arXiv subjects

Yue Zhang

Publications and source records attributed to Yue Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

MobiAgent: Dual-Loop Recursive Policy Self-Improvement for Long-Horizon Mobile Manipulation

Long-horizon mobile manipulation presents significant challenges due to compounding execution errors and capacity interference between locomotion and arm control. While recent Vision-Language-Action models excel at short-horizon tasks, they lack the hierarchical reasoning required for multi-stage objectives. Furthermore, existing hierarchical agents suffer from rigid sub-task mapping, inflexible replanning, and a lack of continuous learning. To address these limitations, we introduce MobiAgent, a dual-loop agentic framework that bridges robust deployment execution and recursive policy self-improvement. During deployment, the Inner Loop decouples high-level reasoning from low-level control through highly composable atomic skills. It employs Vision-Language models for receding-horizon planning and visual reflection, dynamically composing skills to ensure robust error recovery. These skills are executed by specialized flow-matching experts that share a unified VLM backbone, maximizing reusability while mitigating capacity interference. Concurrently, the Outer Loop drives automated lifelong learning by autonomously segmenting and verifying deployment rollouts, clustering them to discover atomic skills, and continuously fine-tuning the skill library without human annotations. Evaluations on RoboCasa, BEHAVIOR-1K, and real-world tasks demonstrate the effectiveness of MobiAgent. It outperforms $π_{0.5}$-TA by 22.5 percentage points on BEHAVIOR-1K and enables robust recovery from execution failures. Through autonomous data recycling, success improves from 7.50% to 27.50% on RoboCasa and from 32.5% to 57.5% on Astribot S1.

cs.RO↗

arXiv2026

From bits to atoms: A survey of cross-layer safety in Embodied AI

Embodied Artificial Intelligence (AI) introduces cyber-physical risks because digital failures and attacks can cause physical harm. This survey organizes embodied AI safety through a four-layer reference architecture -- comprising the physical interaction, system and middleware, algorithm and decision, and cloud-edge and data layers -- and uses it to connect vulnerabilities, propagation paths, and safeguards across the system stack. It traces the evolution of safety paradigms and examines how faults and attacks cross architectural boundaries to produce physical consequences. Furthermore, the survey systematizes composite attack chains, outlines multi-layered defense strategies, and reviews representative embodied safety benchmarks. Ultimately, we advocate for unifying software alignment with rigorous hardware constraints to build trustworthy systems capable of operating safely within human-centric environments.

cs.RO↗

arXiv2026

Ego2Act: Evaluating Goal-Directed Manipulation in Egocentric Video Generation

Video generation models are increasingly being explored as world simulators for embodied planning and learning. To do so effectively, these models must not only generate visually appealing frames, but also predict how environments dynamically evolve when executing goal-directed actions. While evaluating these capabilities is crucial, existing benchmarks focus mainly on single short actions or step-by-step instructions. This leaves multi-step physical reasoning underexplored, especially in egocentric video generation that requires planning to simulate proper execution to accomplish high-level goals by carrying out multiple real-world manipulations. We introduce Ego2Act, a goal-directed benchmark featuring 2,640 videos from 110 real-world tasks across day-to-day settings, varying object clutter and multi-step complexity. Given an initial scene image and a high-level goal, Ego2Act evaluates whether video generation models can produce realistic egocentric videos of a hand manipulating objects to carry out the task. To support scalable evaluation, we also introduce Ego2ActJudge, a reference-free evaluation pipeline that achieves better task completion and physics plausibility evaluation alignment with human consensus compared to relevant baselines. Our findings reveal that models' generated simulations often skip or partially execute steps, leaving later steps missing dependent states, which leads to unfulfilled goal. Furthermore, models consistently fail at fine-grained physical dynamics, particularly during complex object manipulation and persistent world modeling. We hope Ego2Act provides a rigorous testbed for advancing video models toward physically plausible, goal-directed simulation.

