers, a new class of supply-chain threats has emerged, where malicious behaviors are embedded in seemingly benign tools, silently hijacking agent execution, leaking sensitive data, or triggering unauthorized actions. Despite their growing impact, there is currently no comprehensive benchmark for evaluating such threats. To bridge this gap, we introduce SC-Inject-Bench, a large-scale benchmark comprising over 10,000 malicious MCP tools grounded in a taxonomy of 25+ attack types derived from MITRE ATT&CK targeting supply-chain threats. We observe that existing MCP scanners and semantic guardrails perform poorly on this benchmark. Motivated by this finding, we propose ShieldNet, a network-level guardrail framework that detects supply-chain poisoning by observing real network interactions rather than surface-level tool traces. ShieldNet integrates a man-in-the-middle (MITM) proxy and an event extractor to identify critical network behaviors, which are then processed by a lightweight classifier for attack detection. Extensive experiments show that ShieldNet achieves strong detection performance (up to 0.995 F-1 with only 0.8% false positives) while introducing little runtime overhead, substantially outperforming existing MCP scanners and LLM-based guardrails.

FileGram: Grounding Agent Personalization in File-System Behavioral Traces

✇cs.AI, q-bio.NC updates on arXiv.org
By: Shuai Liu · Shulin Tian · Kairui Hu · Yuhao Dong · Zhe Yang · Bo Li · Jingkang Yang · Chen Change Loy · Ziwei Liu
7 April 2026 at 12:00
arXiv:2604.04901v1 Announce Type: cross Abstract: Coworking AI agents operating within local file systems are rapidly emerging as a paradigm in human-AI interaction; however, effective personalization remains limited by severe data constraints, as strict privacy barriers and the difficulty of jointly collecting multimodal real-world traces prevent scalable training and evaluation, and existing methods remain interaction-centric while overlooking dense behavioral traces in file-system operations; to address this gap, we propose FileGram, a comprehensive framework that grounds agent memory and personalization in file-system behavioral traces, comprising three core components: (1) FileGramEngine, a scalable persona-driven data engine that simulates realistic workflows and generates fine-grained multimodal action sequences at scale; (2) FileGramBench, a diagnostic benchmark grounded in file-system behavioral traces for evaluating memory systems on profile reconstruction, trace disentanglement, persona drift detection, and multimodal grounding; and (3) FileGramOS, a bottom-up memory architecture that builds user profiles directly from atomic actions and content deltas rather than dialogue summaries, encoding these traces into procedural, semantic, and episodic channels with query-time abstraction; extensive experiments show that FileGramBench remains challenging for state-of-the-art memory systems and that FileGramEngine and FileGramOS are effective, and by open-sourcing the framework, we hope to support future research on personalized memory-centric file-system agents.

SoSBench: Benchmarking Safety Alignment on Six Scientific Domains

✇cs.AI, q-bio.NC updates on arXiv.org
By: Fengqing Jiang · Fengbo Ma · Zhangchen Xu · Yuetai Li · Zixin Rao · Bhaskar Ramasubramanian · Luyao Niu · Bo Li · Xianyan Chen · Zhen Xiang · Radha Poovendran
7 April 2026 at 12:00
arXiv:2505.21605v3 Announce Type: replace-cross Abstract: Large language models (LLMs) exhibit advancing capabilities in complex tasks, such as reasoning and graduate-level question answering, yet their resilience against misuse, particularly involving scientifically sophisticated risks, remains underexplored. Existing safety benchmarks typically focus either on instructions requiring minimal knowledge comprehension (e.g., ``tell me how to build a bomb") or utilize prompts that are relatively low-risk (e.g., multiple-choice or classification tasks about hazardous content). Consequently, they fail to adequately assess model safety when handling knowledge-intensive, hazardous scenarios. To address this critical gap, we introduce SoSBench, a regulation-grounded, hazard-focused benchmark encompassing six high-risk scientific domains: chemistry, biology, medicine, pharmacology, physics, and psychology. The benchmark comprises 3,000 prompts derived from real-world regulations and laws, systematically expanded via an LLM-assisted evolutionary pipeline that introduces diverse, realistic misuse scenarios (e.g., detailed explosive synthesis instructions involving advanced chemical formulas). We evaluate frontier models within a unified evaluation framework using our SoSBench. Despite their alignment claims, advanced models consistently disclose policy-violating content across all domains, demonstrating alarmingly high rates of harmful responses (e.g., 84.9% for Deepseek-R1 and 50.3% for GPT-4.1). These results highlight significant safety alignment deficiencies and underscore urgent concerns regarding the responsible deployment of powerful LLMs.

