❌

Normal view

Received — 17 March 2026 ⏭

When Right Meets Wrong: Bilateral Context Conditioning with Reward-Confidence Correction for GRPO

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yu Li · Tian Lan · Zhengling Qi
16 March 2026 at 12:00
arXiv:2603.13134v1 Announce Type: new Abstract: Group Relative Policy Optimization (GRPO) has emerged as an effective method for training reasoning models. While it computes advantages based on group mean, GRPO treats each output as an independent sample during the optimization and overlooks a vital structural signal: the natural contrast between correct and incorrect solutions within the same group, thus ignoring the rich, comparative data that could be leveraged by explicitly pitting successful reasoning traces against failed ones. To capitalize on this, we present a contrastive reformulation of GRPO, showing that the GRPO objective implicitly maximizes the margin between the policy ratios of correct and incorrect samples. Building on this insight, we propose Bilateral Context Conditioning (BICC), a mechanism that allows the model to cross-reference successful and failed reasoning traces during the optimization, enabling a direct information flow across samples. We further introduce Reward-Confidence Correction (RCC) to stabilize training by dynamically adjusts the advantage baseline in GRPO using reward-confidence covariance derived from the first-order approximation of the variance-minimizing estimator. Both mechanisms require no additional sampling or auxiliary models and can be adapted to all GRPO variants. Experiments on mathematical reasoning benchmarks demonstrate consistent improvements across comprehensive models and algorithms. Code is available at \href{https://github.com/Skylanding/BiCC}{https://github.com/Skylanding/BiCC}.

Empowering Semantic-Sensitive Underwater Image Enhancement with VLM

✇cs.AI, q-bio.NC updates on arXiv.org
By: Guodong Fan · Shengning Zhou · Genji Yuan · Huiyu Li · Jingchun Zhou · Jinjiang Li
16 March 2026 at 12:00
arXiv:2603.12773v1 Announce Type: cross Abstract: In recent years, learning-based underwater image enhancement (UIE) techniques have rapidly evolved. However, distribution shifts between high-quality enhanced outputs and natural images can hinder semantic cue extraction for downstream vision tasks, thereby limiting the adaptability of existing enhancement models. To address this challenge, this work proposes a new learning mechanism that leverages Vision-Language Models (VLMs) to empower UIE models with semantic-sensitive capabilities. To be concrete, our strategy first generates textual descriptions of key objects from a degraded image via VLMs. Subsequently, a text-image alignment model remaps these relevant descriptions back onto the image to produce a spatial semantic guidance map. This map then steers the UIE network through a dual-guidance mechanism, which combines cross-attention and an explicit alignment loss. This forces the network to focus its restorative power on semantic-sensitive regions during image reconstruction, rather than pursuing a globally uniform improvement, thereby ensuring the faithful restoration of key object features. Experiments confirm that when our strategy is applied to different UIE baselines, significantly boosts their performance on perceptual quality metrics as well as enhances their performance on detection and segmentation tasks, validating its effectiveness and adaptability.

Visual-ERM: Reward Modeling for Visual Equivalence

✇cs.AI, q-bio.NC updates on arXiv.org
By: Ziyu Liu · Shengyuan Ding · Xinyu Fang · Xuanlang Dai · Penghui Yang · Jianze Liang · Jiaqi Wang · Kai Chen · Dahua Lin · Yuhang Zang
16 March 2026 at 12:00
arXiv:2603.13224v1 Announce Type: cross Abstract: Vision-to-code tasks require models to reconstruct structured visual inputs, such as charts, tables, and SVGs, into executable or structured representations with high visual fidelity. While recent Large Vision Language Models (LVLMs) achieve strong results via supervised fine-tuning, reinforcement learning remains challenging due to misaligned reward signals. Existing rewards either rely on textual rules or coarse visual embedding similarity, both of which fail to capture fine-grained visual discrepancies and are vulnerable to reward hacking. We propose Visual Equivalence Reward Model (Visual-ERM), a multimodal generative reward model that provides fine-grained, interpretable, and task-agnostic feedback to evaluate vision-to-code quality directly in the rendered visual space. Integrated into RL, Visual-ERM improves Qwen3-VL-8B-Instruct by +8.4 on chart-to-code and yields consistent gains on table and SVG parsing (+2.7, +4.1 on average), and further strengthens test-time scaling via reflection and revision. We also introduce VisualCritic-RewardBench (VC-RewardBench), a benchmark for judging fine-grained image-to-image discrepancies on structured visual data, where Visual-ERM at 8B decisively outperforms Qwen3-VL-235B-Instruct and approaches leading closed-source models. Our results suggest that fine-grained visual reward supervision is both necessary and sufficient for vision-to-code RL, regardless of task specificity.

