Searcharxiv⌕ Search

arXiv subjects

Ming Xu

Publications and source records attributed to Ming Xu.

At least 19 recordsLinked to original sources
arXiv2026

Distillation as Probability Transport: Routed On-Policy Distillation

On-policy distillation (OPD) transfers teacher knowledge on student-generated trajectories, but efficient sampled objectives reduce the teacher distribution to scalar credit on individual tokens. Such credit indicates whether a token should gain or lose probability, yet leaves the corresponding redistribution unspecified. We recast OPD as teacher-guided probability transport and propose RouteOPD (Routed On-Policy Distillation), which decomposes local teacher--student disagreement into student-excess sources and teacher-deficit destinations and couples them into explicit transport pairs. RouteOPD optimizes pairwise log-odds toward jointly realizable targets obtained from a bounded teacher potential, while adapting the transport budget to the concentration of teacher demand. This formulation directs updates toward teacher-preferred destinations and controls their magnitude within a single transport operator. Experiments across four teacher--student settings and four mathematical-reasoning benchmarks demonstrate that RouteOPD consistently outperforms sampled reverse-KL OPD, with improvements accompanied by higher routing fidelity and lower background leakage. These results demonstrate the effectiveness of explicitly modeling probability transport in on-policy distillation.

cs.LG↗
arXiv2026

When Self-Evolution Backfires: Pre-Commit Gating against Skill Contamination in LLM Agents

Self-evolving agents accumulate capability by distilling reusable skills from their execution trajectories, but we find this process is not monotonic: past a critical pool size, newly added skills degrade performance instead of improving it. We formalize this capability-contamination phase transition and trace it to a structural cause: once a defective skill enters the decision context, it becomes reference material for distilling later skills, forming cross-round contamination chains. We further show the contamination is structurally irreversible: removing a source skill after the fact cannot erase the flawed reasoning its descendants have already inherited, so post-hoc rollback recovers only a small fraction of the lost performance. This makes skill admission a pre-commit necessity rather than a post-hoc fix, and motivates Verifier-as-Gatekeeper (VaG): a progressive trust hierarchy whose three heterogeneous critics - structural validity, behavioral harmlessness, and semantic consistency - filter each skill individually, coupled with a marginal-gain subset selection that removes combinatorial contamination at the top tier before skills reach the runtime context. On Terminal-Bench 2, unconditional accumulation rises to a peak and then degrades, giving back most of its gains as the pool keeps growing, and post-hoc removal of the culprit skills recovers only a small part of the drop - the empirical signature of irreversibility. In contrast, VaG improves every round, reaching 72% pass@1 with a pool roughly 5x smaller, and its frozen skill pool transfers positively to four other backbones and a second benchmark without re-evolution. Ablations confirm the three critics are complementary and mutually non-substitutable, each intercepting a largely disjoint class of harmful skills.

cs.AI↗
arXiv2026

First Observation of Fishbone-Driven Zonal Flows with Fine Reversed Structure in Tokamak Plasmas

We present the first direct experimental observation of fishbone-driven zonal flows in the core of the EAST tokamak. In contrast to the global pattern predicted by previous models and simulations based on the energetic-particle-expulsion mechanism, the observed flows exhibit a fine-scale, radially reversed structure inside the q = 1 rational surface. The flow rises faster and saturates earlier than the fishbone within a single burst, indicating that a beat-driven nonlinear process dominates the early stage rather than the energetic-particle-expulsion mechanism. Global nonlinear gyrokinetic simulations quantitatively reproduce the observed radial profile and reveal that this structure arises from the cancellation of comparable but opposite contributions from thermal ions and electrons. This cancellation mechanism is not captured in previous theoretical frameworks. These findings establish that the fishbone can generate sheared flows with a distinct radial topology, offering a promising pathway for regulating turbulence and improving core confinement.