InfoTok: Information-Theoretic Regularization for Capacity-Constrained Shared Visual Tokenization in Unified MLLMs

✇cs.AI, q-bio.NC updates on arXiv.org
By: Lv Tang · Tianyi Zheng · Bo Li · Xingyu Li
7 April 2026 at 12:00
arXiv:2602.01554v2 Announce Type: replace-cross Abstract: Unified multimodal large language models (MLLMs) aim to unify image understanding and image generation within a single framework, where a shared visual tokenizer serves as the sole interface that maps high-dimensional images into a limited token budget for downstream multimodal reasoning and synthesis. However, existing shared-token designs are largely architecture-driven and lack an explicit criterion for what information should be preserved to simultaneously support semantic abstraction and visual detail. In this paper, we adopt a capacity-constrained perspective, viewing the shared tokenizer as a compute-bounded learner whose finite representational budget should prioritize reusable structure over hard-to-exploit high-entropy variations and redundancy. Motivated by this view, we propose \textbf{\textit{InfoTok}}, an information-regularized tokenization mechanism grounded in the Information Bottleneck (IB) principle. InfoTok explicitly controls information flow from images to shared tokens to multimodal outputs by imposing mutual-information (MI) constraints that enforce a principled trade-off between compression and task relevance, while also encouraging cross-modal consistency. Because MI is intractable for high-dimensional visual representations, we instantiate InfoTok with practical, differentiable dependence estimators, including a variational IB formulation and a Hilbert Schmidt Independence Criterion (HSIC) based alternative. Integrated into three representative unified MLLMs without introducing any additional training data, InfoTok consistently improves both image understanding and generation performance. These results support information-regularized visual tokenization as a sound basis for token learning in unified MLLMs.
Received — 4 April 2026 ⏭

DR-LoRA: Dynamic Rank LoRA for Fine-Tuning Mixture-of-Experts Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Guanzhi Deng · Bo Li · Ronghao Chen · Xiujin Liu · Zhuo Han · Huacan Wang · Lijie Wen · Linqi Song
3 April 2026 at 12:00
arXiv:2601.04823v5 Announce Type: replace Abstract: Mixture-of-Experts (MoE) has become a prominent paradigm for scaling Large Language Models (LLMs). Parameter-efficient fine-tuning methods, such as LoRA, are widely adopted to adapt pretrained MoE LLMs to downstream tasks. However, existing approaches typically assign identical LoRA ranks to all expert modules, ignoring the heterogeneous specialization of pretrained experts. This uniform allocation leads to a resource mismatch: task-relevant experts are under-provisioned, while less relevant ones receive redundant parameters. To address this, we propose DR-LoRA, a Dynamic Rank LoRA framework for fine-tuning pretrained MoE models. Specifically, DR-LoRA initializes all expert LoRA modules with a small active rank and uses an expert saliency score, which combines routing frequency and gradient-based rank importance, to identify which experts would benefit most from additional capacity. It then periodically expands the active ranks of the task-critical expert LoRA, progressively constructing a heterogeneous rank distribution tailored to the target task. Experiments on three MoE models across six tasks show that DR-LoRA consistently outperforms LoRA and other strong baselines, demonstrating that task-adaptive heterogeneous rank allocation is an effective strategy to improve active capacity utilization in MoE fine-tuning.

Learning to Learn-at-Test-Time: Language Agents with Learnable Adaptation Policies