How to Build a Quantum Supercomputer: Scaling from Hundreds to Millions of Qubits

✇cs.AI, q-bio.NC updates on arXiv.org
By: Masoud Mohseni · Artur Scherer · K. Grace Johnson · Oded Wertheim · Matthew Otten · Namit Anand · Navid Anjum Aadit · Yuri Alexeev · Gilad Ben-Shach · Kirk M. Bresniker · Kerem Y. Camsari · Barbara Chapman · Soumitra Chatterjee · Shuvro Chowdhury · Gebremedhin A. Dagnew · Tom Dvir · Aniello Esposito · Farah Fahim · Michael Ferguson · Marco Fiorentino · Archit Gajjar · Katerina Gratsea · Gaurav Gyawali · Christian Heiter · Ali H. Z. Kavaki · Abdullah Khalid · Xiangzhou Kong · Bohdan Kulchytskyy · Elica Kyoseva · Ruoyu Li · P. Aaron Lott · Igor L. Markov · Robert F. McDermott · Lucas Morais · Giacomo Pedretti · Pooja Rao · Eleanor Rieffel · Allyson Silva · John Sorebo · Panagiotis Spentzouris · Ziv Steiner · Boyan Torosov · Davide Venturelli · Robert J. Visser · Zak Webb · Xin Zhan · Yonatan Cohen · Pooya Ronagh · Alan Ho · Raymond G. Beausoleil · John M. Martinis
16 March 2026 at 12:00
arXiv:2411.10406v3 Announce Type: replace-cross Abstract: In the span of four decades, quantum computation has evolved from an intellectual curiosity to a potentially realizable technology. Today, small-scale demonstrations have become possible for quantum algorithmic primitives on hundreds of physical qubits. Nevertheless, there are significant outstanding challenges in quantum hardware, fabrication, software architecture, and algorithms on the path towards a full-stack scalable quantum computing technology. Here, we provide a comprehensive review of these scaling challenges. We show how to facilitate scaling by adopting existing semiconductor technology to build much higher-quality qubits, employing systems engineering approaches, and performing distributed heterogeneous quantum-classical computing. We provide a detailed resource and sensitivity analysis for quantum applications on surface-code error-corrected quantum computers given current, target, and desired hardware specifications based on superconducting qubits, accounting for a realistic distribution of errors. We provide comprehensive resource estimates for several utility-scale applications including quantum chemistry calculations, catalyst design, NMR spectroscopy, and Fermi-Hubbard simulation. We show that orders of magnitude enhancement in performance could be obtained by a combination of hardware improvements and tight quantum-HPC integration. Furthermore, we introduce high-performance architectures for quantum-probabilistic computing with custom-designed accelerators to tackle today's industry-scale classical optimization, machine learning, and quantum simulation tasks in a cost-effective manner.
Received — 14 March 2026 ⏭

Integrated Network Toxicology and Metabolomics Elucidate Mechanisms of Carbosulfan-Induced Respiratory Toxicity in Rats

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Xian Ju · Di Liang · Hongyu Su · Yachun Zhang · Zhenyu Liang · Yiheng Liu · Wenqi Zhao · Dan Zhang · Zhe Chen · Keming Yun
14 March 2026 at 18:00

Int J Mol Sci. 2026 Feb 25;27(5):2170. doi: 10.3390/ijms27052170.