physics.plasm-ph↗
arXiv2026

ShortOPD: Recovering Pruned LLMs with Short-to-Long On-Policy Distillation

Structured pruning is a hardware-friendly way to compress LLMs, but it is mostly validated on multiple-choice recognition tasks, while the same compressed checkpoints can collapse on the free-form generation that deployment actually requires. Two observations trace this gap. First, greedy \textsc{pass}@$1$ nearly vanishes after compression, yet \textsc{pass}@$k$ recovers substantially under repeated sampling: useful generations are demoted, not erased. Second, the recoverable regime fails mainly through suffix repetition. Recovery should therefore train on the compressed model's own on-policy states with dense token-level supervision, which On-Policy Distillation (OPD) provides by reusing the pre-compression model as a frozen teacher. However, long on-policy rollouts spend early recovery budget on low-information repetitive suffixes, delaying loss descent. To mitigate this waste, we propose \textbf{\shortopd}, a short-to-long OPD schedule that detects teacher-confirmed repetitive suffixes, treats the surviving prefix as each rollout's effective length, and allocates future rollout budgets to the effective lengths the policy can currently use. Across math, code, and open-ended generation, \shortopd\ raises the compressed model's score to about $9\times$ its unrecovered value and $1.6$--$4.4\times$ standard recovery recipes (SFT w/o KD, KD, and SeqKD), and it matches a fixed $8192$-token rollout horizon within two points using a quarter of the training time ($8.5$ vs.\ $35.9$ hours) and $71\%$ fewer rollout tokens. We hope this recipe helps move structured pruning beyond marginal gains on perplexity and multiple-choice benchmarks, a step closer to deployment-ready generation quality.

cs.LG↗
arXiv2026

Scalable and Differentiable Point-Cloud Registration Using Maximum Mean Discrepancy

We present MMD-Reg, a novel correspondence-free approach to point-cloud registration that is differentiable and has linear computational complexity in the number of points. We model registration as a nonlinear least-squares problem based on the Maximum Mean Discrepancy, approximated using random Fourier features. The resulting objective can be solved efficiently with standard methods such as Levenberg-Marquardt, and the solution is differentiable via the implicit function theorem. This allows MMD-Reg to be used as a differentiable optimization layer within end-to-end trainable models, supporting registration under challenging conditions such as poor initial alignment and partial overlap. We demonstrate this Neural MMD-Reg formulation by integrating the layer with a set transformer, training the resulting model in supervised and unsupervised settings, and comparing its performance against recent learning-based methods. We also evaluate standalone MMD-Reg, comparing its accuracy and scalability against widely used non-learning-based registration methods.

cs.CV↗
arXiv2026

Structure-Aware Graph Multi-Task Learning for Dynamic Sparse OD Demand Prediction

Origin-Destination (OD) demand prediction is fundamental to intelligent transportation systems, yet real-world OD flows are often dynamically sparse, long-tailed, and characterized by heterogeneous zero-flow patterns. These properties make it difficult to distinguish whether an OD connection is active from how much demand it generates once activated. Many existing methods primarily treat OD prediction as a single flow regression task, which limits their ability to model low-frequency, intermittent, and long-tailed OD interactions. To address these challenges, we propose SAGMTL, a Structure-Aware Graph Multi-Task Learning framework for dynamic sparse OD demand prediction. SAGMTL decomposes OD prediction into structural state modeling and flow intensity estimation, jointly learning regional activity states, OD connection activity, and edge-level flow intensity within a unified framework. Specifically, a node-edge collaborative representation module captures regional semantics, temporal dynamics, and spatial priors through interactive node-edge updates, producing structure-aware representations for dynamic OD interactions. Based on these representations, SAGMTL estimates OD flows by jointly modeling stable demand patterns and short-term fluctuations. A multi-constraint objective further improves sparsity awareness and structural consistency. Experiments on three real-world urban mobility datasets from Beijing, Chengdu, and Nanjing show that SAGMTL achieves superior overall performance compared with state-of-the-art baselines. Further analysis demonstrates that explicitly modeling regional activity, connection states, and flow intensity improves the robustness of dynamic sparse OD demand prediction.

cs.LG↗
arXiv2026

Continuous Cross-Domain Traffic State Prediction via Memory-Augmented Graph Liquid Time-Constant Networks

Traffic state prediction is a fundamental task in intelligent transportation systems. In practical applications, some regions suffer from limited traffic observations due to insufficient sensing infrastructure, making cross-domain knowledge transfer an important solution for data-scarce traffic prediction. However, existing cross-domain traffic prediction methods still face several limitations, including coarse-grained source-target adaptation, limited capability in handling unseen target-domain patterns, and insufficient modeling of continuous traffic dynamics under irregular or heterogeneous temporal conditions. To address these issues, this paper proposes a continuous cross-domain traffic prediction framework, termed Memory-Augmented Graph Liquid Time-Constant Network (MA-GLTC). Specifically, we first construct spatio-temporal units (STUs) to decompose traffic networks into transferable local units, enabling fine-grained knowledge alignment across domains. Then, a graph liquid time-constant network (GLTC) is developed to model graph-coupled traffic evolution in continuous time. Different from generic graph neural ODE-based models, GLTC introduces graph-coupled recurrent conductance into liquid time-constant dynamics, allowing node states to evolve with leakage, adaptive time constants, and neighborhood-aware feedback. Furthermore, a Memory-based Transfer Storage (MTS) mechanism is designed to preserve source-domain knowledge, retrieve matched traffic patterns, and update reliable target-domain patterns when unseen states emerge. Experiments on five public traffic datasets demonstrate that MA-GLTC consistently outperforms representative innerdomain and cross-domain baselines in both short-term and longterm prediction tasks. Compared with the second-best method, MA-GLTC reduces the average prediction errors by 3.02%, 0.33%, 8.92%, 10.09%, and 2.11%, respectively.

