arXiv subjects

Peng Li

Publications and source records attributed to Peng Li.

At least 55 records · Page 3Linked to original sources

arXiv2026

LP-NavOA: Integrated Local Navigation and Obstacle Avoidance for Humanoid Robots under Limited Perception

Humanoid local navigation in cluttered environments must jointly resolve obstacle avoidance, sparse-goal recovery, and stable whole-body locomotion under short-range and partially observable sensing. Explicit planner-control decompositions introduce latency and can mismatch agile humanoid command-tracking limits, while purely reactive controllers may lose the goal after obstacle occlusion. We present LP-NavOA, a limited-perception navigation and obstacle-avoidance framework for humanoid robots. A raycast-conditioned perception-action proximal policy optimization (PPO) locomotion backbone is first trained with a robot-centered circular heading-speed command and a shared command-side safety filter. With this backbone frozen, A-star and waypoint teachers generate rollouts for distilling a recurrent local planner that overwrites only the heading command at deployment, leaving the whole-body policy intact. At runtime, LP-NavOA uses proprioception, short-range local range sensing, and a body-frame goal direction, requiring no global map, waypoint stream, or external planner. In MuJoCo open-wall and indoor layouts, the distilled planner produces obstacle bypassing and post-avoidance goal recovery, raising teacher-calibrated on-time arrival from 38--40\% to 85--97\% and reducing brush/contact-heavy progress relative to a backbone-only controller. Ablations show that dynamic route shaping, teacher-active data collection, and the circular command interface are important for navigation efficiency and for training the 3.0\,m/s backbone. A Unitree G1 deployment analysis demonstrates hardware executability without continuous joystick steering.

cs.RO↗

arXiv2026

A unified resource-pool architecture for high-dimensional direct-detection optical communication

Increasing optical communication capacity without proportionally increasing receiver complexity remains a key challenge for direct-detection links. Conventional systems typically assign wavelength, polarization and intensity to fixed, separately recovered functions, so that alphabet expansion is accompanied by additional demultiplexing, polarization handling, receiver branches and electronic processing. Here we introduce a unified resource-pool architecture for high-dimensional direct-detection optical communication, in which wavelength, polarization and intensity are jointly organized as a composite optical symbol space and recovered through optical-domain joint projection rather than dimension-by-dimension separation. The receiver is implemented with an integrated disordered photonic processor that transforms each composite optical state into a reproducible multi-output electrical fingerprint for single-shot direct recovery. In a dual-wavelength transmission experiment, the system resolves 4096 composite symbols, corresponding to 12 bits per symbol slot, with a bit error rate of 4.25e-4 after 10 km standard-fiber transmission. Additional experiments demonstrate dense polarization alphabets, wavelength-indexed state-space expansion and high-launch-power operation over hollow-core fiber. These results establish disorder-enabled joint projection in an integrated photonic processor as a route to hardware-efficient high-dimensional direct-detection communication beyond conventional dimension-partitioned receiver architecture.

physics.optics↗

arXiv2026

Instability Caused by Integration of IBRs under Strong Grid Connections -- A Practical Case Study on Large-scale Energy Storage Systems

It has been well known that inverter-based resources (IBRs) can lead to converter-driven stability issues under weak grid connections. However, as the number of IBRs increases, instabilities can also occur even under strong grid connections. A practical case is presented to demonstrate this conclusion, using large-scale energy storage systems (ESSs) as an example. In this study, the ESSs induce oscillations with a frequency of 150 Hz in the d-q coordinates while providing both capacitive and inductive reactive power support (achieved by ESS functional control loops) to the connected power system. Theoretical analysis reveals that under strong grid connections, the dynamic interactions among power conversion systems (PCSs) of ESSs can be superimposed and intensified as the ESS scale extends, which reduces oscillation damping and leads to system instability. This indicates that ESS functional control loops also have potential instability risks when providing supports to power systems, which should be carefully examined. Finally, major impact factors are identified to mitigate the oscillations, and the conclusions are validated based on the SIMULINK platform. This paper provides valuable practical insights into system instabilities even under strong grid conditions, emphasizing the importance of functional control design and careful planning of the scale for IBR-dominated systems.

eess.SY↗

arXiv2026

Director: Accelerating Distributed MoE Serving via Online Proactive Expert Placement