ABSTRACT

Carbosulfan is a widely used carbamate insecticide, yet its mechanisms of respiratory toxicity remain poorly understood. This study integrated network toxicology, untargeted metabolomics, and molecular docking to systematically investigate the potential mechanisms of carbosulfan-induced respiratory toxicity in male Sprague Dawley rats. Rats were administered a single oral dose of carbosulfan (125 or 250 mg/kg) and assessed after 12 h. Exposure resulted in significant pathological lung damage, characterized by disrupted alveolar architecture, inflammatory cell infiltration, and increased serum levels of the pro-inflammatory cytokines IL-6, IL-1β, and TNF-α. Network toxicology analysis identified 51 potential targets associated with respiratory toxicity, with core targets including SRC, EGFR, PTGS2, CXCL8, CYP3A4, and NR3C1. Enriched pathways were primarily related to neuroactive ligand-receptor interaction, VEGF signaling, and arachidonic acid metabolism. Untargeted metabolomics revealed significant metabolic perturbations in pathways central to antioxidant defense and energy homeostasis, including glutathione metabolism, the tricarboxylic acid cycle, and arginine biosynthesis. Molecular docking confirmed stable in silico binding affinities between carbosulfan and the predicted core targets. Integrative analysis suggests that carbosulfan exposure is associated with respiratory damage, potentially through interconnected mechanisms involving oxidative stress, inflammation, and disruption of cell signaling and metabolic enzyme systems. However, given the acute high-dose nature of the model and the interpretative integration of multi-omics data, these findings should be considered hypothesis-generating. This study provides a novel system-level perspective on carbosulfan-induced respiratory toxicity and highlights key pathways and targets for future validation in chronic exposure models.

PMID:41828400 | PMC:PMC12984169 | DOI:10.3390/ijms27052170

Integrated Network Toxicology and Metabolomics Elucidate Mechanisms of Carbosulfan-Induced Respiratory Toxicity in Rats

✇Omics In Lung
By: Xian Ju · Di Liang · Hongyu Su · Yachun Zhang · Zhenyu Liang · Yiheng Liu · Wenqi Zhao · Dan Zhang · Zhe Chen · Keming Yun
14 March 2026 at 18:00

Int J Mol Sci. 2026 Feb 25;27(5):2170. doi: 10.3390/ijms27052170.

ABSTRACT

Carbosulfan is a widely used carbamate insecticide, yet its mechanisms of respiratory toxicity remain poorly understood. This study integrated network toxicology, untargeted metabolomics, and molecular docking to systematically investigate the potential mechanisms of carbosulfan-induced respiratory toxicity in male Sprague Dawley rats. Rats were administered a single oral dose of carbosulfan (125 or 250 mg/kg) and assessed after 12 h. Exposure resulted in significant pathological lung damage, characterized by disrupted alveolar architecture, inflammatory cell infiltration, and increased serum levels of the pro-inflammatory cytokines IL-6, IL-1β, and TNF-α. Network toxicology analysis identified 51 potential targets associated with respiratory toxicity, with core targets including SRC, EGFR, PTGS2, CXCL8, CYP3A4, and NR3C1. Enriched pathways were primarily related to neuroactive ligand-receptor interaction, VEGF signaling, and arachidonic acid metabolism. Untargeted metabolomics revealed significant metabolic perturbations in pathways central to antioxidant defense and energy homeostasis, including glutathione metabolism, the tricarboxylic acid cycle, and arginine biosynthesis. Molecular docking confirmed stable in silico binding affinities between carbosulfan and the predicted core targets. Integrative analysis suggests that carbosulfan exposure is associated with respiratory damage, potentially through interconnected mechanisms involving oxidative stress, inflammation, and disruption of cell signaling and metabolic enzyme systems. However, given the acute high-dose nature of the model and the interpretative integration of multi-omics data, these findings should be considered hypothesis-generating. This study provides a novel system-level perspective on carbosulfan-induced respiratory toxicity and highlights key pathways and targets for future validation in chronic exposure models.

PMID:41828400 | PMC:PMC12984169 | DOI:10.3390/ijms27052170

FNDC1 Competitively Binds Gbeta2 to Suppress the beta-Catenin-Destruction Complex and Promote Gastric Cancer Malignancy

✇Omics in Gastric
By: Wenyu Gao · Hao Chen · Fangyu Lin · Jialin Liu · Tingxuan Huang · Han Wang · Xiaojiao Weng · Xinli Wang · Xiaoyan Lin · Tao Jiang
11 March 2026 at 18:00

FASEB J. 2026 Mar 31;40(6):e71634. doi: 10.1096/fj.202503587R.