cs.LG↗
arXiv2026

Improving Human Diving Endurance with a Field-Deployable, Untethered Exoskeleton

Human endurance in underwater locomotion is fundamentally restricted by high energetic demands to overcome drag and the finite supply of self-contained breathing gas. While exoskeleton technology can reduce the metabolic cost of humans in terrestrial locomotion, its potential to enhance human endurance during underwater diving remains entirely unexplored. Here, we present DiveMate, a field-deployable, untethered exoskeleton designed to improve human diving endurance via adaptive kick assistance in real-world underwater environments. During naturalistic diving, DiveMate increases the travel distance using a given energy (breathing gas) by 42.9% and extends dive duration by 54.9% through reducing gas consumption rate. Marked reductions in muscle activation indicate a decrease in physiological exertion, with the net gas consumption rate decreasing by 47.0%. Kinematic characteristics and regularity improvements further underpin efficient energy economy. These results suggest that applying exoskeleton assistance is beneficial for improving human diving endurance and augmenting their ability to explore the aquatic world. This study extends the application frontier of exoskeletons and provides a potential reference for the design and assessment of future underwater assistive devices.

cs.RO↗
arXiv2026

Unconventional crystallization pathway bypassing the intermediate cubic phase in phase-change superlattices

The Ge-Sb-Te (GST) superlattice phase-change material is a promising candidate for overcoming the high power-consumption of phase-change memory (PCM). However, the working mechanism of the superlattice PCM remains controversial. Partial amorphization, which is currently considered the most plausible mechanism, remains hotly debated: how does the partially amorphized GST recrystallize into its superlattice phase instead of the conventionally expected cubic phase? Here, we address this issue using large-scale molecular dynamics simulations enabled by a machine-learning interatomic potential. Starting from a partially melted GST superlattice, we demonstrate that the residual crystalline regions serve as nuclei, enabling the amorphous GST to recrystallize directly into the superlattice phase without passing through the intermediate cubic phase. Moreover, the recrystallized phase is not an ideal superlattice, but rather a structurally ordered and chemically disordered defective superlattice characterized by anti-site defects and stacking faults. The defective superlattice region is also more susceptible to melting than the defect-free superlattice, and thereby can act as the active region of the PCM device. These results help to clarify the longstanding debates concerning the mechanism of superlattice-based PCM.

cond-mat.mtrl-sci↗
arXiv2026

Global Convergence of a Line-Search Filter Differential Dynamic Programming Method

In this article, we establish the global convergence properties of the FilterDDP algorithm, which extends the discrete-time differential dynamic programming (DDP) algorithm of Mayne and Jacobson [\emph{International Journal of Control}, 3, (1966), pp. 85-95] to handle nonlinear constraints over states and controls, in addition to the dynamics. FilterDDP adopts a line-search filter procedure for step acceptance. However, instead of a damped Newton step applied in the general nonlinear programming setting, the computation of a trial point involves applying a backward recursion and a forward simulation. We establish the global convergence of FilterDDP by showing that for a subset of constrained optimal control problems, the this backward-forward procedure satisfies the same properties as a Newton step for the purpose of establishing global convergence of a line-search filter method, following the analysis of W\"achter and Biegler [\emph{SIAM Journal on Optimization}, 16 (2005), pp. 1-31].

math.OC↗
arXiv2026

Securing LLM Agents Need Intent-to-Execution Integrity

This position paper argues that securing LLM agents requires first defining an end-to-end correctness property that specifies when an agent's execution faithfully reflects the user's intent. Modern LLM agents operate over an \emph{intent-to-execution pipeline}, where natural-language instructions are translated into concrete system operations such as tool calls, API requests, and code execution. While recent defenses have made progress in constraining how agents construct tool calls, most existing formulations implicitly assume that tools are trusted. The emergence of systems such as OpenClaw, with open ecosystems of third-party skills and direct access to user environments, breaks this assumption and exposes new failure modes, including malicious or over-privileged components in the execution pipeline. Despite rapid progress in defense mechanisms, there is no adequate correctness property that defines what ``secure'' means for LLM agents, nor a principled way to evaluate the coverage of existing defenses. We observe that LLM agents are structurally analogous to compilers, where security violations correspond to mis-executions that do not preserve user intent. Drawing on this analogy, we identify two fundamental problem sources -- untrusted data ingestion and untrusted tool execution -- and derive four integrity properties that must hold simultaneously: \emph{Tool Integrity}, \emph{Instruction Integrity}, \emph{Judgment Integrity}, and \emph{Data Flow Integrity}. We call their conjunction \emph{intent-to-execution integrity}. Analyzing existing agentic defenses against these properties reveals that current systems provide only partial and non-compositional coverage, leaving fundamental gaps in securing modern LLM agents.

