Search arXiv⌕ Search

arXiv subjects

Yanjun Li

Publications and source records attributed to Yanjun Li.

At least 19 recordsLinked to original sources
arXiv2026

RFPO: Rectified Flow Policy Optimization for Embodied Control

Flow-based policies provide an expressive framework for continuous robot control, but their iterative ODE integration incurs substantial inference cost. Naively reducing the integration budget can severely degrade control, since policies optimized under full-step execution are not explicitly constrained to remain reliable under coarse numerical integration. We refer to this mismatch as the few-step discretization gap. To address this problem, we introduce RFPO, a flow-policy optimization framework for reliable few-step execution. Reward-aware online Reflow rectifies student-induced transport paths during on-policy learning, making the resulting policy more robust to coarse integration. A frozen Gaussian PPO controller supplies complementary action-space supervision at full and intermediate integration budgets, while the deployed policy remains a single flow student executed with one Euler step. Across Unitree Go2, Boston Dynamics Spot, Unitree H1, and Unitree G1, RFPO consistently preserves full-step control performance under one-step execution, with one-step returns remaining within 2.4% of their corresponding 64-step values across both zero and random initialization. On Unitree Go2, one-step execution retains 98.5% of the 64-step reward while reducing onboard mean inference latency from 4.39 ms to 0.08 ms, yielding a 54.9x speedup. Real-robot experiments further validate stable one-step locomotion. Code: https://github.com/AIGeeksGroup/RFPO. Website: https://aigeeksgroup.github.io/RFPO.

cs.RO↗
arXiv2026

DexPolicy: Scheduled Exploration for Trajectory-Guided Dexterous Manipulation

Reinforcement learning (RL) for dexterous manipulation must discover finger-object contacts and then control the object precisely; the action noise that serves the first goal can interfere with the second. In trajectory-guided settings such as ViViDex, where RL refine hand-object trajectories from human video, our baseline PPO runs end near their initial action noise after 5M steps, motivating explicit control of exploration scale. DexPolicy makes that scale an explicit function of training steps, annealing from broad to narrow exploration while holding loss, architecture, reward, and optimizer settings fixed. We study three policy-optimization settings: PPO, critic-free GRPO continuation, and a flow-parameterized PPO variant (FPO). Across five YCB objects and three training seeds, mean deterministic Target success rises from 49.4% to 68.1% (FPO), 14.1% to 45.4% (GRPO), and 32.0% to 35.7% (PPO). On a RealMan RM75 arm with an Inspire/RH56 hand, 360 trials over three objects raise mean Target success from 25.0% to 85.0% (FPO), 10.0% to 63.3% (GRPO), and 8.3% to 43.3% (PPO), with one trained model per object-method condition. PPO component screening favors noise control over the tested optimizer contraction; the selected PPO schedule yields higher mean Target success than linear decay with the same endpoints on three tested objects. Training return, deterministic Target success, and tolerance to execution noise dissociate; schedules should therefore be judged by terminal task success under the intended execution conditions, per task and policy-optimization setting. Code: https://github.com/AIGeeksGroup/DexPolicy. Website: https://aigeeksgroup.github.io/DexPolicy.

cs.RO↗
arXiv2026

Language-Forcing: Advancing Humanoid Loco-Manipulation via Semantic Alignment of Whole-Body World Action Models

World Action Models (WAMs) jointly predict robot actions and future observations. However, their visual prediction objectives do not explicitly distinguish task-relevant transitions from incidental scene changes. For whole-body policies, this distinction matters because task progress depends on coordinated changes in body pose, contact, and object state. We introduce LANGUAGE-FORCING, a training strategy that uses language to emphasize task-relevant changes in the future visual representations of a whole-body WAM. Given paired current and ground- truth short-horizon future frames, a frozen vision-language model assesses task progress and predicts plausible subsequent state changes. We encode these descriptions with a frozen text encoder and align their embeddings with intermediate, action-conditioned future-visual features, complementing the original visual and action prediction objectives. All auxiliary components are used only during training, introducing no additional inference overhead. On HumanoidArena, LANGUAGE-FORCING achieves an average success rate of 79.17%, exceeding the strongest competing policy by 9.46 percentage points and the same WAM without language alignment by 10.77 percentage points. Across five real-world tasks, language alignment raises average success from 53.33% to 71.33% over the same WAM baseline.