ABSTRACT

Gastric cancer (GC) is a leading cause of cancer-related deaths and has high recurrence rate. Although fibronectin domain-containing protein 1 (FNDC1) is implicated in GC progression, its molecular mechanisms remain unclear. Multi-omics analyses (TCGA, GEO datasets) were used to assess FNDC1 expression and clinical correlation. In vitro (cell proliferation, invasion, EMT markers) and in vivo (xenograft) experiments, combined with molecular assays (Co-IP, WB, ChIP), explored FNDC1's function and mechanism. FNDC1 was significantly upregulated in GC, correlating with advanced clinicopathological features and poor prognosis. Knockdown of FNDC1 suppressed GC cell proliferation, invasion, and metastasis by inhibiting EMT and Wnt/β-catenin signaling. Mechanistically, FNDC1 competitively bound the WD5 domain (residues 224-254) of Gβ2, disrupting Gβγ-Dvl1 interaction. This prevented Dvl1 degradation, promoted Axin1 ubiquitination, and destabilized the β-catenin-destruction complex (GSK3 β-APC-Axin1), leading to β-catenin accumulation and Wnt pathway activation. FNDC1 drives GC malignancy by targeting the Gβ2-Dvl1 axis to activate Wnt/β-catenin signaling, suggesting FNDC1 as a novel prognostic biomarker and therapeutic target.

PMID:41808415 | PMC:PMC12976582 | DOI:10.1096/fj.202503587R

Received — 11 March 2026 ⏭

A prior information informed learning architecture for flying trajectory prediction

✇cs.AI, q-bio.NC updates on arXiv.org
By: Xianda Huang · Zidong Han · Ruibo Jin · Zhenyu Wang · Wenyu Li · Xiaoyang Li · Yi Gong
10 March 2026 at 12:00
arXiv:2603.06863v1 Announce Type: cross Abstract: Trajectory prediction for flying objects is critical in domains ranging from sports analytics to aerospace. However, traditional methods struggle with complex physical modeling, computational inefficiencies, and high hardware demands, often neglecting critical trajectory events like landing points. This paper introduces a novel, hardware-efficient trajectory prediction framework that integrates environmental priors with a Dual-Transformer-Cascaded (DTC) architecture. We demonstrate this approach by predicting the landing points of tennis balls in real-world outdoor courts. Using a single industrial camera and YOLO-based detection, we extract high-speed flight coordinates. These coordinates, fused with structural environmental priors (e.g., court boundaries), form a comprehensive dataset fed into our proposed DTC model. A first-level Transformer classifies the trajectory, while a second-level Transformer synthesizes these features to precisely predict the landing point. Extensive ablation and comparative experiments demonstrate that integrating environmental priors within the DTC architecture significantly outperforms existing trajectory prediction frameworks

GCGNet: Graph-Consistent Generative Network for Time Series Forecasting with Exogenous Variables

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zhengyu Li · Xiangfei Qiu · Yuhan Zhu · Xingjian Wu · Jilin Hu · Chenjuan Guo · Bin Yang
10 March 2026 at 12:00
arXiv:2603.08032v1 Announce Type: cross Abstract: Exogenous variables offer valuable supplementary information for predicting future endogenous variables. Forecasting with exogenous variables needs to consider both past-to-future dependencies (i.e., temporal correlations) and the influence of exogenous variables on endogenous variables (i.e., channel correlations). This is pivotal when future exogenous variables are available, because they may directly affect the future endogenous variables. Many methods have been proposed for time series forecasting with exogenous variables, focusing on modeling temporal and channel correlations. However, most of them use a two-step strategy, modeling temporal and channel correlations separately, which limits their ability to capture joint correlations across time and channels. Furthermore, in real-world scenarios, time series are frequently affected by various forms of noises, underscoring the critical importance of robustness in such correlations modeling. To address these limitations, we propose GCGNet, a Graph-Consistent Generative Network for time series forecasting with exogenous variables. Specifically, GCGNet first employs a Variational Generator to produce coarse predictions. A Graph Structure Aligner then further guides it by evaluating the consistency between the generated and true correlations, where the correlations are represented as graphs, and are robust to noises. Finally, a Graph Refiner is proposed to refine the predictions to prevent degeneration and improve accuracy. Extensive experiments on 12 real-world datasets demonstrate that GCGNet outperforms state-of-the-art baselines.