Expert parallelism has become the prevailing paradigm to serve Mixture-of-Experts (MoE) models. Its efficiency depends on the communication and computation latencies of the GPUs, which are linked to the placement of experts in the GPUs. Existing works for optimizing expert placement focus on leveraging past requests' expert activation patterns. However, they demonstrate deficiencies facing diverse and rapidly changing request patterns, calling for an online, proactive approach. Implementing such an approach requires addressing several challenges: the uncertainty associated with incoming requests' expert activation, the cost of expert migration, and the NP-hard complexity in optimization. Therefore, we present Director, a new distributed MoE serving system that minimizes end-to-end latency via prediction-driven, online expert placement. Director uses either a lightweight cascaded predictor or a low-bit quantized replica for expert activation patterns of incoming requests. An online migration module then enacts the changes with near-zero downtime by executing migrations in compute-bound phases, keeping disruption bounded. At its core, a relaxation-based expert placement optimizer operates under capacity constraints, runs in polynomial time, and achieves a $(1+ε)$ approximation ratio. Finally, we implement a prototype and demonstrate, through extensive experiments, a reduction in end-to-end latency of $11\sim55\%$ for popular MoE models (e.g., Mistral, DeepSeek and Qwen) compared to existing work.

cs.LG↗

arXiv2026

Interference of critical dynamics associated with zero modes

We study the interference of critical dynamics associated with zero modes (ICDZM) in the generalized Creutz ladders using closed quench paths that pass through two critical points successively. By reading out the final zero-mode transfer probability, we find rich ICDZM interference patterns dependent on the quench path. In particular, when the closed path links two topologically nontrivial phases, the ICDZM pattern may either vanish or exhibit period doubling. Within the framework of WKB analysis, this phenomenon is well clarified by the interference phase accumulated in the quench procedure. We also demonstrate that the zero-mode transfer probability can be detected by the deviation of the boundary particle number from its initial fractional value, which arises from the blending of bulk modes in the critical dynamics. As an edge defect, the zero-mode transfer probability captures both the ICDZM oscillation and the known anomalous defect production in a non-closed quench path. These results identify ICDZM and the corresponding edge defect as probes for critical dynamics associated with topological zero modes.

quant-ph↗

arXiv2026

From Meta Idea to Advanced Mathematical Discovery -- Human-AI Co-Discovery of Sign-Embedding Quantum Algorithms

AI-assisted mathematics is often evaluated on solving predefined problems. In practice, however, many important advances begin earlier, when a vague research intuition is transformed into a concrete problem, a promising route, and a theorem family worth proving. This report studies that stage through a case study that led to sign-embedding quantum algorithms for matrix equations and matrix functions, foundational primitives in quantum linear algebra and operator-output quantum algorithms. The project began with a human-originated intuition that rational approximation is especially effective for jump-type functions such as the sign function, and might therefore serve as a design principle for quantum algorithms. Rather than merely assisting after the problem was fixed, AI-assisted exploration, including workflows later integrated into the agentic AI-mathematician system AIM, played a key role in expanding this intuition into a route map, comparing candidate formulations, and converging toward sign embedding as the central framework. AIM then helped connect a known matrix-sign identity to wider classes of matrix equations and matrix functions, and drafted proof and complexity calculations. The decisive scientific judgments remained human: selecting which human-AI-expanded routes were worth pursuing, rejecting a Cayley-trapezoidal approximation when its validity required a hidden condition, and refining the Sylvester implementation from a coarse quadratic-gap query route to the final factorized and scaled analysis. The report argues that human-AI co-discovery workflows, with systems such as AIM as important components, are most valuable not as standalone theorem provers, but as research partners for problem formation, connection discovery, derivation, and skeptical review inside a human-gated research loop.

cs.LG↗

arXiv2026

TAHOE: Text-to-SQL with Automated Hint Optimization from Experience

Large Language Models (LLMs) have democratized database access through Text-to-SQL, but moving from prototypes to production remains difficult. Real deployments must handle strict SQL dialects, massive schemas, and evolving user preferences, while supervised fine-tuning is costly and rigid and agentic test-time scaling is expensive. We present Tahoe, a system that treats prompt optimization as a dynamic data management problem. Tahoe uses an error-driven hint learning pipeline across Development and Deployment to consolidate debugging traces into a structured Hint Bank. Compiler feedback is distilled into reusable Syntax Hints for dialect-specific rules, while execution and user feedback are converted into Semantic Hints for schema- and user-specific logic. Tahoe further introduces a Strategy Layer that models conflicting user intents as competing strategies under shared natural-language triggers, with recency signals and post-learning attribution statistics that summarize empirical success, harm, inertness, and support. At inference time, Tahoe retrieves relevant hints and guides the LLM through Logic Planning followed by SQL Synthesis. We implement and evaluate the development-phase workflow, leaving deployment-time human-feedback updates for future work. On Spider 2.0-Snow, Tahoe substantially improves Text-to-SQL without updating model parameters. On 113 supervised Spider 2.0-Snow-0212 examples using GPT-5.5, Tahoe raises pass rate from 61.95 percent to 79.42 percent and pass-at-4 from 72.57 percent to 87.61 percent, achieves 100 percent Snowflake syntax pass rate, and reduces average compiler-feedback critic rounds from 2.79 to 0.12 per sampled candidate. The same Hint Bank also transfers to weaker backbones, including a 19.7 percentage-point pass-rate gain on Doubao-2.0-lite.