cs.CV↗

arXiv2026

Semifactual Credit-Augmented Policy Optimization

Reinforcement learning with verifiable rewards (RLVR) has improved the reasoning capabilities of large language models (LLMs), yet their predictions remain sensitive to task-irrelevant prompt features. We investigate this sensitivity through semifactual prompt interventions that preserve the underlying problem and its answer. Our analysis reveals substantial variation in token-level sensitivity and shows that suppressing high-drift token candidates during decoding improves reasoning accuracy without updating model weights. These findings highlight a limitation of Group Relative Policy Optimization (GRPO), which assigns the same outcome-derived advantage to every response token and may reinforce potential spurious dependence alongside useful reasoning. Motivated by this observation, we introduce Semifactual Credit-Augmented Policy Optimization (SCAPO), a causally inspired variant of GRPO that incorporates semifactual stability into token-level credit assignment. SCAPO measures token probability drift for fixed responses under semifactual interventions and uses normalized stability scores to reduce advantages for relatively unstable tokens during early training, while granting no additional credit for stability alone. On Qwen3-4B-Base and Qwen3-1.7B-Base, SCAPO improves AIME 2024-2026 accuracy over GRPO by 5.63 and 4.17 percentage points, respectively. At both model scales, SCAPO achieves the best results on most evaluated mathematics benchmarks and all evaluated out-of-distribution benchmarks among the compared methods. These results suggest that semifactual stability provides an effective training signal for improving reasoning and generalization through finer-grained credit assignment in RLVR. The code is available at https://github.com/DtYXs/SCAPO.

cs.LG↗

arXiv2026

A Skill-augmented Agentic Framework and Benchmark for Multi-Video Understanding

Multimodal Large Language Models have achieved strong performance in single-video understanding, yet their ability to reason across multiple videos remains limited. Existing approaches typically concatenate multiple videos into a single input and perform direct inference, which introduces training-inference mismatch, information loss from frame compression, and a lack of explicit cross-video coordination. Meanwhile, current multi-video benchmarks primarily emphasize event-level comparison, leaving identity-level matching, fine-grained discrimination, and structured multi-step reasoning underexplored. To address these gaps, we introduce MVX-Bench, a Multi-Video Cross-Dimension Benchmark that reformulates 11 classical computer vision tasks into a unified multi-video question-answering framework, comprising 1,442 questions over 4,255 videos from diverse real-world datasets. We further propose SAMA, a Skill-Augmented Agentic Framework for Multi-Video Understanding, which integrates visual tools, task-specific skills, and a conflict-aware verification mechanism to enable iterative and structured reasoning. Experimental results show that SAMA outperforms strong open-source baselines and GPT on MVX-Bench, and ablations validate the effectiveness of skill design and conflict resolution.

cs.CV↗

arXiv2026

UNBIND: UNlearning By INference-time Directional Steering for Code LLMs

Code large language models acquire programming capabilities from large code corpora, but can also memorize implementations that later require removal. Code unlearning is needed to control their continued reproduction when copyright or security concerns arise. However, targeted and retained code share computational patterns, creating a tension between forgetting specific implementations and preserving general programming ability. We propose \textbf{UNBIND}, a code unlearning framework that separately considers which hidden states correspond to the target code and how to suppress its reproduction. By constructing separate directions for these objectives, UNBIND achieves selective unlearning at inference time while keeping model weights fixed. Our evaluation covers fourteen baselines across two code models and two corpora. UNBIND achieves the highest joint forgetting and utility score in every setting. It reduces target code reproduction by 97.3\% to 99.1\% as measured by F-BLEU, with at most two fewer HumanEval+ and six fewer MBPP+ problems solved than the original models. In repeated extraction tests under a fixed budget, the number of targets yielding exact spans of at least 50 tokens falls from 188--262 to 0--2 out of 300 per setting. No extracted span reaches 100 tokens, and the mean best recovery ratio ranges from 0.43\% to 6.45\%. Multilingual and related-code evaluations further show effective forgetting with limited impact on useful programming capabilities, supporting UNBIND as a practical approach to selective code unlearning.