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zhanzhi Lou · Hui Chen · Yibo Li · Qian Wang · Bryan Hooi
3 April 2026 at 12:00
arXiv:2604.00830v2 Announce Type: replace-cross Abstract: Test-Time Learning (TTL) enables language agents to iteratively refine their performance through repeated interactions with the environment at inference time. At the core of TTL is an adaptation policy that updates the actor policy based on experience from previous episodes, thereby improving future behavior. Existing methods rely on fixed, hand-crafted adaptation policies rather than optimizing them for downstream improvement. We argue that optimal adaptation policies should be learned from task environments, not hand-engineered based on human intuition. To achieve this, we introduce Meta-TTL, a framework that formulates the discovery of effective adaptation policies as a bi-level optimization problem. Within this framework, the inner loop executes the standard TTL process, measuring how effectively a candidate adaptation policy helps an agent correct errors across sequential episodes. Guided by the agent's performance, the outer loop employs evolutionary search over a diverse distribution of training tasks to iteratively refine the adaptation policy. We evaluate Meta-TTL on Jericho and WebArena-Lite across both in-distribution (ID) and out-of-distribution (OOD) settings, using multiple meta-agent backbones. Results on both benchmarks show that Meta-TTL consistently outperforms hand-crafted baselines, suggesting that the optimized adaptation policy encodes transferable strategies that generalize beyond the training task distribution.
Received — 2 April 2026 ⏭

Xuanwu: Evolving General Multimodal Models into an Industrial-Grade Foundation for Content Ecosystems

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zhiqian Zhang · Xu Zhao · Xiaoqing Xu · Guangdong Liang · Weijia Wang · Xiaolei Lv · Bo Li · Jun Gao
1 April 2026 at 12:00
arXiv:2603.29211v1 Announce Type: new Abstract: In recent years, multimodal large models have continued to improve on general benchmarks. However, in real-world content moderation and adversarial settings, mainstream models still suffer from degraded generalization and catastrophic forgetting because of limited fine-grained visual perception and insufficient modeling of long-tail noise. In this paper, we present Xuanwu VL-2B as a case study of how general multimodal models can be developed into an industrial-grade foundation model for content ecosystems. The model adopts a compact InternViT-300M + MLP + Qwen3 1.7B architecture, balancing fine-grained visual perception, language-semantic alignment, and deployment cost within an approximately 2B-parameter budget. To balance business specialization with the retention of general capabilities, we developed a data iteration and curation mechanism and trained the model through a progressive three-stage pipeline: pre-training, mid-training, and post-training. Ablation studies and offline business evaluations show that Xuanwu VL-2B achieves an average score of 67.90 across seven OpenCompass multimodal metrics (vs. 64.27 for InternVL 3.5 2B), an average recall of 94.38% over seven independent business moderation tasks, and a weighted overall recall of 82.82% on policy-violating text in challenging adversarial OCR scenarios, outperforming Gemini-2.5-Pro (76.72%). These results show that, under a limited parameter budget, Xuanwu VL-2B achieves a practical balance among business alignment, visual perception, general capability retention, and deployment cost.

The Model Says Walk: How Surface Heuristics Override Implicit Constraints in LLM Reasoning

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yubo Li · Lu Zhang · Tianchong Jiang · Ramayya Krishnan · Rema Padman
1 April 2026 at 12:00
arXiv:2603.29025v1 Announce Type: cross Abstract: Large language models systematically fail when a salient surface cue conflicts with an unstated feasibility constraint. We study this through a diagnose-measure-bridge-treat framework. Causal-behavioral analysis of the ``car wash problem'' across six models reveals approximately context-independent sigmoid heuristics: the distance cue exerts 8.7 to 38 times more influence than the goal, and token-level attribution shows patterns more consistent with keyword associations than compositional inference. The Heuristic Override Benchmark (HOB) -- 500 instances spanning 4 heuristic by 5 constraint families with minimal pairs and explicitness gradients -- demonstrates generality across 14 models: under strict evaluation (10/10 correct), no model exceeds 75%, and presence constraints are hardest (44%). A minimal hint (e.g., emphasizing the key object) recovers +15 pp on average, suggesting the failure lies in constraint inference rather than missing knowledge; 12/14 models perform worse when the constraint is removed (up to -39 pp), revealing conservative bias. Parametric probes confirm that the sigmoid pattern generalizes to cost, efficiency, and semantic-similarity heuristics; goal-decomposition prompting recovers +6 to 9 pp by forcing models to enumerate preconditions before answering. Together, these results characterize heuristic override as a systematic reasoning vulnerability and provide a benchmark for measuring progress toward resolving it.