cs.CR↗
arXiv2026

Model Checking Matrix Product States against Linear Chain Logic

Matrix product states (MPS) are a standard tensor-network representation for ground states of one-dimensional quantum many-body systems, and they underpin widely used simulation tools such as DMRG. However, while quantum model checking has been developed mainly for quantum programs and communication protocols (with properties expressed along a time axis), there is still no comparable framework for systematically verifying \emph{spatial} and \emph{size-dependent} properties of physical many-body states, where the key parameter is the system size. This paper takes a step toward bridging the gap. We propose \emph{Linear Chain Logic} (LCL), a spatial logic designed to specify physically meaningful properties of periodic MPS families as the system size grows, such as nontriviality on rings and large-size asymptotic patterns. Our approach builds on a simple but powerful connection: every periodic MPS naturally induces a completely positive map (a quantum operation) on its virtual space, so many quantitative features of the MPS can be analysed through the repeated application of the operation. Using this perspective, we derive an effective procedure to compute the inner products of an MPS at a given size and to support richer LCL specifications, without relying on brute-force state expansion. We then develop approximate model-checking algorithms that combine sound bounding with asymptotic structural analysis, enabling scalable reasoning about large system sizes. Experiments on representative MPS families illustrate that our method can automatically verify nontriviality and detect asymptotic spatial regimes in a way that complements traditional numerical techniques.

quant-ph↗
arXiv2026

On the degradations of Binary-Input Discrete Memoryless Channels

For the polar codes introduced by Arikan in 2009, the first code family achieving the capacity of binary-input discrete memoryless channels (BIDMCs) with low-complexity encoding and decoding, it is crucial to evaluate the reliability of the synthetic channels resulted in the code construction. Since the synthetic channels have an output alphabet that grows exponentially with the code length, an effective method for faithfully evaluating their reliability is to replace them with degradations of manageable alphabet size. The main aim of this paper is to find the optimal degradations of symmetric BIDMCs. We determine all the degradations with the minimum probability of error decoding and give a necessary condition for the degradations with the maximum symmetric capacity. Finally, based on this necessary condition, we propose an efficient algorithm for finding degradation schemes that maximize the symmetric capacity.

cs.IT↗
arXiv2026

EOS-Bench: A Comprehensive Benchmark for Earth Observation Satellite Scheduling

Earth observation satellite imaging scheduling is a challenging NP-hard combinatorial optimisation problem central to space mission operations. While next-generation agile Earth observation satellites (EOS) increase operational flexibility, they also significantly raise scheduling complexity. The lack of a unified, open-source benchmark makes it difficult to compare algorithms across studies. This paper introduces EOS-Bench, a comprehensive framework for systematic and reproducible evaluation of scheduling methods. By integrating high-fidelity orbital dynamics and platform constraints, EOS-Bench generates 1,390 scenarios and 13,900 benchmark instances, spanning from small-scale validation cases to large coordination problems with up to 1,000 satellites and 10,000 requests. We further propose a scenario characterisation scheme to quantify structural difficulty based on factors such as opportunity density, task flexibility, conflict intensity, and satellite congestion. A multidimensional evaluation protocol is introduced, assessing performance across five metrics: task profit, completion rate, workload balance, timeliness, and runtime. The framework is evaluated using mixed-integer programming, heuristics, meta-heuristics, and deep reinforcement learning across both agile and non-agile settings. Results show that EOS-Bench effectively distinguishes solver performance across scales and conditions, revealing trade-offs between solution quality and computational efficiency, and providing deeper insight into scenario complexity. EOS-Bench offers a unified and extensible open testbed for advancing research in Earth observation satellite scheduling. The code and data are available at https://github.com/Ethan19YQ/EOS-Bench.