cs.RO↗
arXiv2026

HumanoidTTT: Test-Time Capability Reuse for Efficient Humanoid Control

Recent advances in motion generation and whole-body tracking have enabled humanoid robots to execute increasingly diverse motions, yet the same motion capabilities may be requested repeatedly during continual deployment. Reliable reuse is challenging because intervening motions can change the robot's entry state, making previously successful motions unsafe to replay blindly. Meanwhile, validated capabilities accumulate during deployment, while bounded storage requires deciding which ones are worth retaining. To address these challenges, we present HumanoidTTT, a framework for test-time capability reuse in continual humanoid control. Specifically, we introduce Selective Full-Motion Reuse, which authorizes direct reuse of validated complete motions only from certified applicable entry states, allowing accepted reuse to bypass fresh generation. We further introduce Test-Time Capability Consolidation, which adapts which qualified capabilities persist in a bounded Full-Motion Store using subsequent deployment reuse as feedback. Experiments demonstrate zero unsafe accepts and a 16.4$\times$ end-to-end speedup over fresh generation, while online consolidation improves avoided generator calls by 13.2 per 200 requests over its frozen counterpart. Overall, HumanoidTTT enables reliable and efficient reuse of validated motion capabilities while adaptively retaining useful capabilities throughout continual deployment. Code: https://github.com/AIGeeksGroup/HumanoidTTT. Website: https://aigeeksgroup.github.io/HumanoidTTT.

cs.RO↗
arXiv2026

Characterization of a class of complete permutation quadrinomials over $\mathbb{F}_{2^{2m}}$

Let $q=2^m$, $Q=2^k$, and $1\leq k\leq m-1$. We characterize complete permutation polynomials (CPPs) over $\mathbb{F}_{q^2}$ of the form \[ f(x)=c_0x^{Q+1}+c_1x^{Q+q}+c_2x^{qQ+1} +c_3x^{q(Q+1)},\qquad c_i\in\mathbb{F}_{q^2}. \] We prove that no such CPP exists when $k>1$, and recover the known characterization in the cubic case $k=1$. This completes the classification throughout the stated exponent range. The proof uses the known permutation classification to reduce completeness to linear perturbations of product and monomial models. The required nonpermutation results follow from direct elementary arguments based on the quadratic structure over $\mathbb{F}_2$.

math.NT↗
arXiv2026

Elementary magnons and interacting multi-magnon quasiparticles in the effective spin-$\frac{1}{2}$ kagome-staircase magnet Co$_{3}$V$_{2}$O$_{8}$

The excitation spectrum of an anisotropic magnet provides a direct link between its microscopic Hamiltonian and interaction-driven quasiparticles. Here we use high-resolution time-domain terahertz spectroscopy to map the magnetic excitations of the three-dimensional kagome-staircase compound Co$_{3}$V$_{2}$O$_{8}$ as functions of temperature and magnetic field. At low energies, polarization-resolved spectra identify magnetic-dipole-active one-magnon modes and track their evolution across the ferromagnetic and spin-density-wave phases. Combining their field dependence with previously reported inelastic-neutron-scattering dispersions, we determine an effective spin-$\frac{1}{2}$ Hamiltonian with strongly anisotropic exchange that quantitatively reproduces the one-magnon spectrum. This model provides a noninteracting benchmark for the high-energy response, where we observe sharp branches with field slopes that are two to four times those of the one-magnon modes, together with anticrossings between branches of different magnon numbers. Their sharpness, polarization dependence, and departure from the calculated multi-magnon continua identify them as interacting multi-magnon quasiparticles that can be stabilized by strong exchange anisotropy.

cond-mat.str-el↗
arXiv2026

The First EgoCross Challenge at EgoVis 2026: Cross-Domain Egocentric Video Question Answering

EgoCross is a cross-domain egocentric video question answering benchmark designed to evaluate whether multimodal large language models can generalize beyond common daily-life scenarios. The first EgoCross Challenge was hosted at the Third EgoVis Workshop at CVPR 2026 and evaluated models on first-person videos from four target domains: surgery, industrial assembly, extreme sports, and animal perspectives. Each test example consists of an egocentric video clip, a question, and four candidate answers, from which the model must select the correct option. This technical report introduces the challenge task, benchmark resources, and two official Codabench tracks. The Source-Limited Track restricts participants to the official baseline model and a small support set, whereas the Open-Source Track permits broader choices of models and training data under rules that prohibit the manual construction of target-domain training data. In total, the challenge received more than 1,500 submissions from over 130 participants, with 19 teams participating in the Open-Source Track and 38 teams in the Source-Limited Track. We further present the official leaderboard results and summarize the winning solutions from both tracks. We hope that this report will serve as a useful technical reference for advancing cross-domain egocentric video understanding. All resources, including the challenge data, baseline implementation, and code released by the winning teams, are made publicly available.