Towards Effective and Efficient Graph Alignment without Supervision

✇cs.AI, q-bio.NC updates on arXiv.org
By: Songyang Chen · Youfang Lin · Yu Liu · Shuai Zheng · Lei Zou
10 March 2026 at 12:00
arXiv:2603.08526v1 Announce Type: cross Abstract: Unsupervised graph alignment aims to find the node correspondence across different graphs without any anchor node pairs. Despite the recent efforts utilizing deep learning-based techniques, such as the embedding and optimal transport (OT)-based approaches, we observe their limitations in terms of model accuracy-efficiency tradeoff. By focusing on the exploitation of local and global graph information, we formalize them as the ``local representation, global alignment'' paradigm, and present a new ``global representation and alignment'' paradigm to resolve the mismatch between the two phases in the alignment process. We then propose \underline{Gl}obal representation and \underline{o}ptimal transport-\underline{b}ased \underline{Align}ment (\texttt{GlobAlign}), and its variant, \texttt{GlobAlign-E}, for better \underline{E}fficiency. Our methods are equipped with the global attention mechanism and a hierarchical cross-graph transport cost, able to capture long-range and implicit node dependencies beyond the local graph structure. Furthermore, \texttt{GlobAlign-E} successfully closes the time complexity gap between representative embedding and OT-based methods, reducing OT's cubic complexity to quadratic terms. Through extensive experiments, our methods demonstrate superior performance, with up to a 20\% accuracy improvement over the best competitor. Meanwhile, \texttt{GlobAlign-E} achieves the best efficiency, with an order of magnitude speedup against existing OT-based methods.

Parallelized Planning-Acting for Efficient LLM-based Multi-Agent Systems in Minecraft

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yaoru Li · Shunyu Liu · Tongya Zheng · Li Sun · Mingli Song
10 March 2026 at 12:00
arXiv:2503.03505v2 Announce Type: replace Abstract: Recent advancements in Large Language Model~(LLM)-based Multi-Agent Systems (MAS) have demonstrated remarkable potential for tackling complex decision-making tasks. However, existing frameworks inevitably rely on serialized execution paradigms, where agents must complete sequential LLM planning before taking action. This fundamental constraint severely limits real-time responsiveness and adaptation, which is crucial in dynamic environments with ever-changing scenarios like Minecraft. In this paper, we propose a novel parallelized planning-acting framework for LLM-based MAS, featuring a dual-thread architecture with interruptible execution to enable concurrent planning and acting. Specifically, our framework comprises two core threads: (1) a planning thread driven by a centralized memory system, maintaining synchronization of environmental states and agent communication to support dynamic decision-making; and (2) an acting thread equipped with a comprehensive skill library, enabling automated task execution through recursive decomposition. Extensive experiments on Minecraft demonstrate the effectiveness of the proposed framework.

Improving Visual Object Tracking through Visual Prompting

✇cs.AI, q-bio.NC updates on arXiv.org
By: Shih-Fang Chen · Jun-Cheng Chen · I-Hong Jhuo · Yen-Yu Lin
10 March 2026 at 12:00
arXiv:2409.18901v2 Announce Type: replace-cross Abstract: Learning a discriminative model that distinguishes the specified target from surrounding distractors across frames is essential for generic object tracking (GOT). Dynamic adaptation of target representation against distractors remains challenging because prevailing trackers exhibit limited discriminative capability. To address this issue, we present a new visual prompting mechanism for generic object tracking, termed PiVOT. PiVOT introduces mechanisms that leverage the pretrained foundation model (CLIP) to automatically generate and refine visual prompts online, thereby enabling the tracker to suppress distractors through contrastive guidance. To transfer contrastive knowledge from the foundation model to the tracker, PiVOT automatically propagates this knowledge online and dynamically generates and updates visual prompts. Specifically, it proposes a prompt initialization mechanism that produces an initial visual prompt highlighting potential target locations. The foundation model is then used to refine the prompt based on appearance similarities between candidate objects and reference templates across potential targets. After refinement, the visual prompt better highlights potential target locations and reduces irrelevant prompt information. With the proposed prompting mechanism, the tracker can generate instance-aware feature maps guided by the visual prompts, which are incrementally and automatically updated during tracking, thereby effectively suppressing distractors. Extensive experiments across multiple benchmarks indicate that PiVOT, with the proposed prompting mechanism, can suppress distracting objects and improve tracking performance.