cs.DB↗

arXiv2026

Efficient RWKV-based Representation Learning for 3D Point Clouds

The recent receptance weighted key value (RWKV) model combines RNN-style recurrence, offering a linear-complexity alternative to Transformers' quadratic self-attention for modeling global dependencies. However, when directly applied to point clouds, RWKV, originally developed for sequential text, struggles to capture local geometric structures and model spatial dependencies effectively. To address this, we propose the \textbf{P-RWKV} block, which bridges the gap between sequence modeling and irregular 3D geometry while preserving the efficiency advantages of RWKV. It consists of a Local Perception Expansion (LPE) component to expand contextual perception along the spatio-temporal sequence and a Spatial Context Enhancement (SCE) component to strengthen spatial awareness. To validate the effectiveness of P-RWKV for point cloud understanding, we construct PointER, a single-modality self-supervised representation learning framework whose encoder is composed of stacked P-RWKV blocks. Furthermore, we extend P-RWKV to a cross-modality setting and integrate the proposed core sub-modules into multiple architectures, demonstrating strong plug-and-play flexibility and architectural generality. Extensive experiments show that the P-RWKV block and its key sub-modules achieve competitive performance across various tasks with lower computational cost and inference latency. Code will be released upon acceptance.

cs.CV↗

arXiv2026

Bridging the Agent-World Gap: Text World Models for LLM-based Agents

Large language model (LLM)-based agents are increasingly used in interactive textual environments, from web navigation and code editing to tool use and long-horizon dialogue. Yet many remain largely reactive, mapping observations to actions without an explicit model of how these environments are structured and evolve. This motivates text world models (TWMs): transition models over textual states that, given a state and a candidate action, predict the resulting webpage, terminal output, API response, or user reply, thereby supporting planning, efficient learning, and principled evaluation. We systematically review text world models for LLM-based agents, organized around a formal framework and the agent lifecycle: (1) Foundations, defining text world models and characterizing them by state representation and grounding domain; (2) Construction, taxonomizing LLM-as-WM and code-as-WM paradigms and reviewing methods for building them; (3) Application, examining how world models support agents at training time through experience synthesis and at inference time through planning, verification, and adaptation; and (4) Evaluation, covering both evaluation of the world model itself and its use as an evaluation environment for agents. We aim to consolidate this rapidly developing area, clarify its design space, and highlight open challenges for future research.

cs.CL↗

arXiv2026

General framework for incoherent topological structured light and optical information encoding

Topology provides a powerful language for describing global invariants in physical systems, yet optical topology has been explored predominantly with fully coherent light. Recent studies have shown that incoherent light can host topological structures mediated by coherence singularities; however, a general framework for their construction and control has been lacking. Here, we introduce an incoherent Milnor polynomial, which establishes a theoretical framework for real-space incoherent topological structured light, in which topology and statistical coherence emerge as independent and jointly addressable degrees of freedom. This framework overcomes a fundamental limitation of coherent topological structured light, enabling arbitrary intensity engineering without altering the underlying topological configuration. Experimentally, we realize incoherent Hopf-linked and trefoil-knotted coherence singularities with programmable statistical coherence. We further demonstrate a robust optical information-encoding scheme inspired by Rubik's-cube-like rotations, where statistical coherence determines far-field intensity patterns associated with the cube's initial states, and topological structures govern controlled rotations acting as encryption keys. Our results advance incoherent topological structured light from a physical curiosity to a programmable photonic platform, opening new avenues for optical information encoding, statistical photonics, and coherence-engineered functionalities beyond coherent optical topology.

physics.optics↗

arXiv2026

An Explicit $O(r\log r)$ Threshold for Attaining the Semple--Steel Bound with $r$-State Characters