cs.NI↗
arXiv2026

ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents

Language-model agents are increasingly used as persistent coworkers that assist users across multiple working days. During such workflows, the surrounding environment may change independently of the agent: new emails arrive, calendar entries shift, knowledge-base records are updated, and evidence appears across images, scanned PDFs, audio, video, and spreadsheets. Existing benchmarks do not adequately evaluate this setting because they typically run within a single static episode and remain largely text-centric. We introduce \bench{}, a benchmark for coworker agents built around multi-turn multi-day tasks, a stateful sandboxed service environment whose state evolves between turns, and rule-based verification. The current release contains 100 tasks across 13 professional scenarios, executed against five stateful sandboxed services (filesystem, email, calendar, knowledge base, spreadsheet) and scored by 1537 deterministic Python checkers over post-execution service state; no LLM-as-judge is invoked during scoring. We benchmark seven frontier agent systems. The strongest model reaches 75.8 weighted score, but the best strict Task Success is only 20.0\%, indicating that partial progress is common while complete end-to-end workflow completion remains rare. Turn-level analysis shows that performance drops after the first exogenous environment update, highlighting adaptation to changing state as a key open challenge. We release the benchmark, evaluation harness, and construction pipeline to support reproducible coworker-agent evaluation.

cs.CV↗
arXiv2026

GraSP: Graph-Structured Skill Compositions for LLM Agents

Skill ecosystems for LLM agents have matured rapidly, yet recent benchmarks show that providing agents with more skills does not monotonically improve performance -- focused sets of 2-3 skills outperform comprehensive documentation, and excessive skills actually hurt. The bottleneck has shifted from skill availability to skill orchestration: agents need not more skills, but a structural mechanism to select, compose, and execute them with explicit causal dependencies. We propose GraSP, the first executable skill graph architecture that introduces a compilation layer between skill retrieval and execution. GraSP transforms flat skill sets into typed directed acyclic graphs (DAGs) with precondition-effect edges, executes them with node-level verification, and performs locality-bounded repair through five typed operators -- reducing replanning from O(N) to O(d^h). Across ALFWorld, ScienceWorld, WebShop, and InterCode with eight LLM backbones, GraSP outperforms ReAct, Reflexion, ExpeL, and flat skill baselines in every configuration, improving reward by up to +19 points over the strongest baseline while cutting environment steps by up to 41%. GraSP's advantage grows with task complexity and is robust to both skill over-retrieval and quality degradation, confirming that structured orchestration -- not larger skill libraries -- is the key to reliable agent execution.

cs.CL↗
arXiv2026

ARuleCon: Agentic Security Rule Conversion

Security Information and Event Management (SIEM) systems make it possible for detecting intrusion anomalies in real-time manner by their applied security rules. However, the heterogeneity of vendor-specific rules (e.g., Splunk SPL, Microsoft KQL, IBM AQL, Google YARA-L, and RSA ESA) makes cross-platform rule reuse extremely difficult, requiring deep domain knowledge for reliable conversion. As a result, an autonomous and accurate rule conversion framework can significantly lead to effort savings, preserving the value of existing rules. In this paper, we propose ARuleCon, an agentic SIEM-rule conversion approach. Using ARuleCon, the security professionals do not need to distill the source rules' logic, the documentation of the target rules and ARuleCon can purposely convert to the target vendors without more intervention. To achieve this, ARuleCon is equipped with conversion/schema mismatches, and Python-based consistency check that running both source and target rules in controlled test environments to mitigate subtle semantic drifts. We present a comprehensive evaluation of ARuleCon ranging from textual alignment and the execution success, showcasing ARuleCon can convert rules with high fidelity, outperforming the baseline LLM model by 15% averagely. Finally, we perform case studies and interview with our industry collaborators in Singtel Singapore, which showcases that ARuleCon can significantly save expert's time on understanding cross-SIEM's documentation and remapping logic.

cs.CR↗
arXiv2026

Search-P1: Path-Centric Reward Shaping for Stable and Efficient Agentic RAG Training

Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by incorporating external knowledge, yet traditional single-round retrieval struggles with complex multi-step reasoning. Agentic RAG addresses this by enabling LLMs to dynamically decide when and what to retrieve, but current RL-based training methods suffer from sparse outcome rewards that discard intermediate signals and low sample efficiency where failed samples contribute nothing. We propose Search-P1, a framework that introduces path-centric reward shaping for agentic RAG training, comprising two key components: (1) Path-Centric Reward, which evaluates the structural quality of reasoning trajectories through order-agnostic step coverage and soft scoring that extracts learning signals even from failed samples, and (2) Dual-Track Path Scoring with offline-generated reference planners that assesses paths from both self-consistency and reference-alignment perspectives. Experiments on multiple QA benchmarks demonstrate that Search-P1 achieves significant improvements over Search-R1 and other strong baselines, with an average accuracy gain of 7.7 points.

cs.CL↗