cs.SE↗

arXiv2026

When to Trust Imagination: Adaptive Action Execution for World Action Models

World Action Models (WAMs) jointly predict future visual observations and actions, but typically execute a fixed number of actions before replanning, regardless of whether the imagined future remains consistent with reality. We formulate adaptive WAM execution as future--reality verification and propose Future Forward Dynamics Causal Attention (FFDC), a lightweight verifier that reasons over predicted actions, predicted visual dynamics, real observations, and instruction-conditioned semantics. FFDC allows continued execution when a plan remains reliable and triggers replanning when its validity deteriorates. We combine verification with Mixture-of-Horizon Training to improve long-horizon trajectory coverage. On RoboTwin, FFDC-WAM achieves the highest average success rate among the evaluated methods in both clean and randomized settings. Under randomization, it improves success by 2.54 percentage points over Base-Motus while reducing WAM planning calls by 69.10% and completion time by 33.88%. Notably, across six real-world tasks on a Unitree G1D robot, FFDC-WAM improves average success by 9.17% over Base-Motus-LC16 while reducing calls by 74.06% and completion time by 51.31%.

cs.RO↗

arXiv2026

LAGO: Language-Guided Adaptive Object-Region Focus for Zero-Shot Visual-Text Alignment

Zero-shot visual recognition often depends on localized parts, attributes, and textures that are diluted in global image representations. Existing training-free localized methods alleviate this limitation by evaluating multiple image regions, but their candidate locations are typically generated before class semantics determine where to look, restricting language to post-hoc scoring of already sampled evidence. We introduce LAGO (LAnguage-Guided adaptive Object-region focus), which reframes localized recognition as language-guided directed region discovery. LAGO first establishes a compact, class-agnostic object-centric visual initialization, then constructs a soft semantic prototype from the intermediate prediction to actively steer subsequent search toward class-relevant evidence during inference. This object-first, language-second design addresses the circular dependence between recognizing a class and locating its supporting evidence, while preserving complementary local, contextual, and global cues. Across standard and distribution-shift benchmarks, LAGO transfers with a single fixed configuration and uses candidate regions more effectively under matched budgets. Controlled experiments further isolate the proposed mechanism, showing that allowing language to alter the search trajectory retrieves more relevant regions and improves recognition over applying the same semantic signal only after candidate generation.

cs.CV↗

arXiv2026

Uranus: Building the Next-Generation Simulation Infrastructure for Embodied AI

Scalable simulation is essential for robot data generation, policy training, evaluation, and safe iteration, yet real-world interaction is costly and conventional simulators require labor-intensive construction. We present Uranus, a data-driven robot simulator built around a joint-trajectory-conditioned autoregressive diffusion model. Uranus offers three key capabilities: (1) streaming, open-ended rollout, which receives future joint-position trajectories online and autoregressively generates one latent frame per step, corresponding to four RGB frames, without a fixed horizon; (2) low-latency generation, achieving 24 FPS after inference optimization; and (3) scalable, extensible robot control, providing a unified interface for synchronized multi-view generation across diverse robot embodiments and camera configurations. We conduct comprehensive quantitative and qualitative evaluations on both in-distribution and out-of-distribution data, providing an objective assessment of Uranus and clearly identifying its current limitations. We release the code and model weights to empower the community with practical tools and insights.

cs.RO↗

arXiv2026

Starfish: Rebalancing Multi-Party Off-Chain Payment Channels

Blockchain technology has revolutionized the way transactions are executed, but scalability remains a major challenge. Payment Channel Network (PCN), as a Layer-2 scaling solution, has been proposed to address this issue. However, skewed payments can deplete the balance of one party within a channel, restricting the ability of PCNs to transact through a path and subsequently reducing the transaction success rate. To address this issue, the technology of rebalancing has been proposed. However, existing rebalancing strategies in PCNs are limited in their efficiency and applicability under realistic network topologies. Cycle-based approaches only enable rebalancing among nodes that form cycles, while non-cycle-based approaches incur high on-chain operation costs. In this study, we propose Starfish, a rebalancing approach designed for the locally star-like structures formed by high-degree routing hubs and their incident channels in real-world PCNs. By leveraging a hub-centric structure, Starfish reduces the on-chain overhead of rebalancing from quadratic to linear in the number of channels, achieving high rebalancing efficiency. We formally prove the security of Starfish and conduct comparative experiments against existing rebalancing techniques to demonstrate its effectiveness.