cs.CV↗
arXiv2026

Spectral Heat Flow for Conservative Token Condensation in Vision-Language Models

Vision-Language Models (VLMs) are costly at inference time because they must process long sequences of visual tokens. Existing token pruning methods often degrade under high compression by blindly discarding information, breaking spatial structure or collapsing diversity. We propose SpecFlow, a training-free framework that shifts the paradigm from destructive pruning to conservative condensation, strictly enforcing spatial coverage and statistical conservation to ensure stability. Treating visual tokens as nodes in a $k$NN graph, SpecFlow (i) computes a stable importance field via spectral heat flow to preserve structural coherence, (ii) allocates budgets via adaptive spatial partitioning to guarantee coverage, and (iii) aggregates discarded information into coreset sinks to maintain statistical conservation. The method is plug-and-play, requires no fine-tuning, and is compatible with FlashAttention. Experiments confirm that our SpecFlow outperforms SOTA methods across tasks, VLM architectures, and pruning ratios. Notably, LLaVA-1.5 with SpecFlow retains 95.6% of original performance despite pruning 88.9% of visual tokens, offering an exceptional efficiency-accuracy balance. Code is available at https://github.com/Lzy-dot/SpecFlow

cs.CV↗
arXiv2026

Learning Mechanistic Reasoning for Chemical Reactions with Large Language Models

Reaction mechanisms consist of the step-by-step sequences of elementary reactions that explain chemical transformations. Learning the mechanism logic is therefore essential for enhancing the fundamental chemical intelligence of large language models (LLMs). The stepwise deduction of reaction mechanism aligns naturally with the reasoning paradigms of reasoning LLMs. However, current chemical LLMs primarily emphasize coarse-grained name reactions for product prediction and retrosynthesis, often leading to physical inconsistencies and hallucinations. In contrast, specialized small-scale generative models for mechanism inference typically suffer from restricted generalization capacity across diverse chemical spaces. To overcome these limitations, we built a novel, large-scale reasoning dataset of reaction mechanisms. Furthermore, we established the FukuyamaBench, a difficult benchmark derived from Fukuyama's Advanced Organic Reaction Mechanism book, to rigorously evaluate model performance on hierarchical mechanism reasoning. Our fine-tuned Qwen3-30B-A3B achieves 8.3% exact pathway match on FukuyamaBench Set~A, surpassing the specialized FlowER model (5.1%), demonstrating that mechanism-aware training substantially enhances chemical reasoning in language models.

cs.LG↗
arXiv2026

DragMesh-2: Physically Plausible Dexterous Hand-Object Interaction with Articulated Objects

Dexterous interaction with articulated objects is important for household, assistive, and humanoid manipulation, where multi-finger hands can provide compliant contact patterns beyond parallel-jaw grasping. However, articulated-object manipulation differs from static-object manipulation: the target part cannot be directly actuated, and its motion must emerge through sustained physical hand--handle contact. This makes the transition from object-centric articulated generation to hand-driven dexterous hand--object interaction non-trivial, since geometric trajectory replay or open-loop execution does not model the contact dynamics required to move the articulated part. Moreover, policies trained only for task completion under fixed dynamics can overfit nominal contact loads, especially without tactile or force feedback, and may degrade when the contact load changes. To address these challenges, we present DragMesh-2, a contact-driven framework for dexterous interaction with articulated objects that extends articulated interaction from object-centric generation to hand-driven dexterous hand--object interaction, where articulated motion must arise through physical contact. We further propose PICA, a physically informed contact-aware training mechanism that injects physical signals into policy learning without tactile or force feedback, improving robustness and task success under changing contact loads. Finally, we conduct systematic evaluation across multiple damping conditions and articulated-object categories to study robustness under contact-load variation, and provide a pure-geometry dexterous interaction resource to support future loco-manipulation and humanoid hand--object interaction research. Across seven GAPartNet objects, DragMesh-2 achieves stronger robustness under contact-load variation than the compared methods while maintaining high task success across damping conditions.