A Simple "Motivation" Can Enhance Reinforcement Finetuning of Large Reasoning Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Junjie Zhang · Guozheng Ma · Shunyu Liu · Haoyu Wang · Jiaxing Huang · Ting-En Lin · Fei Huang · Yongbin Li · Dacheng Tao
10 March 2026 at 12:00
arXiv:2506.18485v3 Announce Type: replace-cross Abstract: Reinforcement Learning with Verifiable Rewards~(RLVR) has emerged as a powerful learn-to-reason paradigm for large reasoning models to tackle complex tasks. However, the current RLVR paradigm is still not efficient enough, as it works in a trial-and-error manner. To perform better, the model needs to explore the reward space by numerously generating responses and learn from fragmented reward signals, blind to the overall reward patterns. Fortunately, verifiable rewards make the natural language description of the reward function possible, and meanwhile, LLMs have demonstrated strong in-context learning ability. This motivates us to explore if large reasoning models can benefit from a \textbf{motivation} of the task, \textit{i.e.}, awareness of the reward function, during the reinforcement finetuning process, as we humans sometimes do when learning. In this paper, we introduce \textit{\textbf{M}otivation-\textbf{e}nhanced \textbf{R}einforcement \textbf{F}inetuning}~(\textbf{MeRF}), an intuitive yet effective method enhancing reinforcement finetuning of LLMs by involving \emph{``telling LLMs rules of the game''}. Specifically, \textbf{MeRF} directly injects the reward specification into the prompt, which serves as an in-context motivation for the model to be aware of the optimization objective. This simple modification leverages the in-context learning ability of LLMs, aligning generation with optimization, thereby incentivizing the model to generate desired outputs from both inner motivation and external reward. Empirical evaluations demonstrate that \textbf{MeRF} achieves substantial performance gains over the RLVR baseline. Moreover, ablation studies show that MeRF performs better with greater consistency between the in-context motivation and the external reward function, while the model also demonstrates an ability to adapt to misleading motivations through reinforcement finetuning.

Tiny but Mighty: A Software-Hardware Co-Design Approach for Efficient Multimodal Inference on Battery-Powered Small Devices

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yilong Li · Shuai Zhang · Yijing Zeng · Hao Zhang · Xinmiao Xiong · Jingyu Liu · Pan Hu · Suman Banerjee
10 March 2026 at 12:00
arXiv:2510.05109v5 Announce Type: replace-cross Abstract: Large Multimodal Models (LMMs) are inherently modular, consisting of vision and audio encoders, projectors, and large language models. Yet, they are almost always executed monolithically, which underutilizes the heterogeneous accelerators (NPUs, GPUs, DSPs) in modern SoCs and leads to high end-to-end latency. In this paper, we present NANOMIND, a hardware--software co-design inference framework for Large Multimodal Models (LMMs) that breaks large models into modular ``bricks'' (vision, language, audio, etc.) and maps each to its ideal accelerator. The key insight is that large models can be broken into modular components and scheduled to run on the most appropriate compute units. It performs module-level dynamic offloading across accelerators on unified-memory SoCs. By combining customized hardware design, system-level scheduling, and optimized low-bit computation kernels, we demonstrate our framework with a compact, battery-powered device capable of running LMMs entirely on device. This prototype functions as a self-contained intelligent assistant that requires no network connectivity, while achieving higher throughput and superior power efficiency under strict resource constraints. The design further bypasses CPU bottlenecks and reduces redundant memory usage through token-aware buffer management and module-level coordination. Our system outperforms existing implementations in resource efficiency, cutting energy consumption by 42.3\% and GPU memory usage by 11.2\%. This enables a battery-powered device to run LLaVA-OneVision with a camera for nearly 20.8 hours.

DropVLA: An Action-Level Backdoor Attack on Vision-Language-Action Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zonghuan Xu · Jiayu Li · Yunhan Zhao · Xiang Zheng · Xingjun Ma · Yu-Gang Jiang
10 March 2026 at 12:00
arXiv:2510.10932v4 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models map multimodal perception and language instructions to executable robot actions, making them particularly vulnerable to behavioral backdoor manipulation: a hidden trigger introduced during training can induce unintended physical actions while nominal task performance remains intact. Prior work on VLA backdoors primarily studies untargeted attacks or task-level hijacking, leaving fine-grained control over individual actions largely unexplored. In this work, we present DropVLA, an action-level backdoor attack that forces a reusable action primitive (e.g., open_gripper) to execute at attacker-chosen decision points under a realistic pipeline-black-box setting with limited data-poisoning access, using a window-consistent relabeling scheme for chunked fine-tuning. On OpenVLA-7B evaluated with LIBERO, vision-only poisoning achieves 98.67%-99.83% attack success rate (ASR) with only 0.31% poisoned episodes while preserving 98.50%-99.17% clean-task retention, and successfully triggers the targeted action within 25 control steps at 500 Hz (0.05 s). Text-only triggers are unstable at low poisoning budgets, and combining text with vision provides no consistent ASR improvement over vision-only attacks. The backdoor remains robust to moderate trigger variations and transfers across evaluation suites (96.27%, 99.09%), whereas text-only largely fails (0.72%). We further validate physical-world feasibility on a 7-DoF Franka arm with pi0-fast, demonstrating non-trivial attack efficacy under camera-relative motion that induces image-plane trigger drift. These results reveal that VLA models can be covertly steered at the granularity of safety-critical actions with minimal poisoning and without observable degradation of nominal performance.