cs.CR↗

arXiv2026

Orthogonal Policy Learning with Ordinal Outcomes

Policy learning methods based on conditional average treatment effects can obscure subpopulation heterogeneity when applied to ordinal outcomes. We develop a policy learning framework for ordinal outcomes with heterogeneous utilities for individuals who strictly benefit from treatment and those who do not. Since the probability of strict benefit is only partially identified without further conditions, we adopt a minimax strategy which minimizes the worst-case regret over the identification region. To estimate the resulting nonsmooth objective, we combine smooth approximations with Neyman orthogonalization to remove first-order bias from nuisance estimation. We also derive an excess worst-case regret bound over a restricted policy class. The proposed method is validated through extensive simulations and an application to the 2022 Survey of Income and Program Participation (SIPP) data.

stat.ME↗

arXiv2026

Entropy Dilution Mechanism and Dark Matter Isocurvature

We explore dark matter isocurvature perturbations in cosmological scenarios with entropy dilution resulting from a late decaying massive state that temporarily dominates the universe before big-bang nucleosynthesis. The dilution mechanism features a non-standard scaling window of the radiation energy density, which allows for a new way to derive the entropy dilution factor. On super-horizon scales, the same scaling window also governs the evolution of isocurvature, and we present an analytic approach to solving the perturbation equations. We characterize various regimes where dark matter isocurvature survives and is subject to constraints from the cosmic microwave background observations. For cosmologically viable scenarios, we find that dark matter is dominantly produced from its dilutor decay, and the isocurvature is suppressed by the entropy dilution factor to the $4/3$ power. Our work demonstrates the complementarity between isocurvature and collisionless damping of dark matter as powerful probes of the entropy dilution mechanism.

hep-ph↗

arXiv2026

An Industrial-Scale Sequential Recommender for LinkedIn Feed Ranking

LinkedIn Feed enables professionals worldwide to discover relevant content, build connections, and share knowledge at scale. We present Feed Sequential Recommender (Feed SR), a transformer-based sequential ranking model for LinkedIn Feed that replaces a DCNv2-based ranker and meets strict production constraints. We detail the modeling choices, training techniques, and serving optimizations that enable deployment at a scale of 1.2 billion members. Feed SR has been serving the majority of LinkedIn's Feed traffic for over three months and shows significant improvements in member engagement (+2.10% time spent, +3.52% like, comments, or reshares) in online A/B tests compared to the existing production model. We also describe our deployment experience with alternative sequential and LLM-based ranking architectures and why Feed SR provided the best combination of online metrics and production efficiency.

cs.IR↗

arXiv2026

Lyα forest bounds on sterile neutrino production via neutrino self-interactions

Sterile neutrinos in the keV mass range have long been considered a well-motivated dark matter (DM) candidate. In this work, we explore a sterile neutrino production mechanism through active neutrino self-interactions in the early universe, assuming that they constitute the full DM abundance. We implement a self-consistent treatment of the sterile-neutrino free streaming and the active-neutrino self-interactions on structure formation, which yield a unique scale-dependent modification to the linear matter power spectrum. We then set bounds on this scenario using a combination of the cosmic microwave background and Ly$α$ forest constraints. Specifically, we utilize the two recent likelihoods derived from eBOSS data: (i) an effective field theory (EFT) based full-shape likelihood and (ii) a compressed likelihood obtained from the PRIYA-simulation emulator. We produce some of the most stringent observational constraints to date on sterile neutrino DM, comparable to the bounds from the most stringent laboratory constraints.