cs.RO↗
arXiv2026

MotionVLA: Vision-Language-Action Model for Humanoid Motion

Generating realistic humanoid motion from scene images and text involves both low-frequency pose semantics and high-frequency physical dynamics. However, many existing methods tokenize motion with a single shared codebook, forcing heterogeneous motion signals into the same quantization space. Our frequency-domain analysis of human motion data reveals a clear mismatch between single-codebook quantization and motion statistics: five DCT coefficients capture 93% of joint-position energy but only 37% of joint-velocity energy, which can bias quantization toward pose statistics and under-represent high-frequency velocity components. A second challenge lies in adapting a standard autoregressive model to effectively model high-frequency physical signals in motion sequences. Therefore, we propose DSFT, a dual-stream frequency tokenizer that separates motion into Base and physical streams and compresses them independently with DCT truncation and BPE. Furthermore, we present MotionVLA, a Qwen3.5-based model that arranges Base and physical tokens in a unified sequence, where Phys tokens are predicted after Base tokens. Experiments on HumanML3D and MBench show that, despite using a lightweight 2B backbone, MotionVLA reduces the Diversity gap to real data by over 50% on HumanML3D and improves Motion-Condition Consistency by 3.8% on MBench, supporting frequency-aware dual-stream decoupling as an effective formulation for autoregressive motion generation. Code: https://github.com/AIGeeksGroup/MotionVLA. Website: https://aigeeksgroup.github.io/MotionVLA.

cs.CV↗
arXiv2026

Weyl points enabling significant enhancement of thermoelectric performance in an antiferromagnetic van der Waals metal GdTe3

Magneto-thermoelectric (MTE) effect has demonstrated significant ad-vantages in achieving optimal thermoelectric (TE) properties compared to conventional methods. Topological materials pro-vide a unique platform for investigating the MTE effect, leveraging their exotic electronic structure topology. In this study, we report that the topological material GdTe3 exhibits an unsaturated power factor of up to 18846 μW m-1 K-1 under a magnetic field of 13.5 T at 20 K, which represents the highest value observed in metallic systems and surpasses most state-of-the-art TE materials. The relative enhancement under magnetic field in thermopower and power factor reaches 873% and 1075%, respectively, attributed to the Weyl points contribution resulting from the field-induced topological transition, as confirmed by our theoretical calculations. Our findings demonstrate a promising candidate for solid-state cooling and reveal the substantial contribution of Weyl points to TE enhancement, thereby offering a novel approach to optimizing TE properties in topological materials.

cond-mat.mtrl-sci↗
arXiv2026

EgoCross: Benchmarking Multimodal Large Language Models for Cross-Domain Egocentric Video Question Answering

Recent advances in Multimodal Large Language Models (MLLMs) have significantly pushed the frontier of egocentric video question answering (EgocentricQA). However, existing benchmarks and studies are mainly limited to common daily activities such as cooking and cleaning. In contrast, real-world deployment inevitably encounters domain shifts, where target domains differ substantially in both visual style and semantic content. To bridge this gap, we introduce \textbf{EgoCross}, a comprehensive benchmark designed to evaluate the cross-domain generalization of MLLMs in EgocentricQA. EgoCross covers four diverse and challenging domains, including surgery, industry, extreme sports, and animal perspective, representing realistic and high-impact application scenarios. It comprises approximately 1,000 QA pairs across 798 video clips, spanning four key QA tasks: prediction, recognition, localization, and counting. Each QA pair provides both OpenQA and CloseQA formats to support fine-grained evaluation. Extensive experiments show that most existing MLLMs, whether general-purpose or egocentric-specialized, struggle to generalize to domains beyond daily life, highlighting the limitations of current models. Furthermore, we conduct several pilot studies, e.g., fine-tuning and reinforcement learning, to explore potential improvements. We hope EgoCross and our accompanying analysis will serve as a foundation for advancing domain-adaptive, robust egocentric video understanding.

cs.CV↗
arXiv2026

AutoVulnPHP: LLM-Powered Two-Stage PHP Vulnerability Detection and Automated Localization

PHP's dominance in web development is undermined by security challenges: static analysis lacks semantic depth, causing high false positives; dynamic analysis is computationally expensive; and automated vulnerability localization suffers from coarse granularity and imprecise context. Additionally, the absence of large-scale PHP vulnerability datasets and fragmented toolchains hinder real-world deployment. We present AutoVulnPHP, an end-to-end framework coupling two-stage vulnerability detection with fine-grained automated localization. SIFT-VulMiner (Structural Inference for Flaw Triage Vulnerability Miner) generates vulnerability hypotheses using AST structures enhanced with data flow. SAFE-VulMiner (Semantic Analysis for Flaw Evaluation Vulnerability Miner) verifies candidates through pretrained code encoder embeddings, eliminating false positives. ISAL (Incremental Sequence Analysis for Localization) pinpoints root causes via syntax-guided tracing, chain-of-thought LLM inference, and causal consistency checks to ensure precision. We contribute PHPVD, the first large-scale PHP vulnerability dataset with 26,614 files (5.2M LOC) across seven vulnerability types. On public benchmarks and PHPVD, AutoVulnPHP achieves 99.7% detection accuracy, 99.5% F1 score, and 81.0% localization rate. Deployed on real-world repositories, it discovered 429 previously unknown vulnerabilities, 351 assigned CVE identifiers, validating its practical effectiveness.