Ego-Vision World Model for Humanoid Contact Planning

✇cs.AI, q-bio.NC updates on arXiv.org
By: Hang Liu · Yuman Gao · Sangli Teng · Yufeng Chi · Yakun Sophia Shao · Zhongyu Li · Maani Ghaffari · Koushil Sreenath
10 March 2026 at 12:00
arXiv:2510.11682v2 Announce Type: replace-cross Abstract: Enabling humanoid robots to exploit physical contact, rather than simply avoid collisions, is crucial for autonomy in unstructured environments. Traditional optimization-based planners struggle with contact complexity, while on-policy reinforcement learning (RL) is sample-inefficient and has limited multi-task ability. We propose a framework combining a learned world model with sampling-based Model Predictive Control (MPC), trained on a demonstration-free offline dataset to predict future outcomes in a compressed latent space. To address sparse contact rewards and sensor noise, the MPC uses a learned surrogate value function for dense, robust planning. Our single, scalable model supports contact-aware tasks, including wall support after perturbation, blocking incoming objects, and traversing height-limited arches, with improved sample efficiency and multi-task capability over on-policy RL. Deployed on a physical humanoid, our system achieves robust, real-time contact planning from proprioception and ego-centric depth images. Code and dataset are available at our website: https://ego-vcp.github.io/

DevBench: A Realistic, Developer-Informed Benchmark for Code Generation Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Pareesa Ameneh Golnari · Adarsh Kumarappan · Wen Wen · Xiaoyu Liu · Gabriel Ryan · Yuting Sun · Shengyu Fu · Elsie Nallipogu
10 March 2026 at 12:00
arXiv:2601.11895v2 Announce Type: replace-cross Abstract: DevBench is a telemetry-driven benchmark designed to evaluate Large Language Models (LLMs) on realistic code completion tasks. It includes 1,800 evaluation instances across six programming languages and six task categories derived from real developer telemetry, such as API usage and code purpose understanding. Unlike prior benchmarks, it emphasizes ecological validity, avoids training data contamination, and enables detailed diagnostics. The evaluation combines functional correctness, similarity-based metrics, and LLM-judge assessments focused on usefulness and contextual relevance. 9 state-of-the-art models were assessed, revealing differences in syntactic precision, semantic reasoning, and practical utility. Our benchmark provides actionable insights to guide model selection and improvement-detail that is often missing from other benchmarks but is essential for both practical deployment and targeted model development.
Received — 5 March 2026 ⏭

Mozi: Governed Autonomy for Drug Discovery LLM Agents

✇cs.AI, q-bio.NC updates on arXiv.org
By: He Cao · Siyu Liu · Fan Zhang · Zijing Liu · Hao Li · Bin Feng · Shengyuan Bai · Leqing Chen · Kai Xie · Yu Li
5 March 2026 at 13:00
arXiv:2603.03655v1 Announce Type: new Abstract: Tool-augmented large language model (LLM) agents promise to unify scientific reasoning with computation, yet their deployment in high-stakes domains like drug discovery is bottlenecked by two critical barriers: unconstrained tool-use governance and poor long-horizon reliability. In dependency-heavy pharmaceutical pipelines, autonomous agents often drift into irreproducible trajectories, where early-stage hallucinations multiplicatively compound into downstream failures. To overcome this, we present Mozi, a dual-layer architecture that bridges the flexibility of generative AI with the deterministic rigor of computational biology. Layer A (Control Plane) establishes a governed supervisor--worker hierarchy that enforces role-based tool isolation, limits execution to constrained action spaces, and drives reflection-based replanning. Layer B (Workflow Plane) operationalizes canonical drug discovery stages -- from Target Identification to Lead Optimization -- as stateful, composable skill graphs. This layer integrates strict data contracts and strategic human-in-the-loop (HITL) checkpoints to safeguard scientific validity at high-uncertainty decision boundaries. Operating on the design principle of ``free-form reasoning for safe tasks, structured execution for long-horizon pipelines,'' Mozi provides built-in robustness mechanisms and trace-level audibility to completely mitigate error accumulation. We evaluate Mozi on PharmaBench, a curated benchmark for biomedical agents, demonstrating superior orchestration accuracy over existing baselines. Furthermore, through end-to-end therapeutic case studies, we demonstrate Mozi's ability to navigate massive chemical spaces, enforce stringent toxicity filters, and generate highly competitive in silico candidates, effectively transforming the LLM from a fragile conversationalist into a reliable, governed co-scientist.