Electric dipole moment drives the dynamics of the TNFR1 complex I signalosome

✇Nature - Issue - nature.com science feeds
By: Jianping Liu · Jing Zhao · Jiayang Gao · Kun Zhao · Yaoyao Han · Jing Yang · Zefei Li · Jianyu Ye · Ziyu Sun · Fengyi Wang · Xinyi Liu · Zekai Li · Siyu Ji · Bo Liu · Cong Liu · Yixiao Zhang · Junying Yuan · James J. Chou
1 April 2026 at 08:00

Nature, Published online: 01 April 2026; doi:10.1038/s41586-026-10304-1

Long-range interactions mediated by protein electric dipole moments have a role in driving the assembly and disassembly of super-signalling complex I for promoting NF-κB signalling.

Integrative Multi-omics Analysis of Buti Huatan Tang in Chronic Obstructive Pulmonary Disease

✇Omics In Lung
By: Xin Zha · Wen-Wei Gong · Xue-Ying Li · Qing Xia · Jing-Hua Ruan · Rong Tan · Zhu-Sheng Zhu · Shao-Bo Liu
30 March 2026 at 18:00

J Vis Exp. 2026 Mar 13;(229). doi: 10.3791/70383.

ABSTRACT

This study utilized a multi-omics and computational biology framework to investigate the therapeutic potential of the Traditional Chinese Medicine (TCM) formula Buti Huatan Tang (BTHTT) against chronic obstructive pulmonary disease (COPD). Significant physiological improvements were observed in a rat model following BTHTT intervention. Histological analysis showed a reversal of lung pathological damage, while biochemical assays, and transcriptomics confirmed the normalization of IL-1β and IL-1R2 levels. Additionally, metabolic profiling revealed that BTHTT corrected disruptions in T3 and T4 thyroid hormone levels. A negative correlation was observed between the IL-1β/IL-1R2 axis and these thyroid hormones, indicating that their regulation is associated with the formula's therapeutic effect. Beyond direct measurements, machine learning algorithms identified ten COPD signature genes from clinical databases. Pathway enrichment analysis suggests that BTHTT may act through cytokine-cytokine-receptor interactions and thyroid hormone synthesis pathways. Furthermore, while 283 components were identified in vivo, compounds such as tanshinone IIA and cryptotanshinone are currently considered candidate active substances. Their role as primary drivers is supported by a model in which they stably bind to IL-1R2; this inference is based on molecular docking and molecular dynamics (MD) simulations rather than direct experimental isolation. Overall, the data support a model in which BTHTT exerts a multi-target effect on COPD by modulating inflammation and metabolic homeostasis. This integrated approach provides a refined scientific basis for the clinical application of BTHTT and highlights specific pathways for future experimental validation.

PMID:41911070 | DOI:10.3791/70383

Integrative Multi-omics Analysis of Buti Huatan Tang in Chronic Obstructive Pulmonary Disease

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Xin Zha · Wen-Wei Gong · Xue-Ying Li · Qing Xia · Jing-Hua Ruan · Rong Tan · Zhu-Sheng Zhu · Shao-Bo Liu
30 March 2026 at 18:00

J Vis Exp. 2026 Mar 13;(229). doi: 10.3791/70383.

ABSTRACT

This study utilized a multi-omics and computational biology framework to investigate the therapeutic potential of the Traditional Chinese Medicine (TCM) formula Buti Huatan Tang (BTHTT) against chronic obstructive pulmonary disease (COPD). Significant physiological improvements were observed in a rat model following BTHTT intervention. Histological analysis showed a reversal of lung pathological damage, while biochemical assays, and transcriptomics confirmed the normalization of IL-1β and IL-1R2 levels. Additionally, metabolic profiling revealed that BTHTT corrected disruptions in T3 and T4 thyroid hormone levels. A negative correlation was observed between the IL-1β/IL-1R2 axis and these thyroid hormones, indicating that their regulation is associated with the formula's therapeutic effect. Beyond direct measurements, machine learning algorithms identified ten COPD signature genes from clinical databases. Pathway enrichment analysis suggests that BTHTT may act through cytokine-cytokine-receptor interactions and thyroid hormone synthesis pathways. Furthermore, while 283 components were identified in vivo, compounds such as tanshinone IIA and cryptotanshinone are currently considered candidate active substances. Their role as primary drivers is supported by a model in which they stably bind to IL-1R2; this inference is based on molecular docking and molecular dynamics (MD) simulations rather than direct experimental isolation. Overall, the data support a model in which BTHTT exerts a multi-target effect on COPD by modulating inflammation and metabolic homeostasis. This integrated approach provides a refined scientific basis for the clinical application of BTHTT and highlights specific pathways for future experimental validation.