Let $d_r(n)$ be the maximum, over all binary phylogenetic trees with $n$ leaves, of the minimum number of $r$-state characters required to define the tree. Semple and Steel proved that $d_r(n)\geq\lceil(n-3)/(r-1)\rceil$, and Bordewich and Semple proved that equality holds for each fixed $r$ and all sufficiently large $n$. We study the corresponding threshold $n_r$, the least $N$ for which equality holds for every $n\geq N$. The Bordewich--Semple construction yields an explicit polynomial upper bound of order $O(r^5)$ for this threshold. We prove the near-linear estimate \[ 3r+1\leq n_r\leq \ceil{64(r-1)\log_2(r+1)}+3\qquad(r\geq4). \] The proof constructs, for every binary phylogenetic tree with $m=n-3$ internal edges, a linked quartet certificate whose conflict graph has maximum degree at most $16\lceil\log_2(m+2)\rceil+4$. Equitable coloring then packs the certificate into exactly $\lceil m/(r-1)\rceil$ $r$-state characters once $m\geq\lceil64(r-1)\log_2(r+1)\rceil$. We also include the lower bound $n_r\geq3r+1$, obtained from the snowflake obstruction, and state the natural conjecture that this lower bound is the exact threshold for all $r\geq4$. The conjectural endpoint is consistent with the known small-state thresholds: $n_4=13$ and $n_5=16$, while the cases $r=2,3$ are also explicitly classified.

math.CO↗

arXiv2026

Bipolar-doped superconducting infinite-layer cuprates

Distilling the intrinsic physics of the superconducting CuO2 plane from the complexities of charge-reservoir layers is a defining challenge in high-temperature superconductivity. While superconducting electron-doped infinite-layer cuprates have been synthesized, controllable and uniform hole doping has long remained elusive despite exploratory attempts, limiting spectroscopic insights. Here, we realize bipolar doping across infinite-layer (Sr,Eu)CuO2 and (Ca,Li)CuO2+δ single-crystalline thin films, mapping the electronic phase diagram. Both electron- and hole-doped films show pronounced electrical resistance anisotropy, indicating the quasi-two-dimensional nature of the CuO2 planes. Angle-resolved photoemission spectroscopy across electron- and hole-doped regimes reveals persistent antiferromagnetic band folding coexisting with superconductivity. Remarkably, at a hole doping ~0.07 determined by Luttinger volume, the antiferromagnetic folding emerges from Fermi arcs within the film's single Fermi surface, with the onset superconducting transition temperature exceeding 60 K. These findings redefine the interplay between magnetic order and superconductivity and establish a definitive platform to investigate the intrinsic mechanism of high-temperature superconducting cuprates.

cond-mat.supr-con↗

arXiv2026

Exact Leaf Powers on Cycles, Ladders, Crowns, and Multipartite Block Graphs

Exact $k$-leaf powers are graphs whose edges are exactly the pairs of leaves at distance $k$ in a tree. We prove explicit structure theorems for exact leaf powers on several representative graph families that test different exact-distance phenomena. Our most detailed root-classification theorem concerns chordless cycles: all exact $5$-leaf roots of $C_\ell$, $\ell\ge 8$, are described by a complete terminal block language. We also prove that the $t$-square ladder $L_t$ is an exact $5$-leaf power if and only if $t\le 2$. In contrast, dense bipartite square structures are often representable: among block-complete multipartite graphs, the exact $5$-leaf powers are precisely the bipartite members, and every bipartite co-cluster graph, including every crown $K_{n,n}-M$, is an exact $k$-leaf power for every $k\ge 5$. Finally, we give parity classifications for complete multipartite graphs and multipartite block graphs at larger exact distances, and isolate a sharp fan boundary at exact distance six.

math.CO↗

arXiv2026

ReGuLaR: Relation-Grounded Latent Reasoning for Large Vision-Language Models

Chain-of-thought (CoT) reasoning has significantly improved the reasoning ability of large vision-language models (LVLMs) by verbalizing intermediate reasoning steps in natural language. However, such discrete textual rationales are often insufficient for encoding continuous visual evidence. Recent work addresses this limitation by moving reasoning into continuous latent space. Despite promising progress, existing methods leave latent reasoning insufficiently connected to the compositional and relational structure of visual evidence. To address this gap, we introduce ReGuLaR, a relation grounded latent reasoning framework that explicitly grounds latent states in these critical yet overlooked visual evidence. ReGuLaR uses a training-time ReGFormer to focus latent reasoning on question-relevant objects and inter-object relations, while at inference time the model reasons and generates answers without invoking the ReGFormer. To support training ReGuLaR, we construct RGROUNDING-351K, a real-world vision-language dataset annotated with key object bounding boxes and inter-object relations. Extensive experiments across diverse benchmarks show that ReGuLaR consistently outperforms existing approaches and achieves state-of-the-art performance. We include our code in the submission and will release the code and training data publicly upon acceptance.