cs.CR↗
arXiv2025

Apo2Mol: 3D Molecule Generation via Dynamic Pocket-Aware Diffusion Models

Deep generative models are rapidly advancing structure-based drug design, offering substantial promise for generating small molecule ligands that bind to specific protein targets. However, most current approaches assume a rigid protein binding pocket, neglecting the intrinsic flexibility of proteins and the conformational rearrangements induced by ligand binding, limiting their applicability in practical drug discovery. Here, we propose Apo2Mol, a diffusion-based generative framework for 3D molecule design that explicitly accounts for conformational flexibility in protein binding pockets. To support this, we curate a dataset of over 24,000 experimentally resolved apo-holo structure pairs from the Protein Data Bank, enabling the characterization of protein structure changes associated with ligand binding. Apo2Mol employs a full-atom hierarchical graph-based diffusion model that simultaneously generates 3D ligand molecules and their corresponding holo pocket conformations from input apo states. Empirical studies demonstrate that Apo2Mol can achieve state-of-the-art performance in generating high-affinity ligands and accurately capture realistic protein pocket conformational changes.

q-bio.BM↗
arXiv2025

Unveiling the delicate "hidden" interface conditions in WS2 flakes by advanced atomic force microscopy

The delicate interfacial conditions and behaviors play critical roles in determining the valuable physical properties of two-dimensional materials and their heterostructures on substrates. However, directly probing these complex interface conditions remains challenging. Here, we reveal the coupled in-plane strain and out-of-plane bonding conditions in strain-engineered WS2 flakes by combining dual-harmonic electrostatic force microscopy (DH-EFM) and scanning microwave impedance microscopy (sMIM). A striking contradiction is observed between the compressive-strain-induced larger bandgap (lower electrical conductivity) detected by DH-EFM, and the enhanced conductivity probed by sMIM. Comparative measurements under different sMIM modes demonstrate that this contradiction originates from a tip-loading-force-induced dynamic puckering effect, which is governed by the interfacial bonding strength. Furthermore, the progressive accumulation and subsequent release of conductivity during forward/backward sMIM-contact scans further confirms this dynamic puckering behavior, revealing pronounced differences in interface conditions between the open- and closed-ring regions of WS2. This work resolves the correlation between electrical properties and interface conditions, and provides fundamental insights for interface-engineered devices.

cond-mat.mtrl-sci↗
arXiv2025

HSACNet: Hierarchical Scale-Aware Consistency Regularized Semi-Supervised Change Detection

Semi-supervised change detection (SSCD) aims to detect changes between bi-temporal remote sensing images by utilizing limited labeled data and abundant unlabeled data. Existing methods struggle in complex scenarios, exhibiting poor performance when confronted with noisy data. They typically neglect intra-layer multi-scale features while emphasizing inter-layer fusion, harming the integrity of change objects with different scales. In this paper, we propose HSACNet, a Hierarchical Scale-Aware Consistency regularized Network for SSCD. Specifically, we integrate Segment Anything Model 2 (SAM2), using its Hiera backbone as the encoder to extract inter-layer multi-scale features and applying adapters for parameter-efficient fine-tuning. Moreover, we design a Scale-Aware Differential Attention Module (SADAM) that can precisely capture intra-layer multi-scale change features and suppress noise. Additionally, a dual-augmentation consistency regularization strategy is adopted to effectively utilize the unlabeled data. Extensive experiments across four CD benchmarks demonstrate that our HSACNet achieves state-of-the-art performance, with reduced parameters and computational cost.

cs.CV↗
arXiv2025

Large Language Model for Extracting Complex Contract Information in Industrial Scenes

This paper proposes a high-quality dataset construction method for complex contract information extraction tasks in industrial scenarios and fine-tunes a large language model based on this dataset. Firstly, cluster analysis is performed on industrial contract texts, and GPT-4 and GPT-3.5 are used to extract key information from the original contract data, obtaining high-quality data annotations. Secondly, data augmentation is achieved by constructing new texts, and GPT-3.5 generates unstructured contract texts from randomly combined keywords, improving model robustness. Finally, the large language model is fine-tuned based on the high-quality dataset. Experimental results show that the model achieves excellent overall performance while ensuring high field recall and precision and considering parsing efficiency. LoRA, data balancing, and data augmentation effectively enhance model accuracy and robustness. The proposed method provides a novel and efficient solution for industrial contract information extraction tasks.

cs.CL↗