PMID:41911070 | DOI:10.3791/70383

A multicenter randomized clinical trial of portable transcranial alternating current stimulation for major depressive disorder

✇npj Digital Medicine
By: Yunhao Wu · Chuanbin Ning · Lei Shi · Jijun Wang · Kuanghao Ye · Ziyu Shen · Yuhan Wang · Duanwei Wang · Huihui Guo · Xian Qiu · Yixin Pan · Halimureti Paerhati · Debo Li · Yu Mo · Jing Wan · Huilin Hu · Yu Cao · Bomin Sun · Xin Lv
28 March 2026 at 08:00

npj Digital Medicine, Published online: 28 March 2026; doi:10.1038/s41746-026-02575-9

A multicenter randomized clinical trial of portable transcranial alternating current stimulation for major depressive disorder
Received — 26 March 2026 ⏭

DreamAudio: Customized Text-to-Audio Generation with Diffusion Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yi Yuan · Xubo Liu · Haohe Liu · Xiyuan Kang · Zhuo Chen · Yuxuan Wang · Mark D. Plumbley · Wenwu Wang
25 March 2026 at 12:00
arXiv:2509.06027v2 Announce Type: replace-cross Abstract: With the development of large-scale diffusion-based and language-modeling-based generative models, impressive progress has been achieved in text-to-audio generation. Despite producing high-quality outputs, existing text-to-audio models mainly aim to generate semantically aligned sound and fall short of controlling fine-grained acoustic characteristics of specific sounds. As a result, users who need specific sound content may find it difficult to generate the desired audio clips. In this paper, we present DreamAudio for customized text-to-audio generation (CTTA). Specifically, we introduce a new framework that is designed to enable the model to identify auditory information from user-provided reference concepts for audio generation. Given a few reference audio samples containing personalized audio events, our system can generate new audio samples that include these specific events. In addition, two types of datasets are developed for training and testing the proposed systems. The experiments show that DreamAudio generates audio samples that are highly consistent with the customized audio features and aligned well with the input text prompts. Furthermore, DreamAudio offers comparable performance in general text-to-audio tasks. We also provide a human-involved dataset containing audio events from real-world CTTA cases as the benchmark for customized generation tasks.
Received — 17 March 2026 ⏭

SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks

✇cs.AI, q-bio.NC updates on arXiv.org
By: Xiangyi Li · Wenbo Chen · Yimin Liu · Shenghan Zheng · Xiaokun Chen · Yifeng He · Yubo Li · Bingran You · Haotian Shen · Jiankai Sun · Shuyi Wang · Binxu Li · Qunhong Zeng · Di Wang · Xuandong Zhao · Yuanli Wang · Roey Ben Chaim · Zonglin Di · Yipeng Gao · Junwei He · Yizhuo He · Liqiang Jing · Luyang Kong · Xin Lan · Jiachen Li · Songlin Li · Yijiang Li · Yueqian Lin · Xinyi Liu · Xuanqing Liu · Haoran Lyu · Ze Ma · Bowei Wang · Runhui Wang · Tianyu Wang · Wengao Ye · Yue Zhang · Hanwen Xing · Yiqi Xue · Steven Dillmann · Han-chung Lee
16 March 2026 at 12:00
arXiv:2602.12670v3 Announce Type: replace Abstract: Agent Skills are structured packages of procedural knowledge that augment LLM agents at inference time. Despite rapid adoption, there is no standard way to measure whether they actually help. We present SkillsBench, a benchmark of 86 tasks across 11 domains paired with curated Skills and deterministic verifiers. Each task is evaluated under three conditions: no Skills, curated Skills, and self-generated Skills. We test 7 agent-model configurations over 7,308 trajectories. Curated Skills raise average pass rate by 16.2 percentage points(pp), but effects vary widely by domain (+4.5pp for Software Engineering to +51.9pp for Healthcare) and 16 of 84 tasks show negative deltas. Self-generated Skills provide no benefit on average, showing that models cannot reliably author the procedural knowledge they benefit from consuming. Focused Skills with 2--3 modules outperform comprehensive documentation, and smaller models with Skills can match larger models without them.
Received — 11 March 2026 ⏭