cs.CV↗

arXiv2026

PetroBench: A Benchmark for Large Language Models in Petroleum Engineering

Large Language Models are increasingly applied in the petroleum industry, highlighting the need for a domain-specific evaluation framework. This study develops a benchmark for LLMs in petroleum engineering, including a three-stage process of data preprocessing, quality filtering, and multi-model validation. Using expert review, a standardized question bank with strong domain relevance and discriminative capability was constructed. The benchmark covers production, reservoir, and drilling engineering, with 1,200 questions across multiple-choice, true or false, term definition, and short-answer formats. Eight mainstream LLMs were evaluated under a unified API environment. Results show that models performed better on subjective than objective questions, indicating weaknesses in factual knowledge discrimination. The highest accuracies for multiple-choice and true or false questions were 65.3% and 74.3%, respectively. Gemini-3-Pro, Kimi-K2.5, and Claude-Opus-4.6-Thinking achieved the best overall scores of 72%-74%. Models performed best in production engineering and weakest in reservoir engineering. Chinese models showed advantages in multiple-choice questions, while international models performed slightly better in short-answer questions. The benchmark provides a reproducible and practical reference for evaluating and deploying LLMs in petroleum engineering.

cs.AI↗

arXiv2026

Diffuse to Detect: Generative Diffusion Models for Unsupervised IC Anomaly Detection

Latent defect screening is challenged by extremely low failure rates, high-dimensional test data, and absence of labeled anomalies. We propose the first unsupervised anomaly detection framework incorporating a Diffusion Transformer. Raw test measurements are first compressed by an autoencoder, then reshaped into a structured token sequence enriched with sinusoidal and per-device wafer-position embeddings. Anomaly scores are derived from the noise-prediction error over mid-range diffusion timesteps, enabling fast wafer-scale screening without any labeled defects or manual feature engineering. Our approach achieves state-of-the-art performance on industrial 16nm IC test data under extreme class imbalance, offering interpretable failure localization through latent-space reconstruction residuals.

cs.LG↗

arXiv2026

A High-Performance, Cross-Platform Open-Source Solver for the Incompressible Navier-Stokes Equations in FEALPy

To address the dual challenges of performance portability across heterogeneous hardware and the high usability barriers of conventional computational fluid dynamics (CFD) software, this paper introduces FEALPy.CFD, a high performance, open-source solver for the incompressible Navier-Stokes equations developed within the FEALPy framework. The solver's core innovation is its backend-agnostic design, which supports multiple computational backends like NumPy, PyTorch, and JAX to enable seamless switching between CPU and GPU computations with minimal code modification, thereby maximizing hardware utilization and code portability. Its highly modular architecture provides a library of composable components for various spatial discretization schemes, greatly simplifying the development and integration of new algorithms.Validation on benchmark cases confirms that the implemented numerical schemes achieve their theoretical orders of convergence. Furthermore, the capability to select a suitable backend architecture for different computational tasks fully leverages the hardware's potential, delivering substantial efficiency gains. By lowering the technical barrier to high-performance, cross-platform fluid dynamics simulation, FEALPy.CFD offers a powerful and accessible tool for academic research, engineering applications, and reproducible computational science.

physics.flu-dyn↗

arXiv2026

GraphFlow: A Graph-Based Workflow Management for Efficient LLM-Agent Serving

Large Language Model (LLM)-based agents demonstrate strong reasoning and execution capabilities on complex tasks when guided by structured instructions, commonly referred to as workflows. However, existing workflow-assisted agent serving systems typically rely on predefined templates and shallow matching mechanisms, which limit their ability to capture deep semantic relationships and generalize to previously unseen tasks. To address these limitations, we propose a new workflow management paradigm that represents workflows using a unified graph, termed wGraph, where each node corresponds to an atomic operation. wGraph serves as a shared substrate from which task-specific workflows are dynamically instantiated. Building on wGraph primitives, we introduce GraphFlow, a system that efficiently integrates workflows into agent serving through two key designs. First, adaptive workflow generation dynamically constructs workflows from wGraph based on task semantics and constraint requirements. Second, workflow state management exploits wGraph structure to efficiently manage Key-Value (KV) caches, reducing redundant computation during agent serving. Extensive experiments across five benchmark datasets show that GraphFlow consistently outperforms state-of-the-art methods, yielding an average performance improvement of approximately 4.95 percentage points, while achieving an approximately 4$\times$ reduction in memory footprint.

cs.LG↗