Can Large Language Models Derive New Knowledge? A Dynamic Benchmark for Biological Knowledge Discovery

✇cs.AI, q-bio.NC updates on arXiv.org
By: Chaoqun Yang · Xinyu Lin · Shulin Li · Wenjie Wang · Ruihan Guo · Fuli Feng · Tat-Seng Chua
5 March 2026 at 13:00
arXiv:2603.03322v1 Announce Type: cross Abstract: Recent advancements in Large Language Model (LLM) agents have demonstrated remarkable potential in automatic knowledge discovery. However, rigorously evaluating an AI's capacity for knowledge discovery remains a critical challenge. Existing benchmarks predominantly rely on static datasets, leading to inevitable data contamination where models have likely seen the evaluation knowledge during training. Furthermore, the rapid release cycles of modern LLMs render static benchmarks quickly outdated, failing to assess the ability to discover truly new knowledge. To address these limitations, we propose DBench-Bio, a dynamic and fully automated benchmark designed to evaluate AI's biological knowledge discovery ability. DBench-Bio employs a three-stage pipeline: (1) data acquisition of rigorous, authoritative paper abstracts; (2) QA extraction utilizing LLMs to synthesize scientific hypothesis questions and corresponding discovery answers; and (3) QA filter to ensure quality based on relevance, clarity, and centrality. We instantiate this pipeline to construct a monthly-updated benchmark covering 12 biomedical sub-domains. Extensive evaluations of SOTA models reveal current limitations in discovering new knowledge. Our work provides the first dynamic, automatic framework for assessing the new knowledge discovery capabilities of AI systems, establishing a living, evolving resource for AI research community to catalyze the development of knowledge discovery.

CubeComposer: Spatio-Temporal Autoregressive 4K 360{\deg} Video Generation from Perspective Video

✇cs.AI, q-bio.NC updates on arXiv.org
By: Lingen Li · Guangzhi Wang · Xiaoyu Li · Zhaoyang Zhang · Qi Dou · Jinwei Gu · Tianfan Xue · Ying Shan
5 March 2026 at 13:00
arXiv:2603.04291v1 Announce Type: cross Abstract: Generating high-quality 360{\deg} panoramic videos from perspective input is one of the crucial applications for virtual reality (VR), whereby high-resolution videos are especially important for immersive experience. Existing methods are constrained by computational limitations of vanilla diffusion models, only supporting $\leq$ 1K resolution native generation and relying on suboptimal post super-resolution to increase resolution. We introduce CubeComposer, a novel spatio-temporal autoregressive diffusion model that natively generates 4K-resolution 360{\deg} videos. By decomposing videos into cubemap representations with six faces, CubeComposer autoregressively synthesizes content in a well-planned spatio-temporal order, reducing memory demands while enabling high-resolution output. Specifically, to address challenges in multi-dimensional autoregression, we propose: (1) a spatio-temporal autoregressive strategy that orchestrates 360{\deg} video generation across cube faces and time windows for coherent synthesis; (2) a cube face context management mechanism, equipped with a sparse context attention design to improve efficiency; and (3) continuity-aware techniques, including cube-aware positional encoding, padding, and blending to eliminate boundary seams. Extensive experiments on benchmark datasets demonstrate that CubeComposer outperforms state-of-the-art methods in native resolution and visual quality, supporting practical VR application scenarios. Project page: https://lg-li.github.io/project/cubecomposer
❌