Deep Expert Injection for Anchoring Retinal VLMs with Domain-Specific Knowledge

✇cs.AI, q-bio.NC updates on arXiv.org
By: Shuai Lu · Meng Wang · Jia Guo · Jiawei Du · Bo Liu · Shengzhu Yang · Weihang Zhang · Huazhu Fu · Huiqi Li
10 March 2026 at 12:00
arXiv:2603.07131v1 Announce Type: cross Abstract: Large Vision Language Models (LVLMs) show immense potential for automated ophthalmic diagnosis. However, their clinical deployment is severely hindered by lacking domain-specific knowledge. In this work, we identify two structural deficiencies hindering reliable medical reasoning: 1) the Perception Gap, where general-purpose visual encoders fail to resolve fine-grained pathological cues (e.g., microaneurysms); and 2) the Reasoning Gap, where sparse visual evidence is progressively overridden by massive language priors in deeper transformer layers, leading to ungrounded hallucinations. To bridge these gaps, we propose EyExIn, a data-efficient framework designed to anchor retinal VLMs with expert knowledge via a Deep Expert Injection mechanism. Our architecture employs an Expert-Aware Dual-Stream encoding strategy that decouples visual representation into a general stream for anatomical context and a specialized expert stream for pathological semantics. To ensure high-fidelity integration, we design a Semantic-Adaptive Gated Fusion module, which dynamically amplifies subtle lesion signals while filtering irrelevant background noise. Furthermore, we introduce Adaptive Deep Expert Injection to embed persistent "Vision Anchors" by integrating fused visual features as residual biases directly into intermediate LLM layers. This mechanism creates a visual shortcut that forces the reasoning stack to remain strictly grounded in visual evidence. Extensive experiments across four benchmarks demonstrate that our model consistently outperforms massive proprietary systems. EyExIn significantly enhances domain-specific knowledge embedding and achieves state-of-the-art precision in ophthalmic visual question answering, advancing the development of trustworthy ophthalmic AI.

Think, Speak, Decide: Language-Augmented Multi-Agent Reinforcement Learning for Economic Decision-Making

✇cs.AI, q-bio.NC updates on arXiv.org
By: Heyang Ma · Qirui Mi · Qipeng Yang · Zijun Fan · Bo Li · Haifeng Zhang
10 March 2026 at 12:00
arXiv:2511.12876v3 Announce Type: replace Abstract: Economic decision-making depends not only on structured signals such as prices and taxes, but also on unstructured language, including peer dialogue and media narratives. While multi-agent reinforcement learning (MARL) has shown promise in optimizing economic decisions, it struggles with the semantic ambiguity and contextual richness of language. We propose LAMP (Language-Augmented Multi-Agent Policy), a framework that integrates language into economic decision-making and narrows the gap to real-world settings. LAMP follows a Think-Speak-Decide pipeline: (1) Think interprets numerical observations to extract short-term shocks and long-term trends, caching high-value reasoning trajectories; (2) Speak crafts and exchanges strategic messages based on reasoning, updating beliefs by parsing peer communications; and (3) Decide fuses numerical data, reasoning, and reflections into a MARL policy to optimize language-augmented decision-making. Experiments in economic simulation show that LAMP outperforms both MARL and LLM-only baselines in cumulative return (+63.5%, +34.0%), robustness (+18.8%, +59.4%), and interpretability. These results demonstrate the potential of language-augmented policies to deliver more effective and robust economic strategies.

SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks

✇cs.AI, q-bio.NC updates on arXiv.org
By: Xiangyi Li · Wenbo Chen · Yimin Liu · Shenghan Zheng · Xiaokun Chen · Yifeng He · Yubo Li · Bingran You · Haotian Shen · Jiankai Sun · Shuyi Wang · Binxu Li · Qunhong Zeng · Di Wang · Xuandong Zhao · Yuanli Wang · Roey Ben Chaim · Zonglin Di · Yipeng Gao · Junwei He · Yizhuo He · Liqiang Jing · Luyang Kong · Xin Lan · Jiachen Li · Songlin Li · Yijiang Li · Yueqian Lin · Xinyi Liu · Xuanqing Liu · Haoran Lyu · Ze Ma · Bowei Wang · Runhui Wang · Tianyu Wang · Wengao Ye · Yue Zhang · Hanwen Xing · Yiqi Xue · Steven Dillmann · Han-chung Lee
10 March 2026 at 12:00
arXiv:2602.12670v2 Announce Type: replace Abstract: Agent Skills are structured packages of procedural knowledge that augment LLM agents at inference time. Despite rapid adoption, there is no standard way to measure whether they actually help. We present SkillsBench, a benchmark of 86 tasks across 11 domains paired with curated Skills and deterministic verifiers. Each task is evaluated under three conditions: no Skills, curated Skills, and self-generated Skills. We test 7 agent-model configurations over 7,308 trajectories. Curated Skills raise average pass rate by 16.2 percentage points(pp), but effects vary widely by domain (+4.5pp for Software Engineering to +51.9pp for Healthcare) and 16 of 84 tasks show negative deltas. Self-generated Skills provide no benefit on average, showing that models cannot reliably author the procedural knowledge they benefit from consuming. Focused Skills with 2--3 modules outperform comprehensive documentation, and smaller models with Skills can match larger models without them.

Arginine methylation-dependent stabilization of SUV39H1 promotes breast cancer growth

✇Oncogene - Issue - nature.com science feeds
By: Wei Zhao · Ge Wang · Peng Wang · Bo Ma · Bo Liu · Yenan Fu · Youzhi Tang · Xinwei Duan · Kunhao Zhou · Jing Zhang · Wei-Guo Zhu · Hongquan Zhang · Yu Yu
7 March 2026 at 08:00

Oncogene, Published online: 07 March 2026; doi:10.1038/s41388-026-03712-0

Arginine methylation-dependent stabilization of SUV39H1 promotes breast cancer growth
Received — 5 March 2026 ⏭

Towards Realistic Personalization: Evaluating Long-Horizon Preference Following in Personalized User-LLM Interactions

✇cs.AI, q-bio.NC updates on arXiv.org
By: Qianyun Guo · Yibo Li · Yue Liu · Bryan Hooi
5 March 2026 at 13:00
arXiv:2603.04191v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly serving as personal assistants, where users share complex and diverse preferences over extended interactions. However, assessing how well LLMs can follow these preferences in realistic, long-term situations remains underexplored. This work proposes RealPref, a benchmark for evaluating realistic preference-following in personalized user-LLM interactions. RealPref features 100 user profiles, 1300 personalized preferences, four types of preference expression (ranging from explicit to implicit), and long-horizon interaction histories. It includes three types of test questions (multiple-choice, true-or-false, and open-ended), with detailed rubrics for LLM-as-a-judge evaluation. Results indicate that LLM performance significantly drops as context length grows and preference expression becomes more implicit, and that generalizing user preference understanding to unseen scenarios poses further challenges. RealPref and these findings provide a foundation for future research to develop user-aware LLM assistants that better adapt to individual needs. The code is available at https://github.com/GG14127/RealPref.

Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning

✇cs.AI, q-bio.NC updates on arXiv.org
By: Huihan Liu · Changyeon Kim · Bo Liu · Minghuan Liu · Yuke Zhu
5 March 2026 at 13:00
arXiv:2603.03818v1 Announce Type: cross Abstract: Continual learning is a long-standing challenge in robot policy learning, where a policy must acquire new skills over time without catastrophically forgetting previously learned ones. While prior work has extensively studied continual learning in relatively small behavior cloning (BC) policy models trained from scratch, its behavior in modern large-scale pretrained Vision-Language-Action (VLA) models remains underexplored. In this work, we found that pretrained VLAs are remarkably resistant to forgetting compared with smaller policy models trained from scratch. Simple Experience Replay (ER) works surprisingly well on VLAs, sometimes achieving zero forgetting even with a small replay data size. Our analysis reveals that pretraining plays a critical role in downstream continual learning performance: large pretrained models mitigate forgetting with a small replay buffer size while maintaining strong forward learning capabilities. Furthermore, we found that VLAs can retain relevant knowledge from prior tasks despite performance degradation during learning new tasks. This knowledge retention enables rapid recovery of seemingly forgotten skills through finetuning. Together, these insights imply that large-scale pretraining fundamentally changes the dynamics of continual learning, enabling models to continually acquire new skills over time with simple replay. Code and more information can be found at https://ut-austin-rpl.github.io/continual-vla
❌