astro-ph.CO↗

arXiv2026

Precision Electroweak Constraints on Neutrinophilic Scalars

Strong self-interaction among the active neutrinos mediated by a neutrinophilic scalar is a well-motivated target of particle physics and cosmological probes. In this article, we present precision electroweak constraints on models for neutrino self-interaction. We first work in the simplified model where the finite radiative corrections are obtained with the guidance of gauge invariance. These corrections are logarithmically enhanced for small mediator masses. We point out the importance of neutrino charged-current coupling correction and its impact on the $Δr$ parameter and Fermi constant measurements. This effect was overlooked previously and allows us to derive the leading constraint on the neutrinophilic couplings for mediator mass above a few hundred MeV. We investigate the robustness of the result in a concrete UV completion which further includes a TeV-scale $SU(2)_L$ triplet scalar and find the simplified model constraints continue to hold for a wide range of parameter space. We pin down moving parts in the UV complete model and conditions when the contributions from heavy particle loops are no longer negligible. Our results serve as a useful road map for future explorations of the self-interacting neutrino paradigm.

hep-ph↗

arXiv2026

Current-Driven Magnetization Switching via Zhang-Li Torque

Current-driven magnetization switching is currently governed by two established mechanisms: the Slonczewski spin-transfer torque (STT) and the spin-orbit torque (SOT). Here we identify the Zhang-Li STT as a third mechanism that operates on a different physical footing. Using micromagnetic simulations, we show that the interfacial Dzyaloshinskii-Moriya interaction (iDMI) inevitably creates spatial magnetization gradients at device edges, enabling the Zhang-Li torque to drive deterministic, field-assisted switching that fully reproduces all hallmark behaviors of SOT, including polarity control and effective-field characteristics. Our work establishes the Zhang-Li torque as an autonomous switching pathway, redefining the physical picture of current-induced magnetization reversal and providing new design principles for spintronic devices.

cond-mat.mes-hall↗

arXiv2026

NovGauge: A Fine-Grained Benchmark for Diagnosing LLMs' Capability in Paper Novelty Assessment

Large language models (LLMs) are increasingly used in peer review at major AI conferences, yet novelty remains a persistent weak point. Existing benchmarks assess novelty as a single holistic score, making it difficult to diagnose which dimension a model misjudges or whether its evidence is faithful. We present NovGauge, a human-anchored benchmark for fine-grained novelty assessment diagnosis. The benchmark contains 619 paper pairs and 50 multi-paper sets, drawn from two expert sources: ICLR reviewer overlap claims and survey co-citations. Instances are independently labeled along three dimensions: task, problem, and method, capturing application goals, technical challenges, and solution approaches. We propose a cascading diagnostic pipeline that verifies per-dimension correctness, evidence grounding, and logical support. Evaluation of 18 LLMs shows hallucination rates ranging from 0% to 39% across dimensions, and among non-hallucinated correct-positive judgments, over 70% cite evidence fails to logically support the stated reason. The best-performing model, GPT-5.5, achieves 43-72% Verified F1 across dimensions, while most models retain less than half of their raw F1 after faithfulness verification. These results suggest that current LLMs remain far from reliable scientific novelty assessment, particularly when correctness is conditioned on faithful evidence grounding.

cs.AI↗

arXiv2026

AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories

Maintaining spatial world consistency over long horizons remains a central challenge for camera-controllable video generation. Existing memory-based approaches often condition generation on globally reconstructed 3D scenes by rendering anchor videos from the reconstructed geometry in the history. However, reconstructing a global 3D scene from multiple views inevitably introduces cross-view misalignment, as pose and depth estimation errors cause the same surfaces to be reconstructed at slightly different 3D locations across views. When fused, these inconsistencies accumulate into noisy geometry that contaminates the conditioning signals and degrades generation quality. We introduce AnchorWeave, a memory-augmented video generation framework that replaces a single misaligned global memory with multiple clean local geometric memories and learns to reconcile their cross-view inconsistencies. To this end, AnchorWeave performs coverage-driven local memory retrieval aligned with the target trajectory and integrates the selected local memories through a multi-anchor weaving controller during generation. Extensive experiments demonstrate that AnchorWeave significantly improves long-term scene consistency while maintaining strong visual quality, with ablation and analysis studies further validating the effectiveness of local geometric conditioning, multi-anchor control, and coverage-driven retrieval.

cs.CV↗