Yanjie Li

Wenqiang Li, Weijun Li, Lina Yu et al.

Symbolic regression (SR) is a powerful technique for discovering the underlying mathematical expressions from observed data. Inspired by the success of deep learning, recent deep generative SR methods have shown promising results. However, these methods face difficulties in processing high-dimensional problems and learning constants due to the large search space, and they don't scale well to unseen problems. In this work, we propose DySymNet, a novel neural-guided Dynamic Symbolic Network for SR. Instead of searching for expressions within a large search space, we explore symbolic networks with various structures, guided by reinforcement learning, and optimize them to identify expressions that better-fitting the data. Based on extensive numerical experiments on low-dimensional public standard benchmarks and the well-known SRBench with more variables, DySymNet shows clear superiority over several representative baseline models. Open source code is available at https://github.com/AILWQ/DySymNet.

Yueyang Weng, Xiaopeng Zhang, Yongjin Mu et al.

Action chunking is a widely adopted approach in Learning from Demonstration (LfD). By modeling multi-step action chunks rather than single-step actions, action chunking significantly enhances modeling capabilities for human expert policies. However, because action chunking makes a single decision only after a complete action block has been executed, the resulting reduction in decision frequency restricts the utilization of real-time observations, impairing reactivity in dynamic or noisy environments. Existing efforts to address this issue have primarily resorted to trading off reactivity against decision consistency, without achieving both. To address this limitation, we propose a novel algorithm, Temporal Action Selection (TAS), which caches predicted action chunks from multiple timesteps and dynamically selects the optimal action through a lightweight selector network. TAS achieves balanced optimization across both reactivity and decision consistency. Experiments across multiple tasks with diverse base policy architectures show that TAS significantly improves success rates. Furthermore, integrating TAS as a base policy with residual reinforcement learning (RL) improves both training efficiency and the performance ceiling. Experiments in both simulation and physical robots confirm the method's efficacy.

Meituan LongCat Team, Bin Xiao, Chao Wang et al.

The prevailing Next-Token Prediction (NTP) paradigm has driven the success of large language models through discrete autoregressive modeling. However, contemporary multimodal systems remain language-centric, often treating non-linguistic modalities as external attachments, leading to fragmented architectures and suboptimal integration. To transcend this limitation, we introduce Discrete Native Autoregressive (DiNA), a unified framework that represents multimodal information within a shared discrete space, enabling a consistent and principled autoregressive modeling across modalities. A key innovation is the Discrete Native Any-resolution Visual Transformer (dNaViT), which performs tokenization and de-tokenization at arbitrary resolutions, transforming continuous visual signals into hierarchical discrete tokens. Building on this foundation, we develop LongCat-Next, a native multimodal model that processes text, vision, and audio under a single autoregressive objective with minimal modality-specific design. As an industrial-strength foundation model, it excels at seeing, painting, and talking within a single framework, achieving strong performance across a wide range of multimodal benchmarks. In particular, LongCat-Next addresses the long-standing performance ceiling of discrete vision modeling on understanding tasks and provides a unified approach to effectively reconcile the conflict between understanding and generation. As an attempt toward native multimodality, we open-source the LongCat-Next and its tokenizers, hoping to foster further research and development in the community. GitHub: https://github.com/meituan-longcat/LongCat-Next

Yanjie Li, Sihong Shao

This paper studies the Fourier spectral method (FSM) for the Schrödinger equation with singular potentials $V \in H^{s}$, where $s > \max\{d/2-2,-1\}$ and $d$ denotes the spatial dimension. This setting includes a broad class of singular potentials, such as the 3D Coulomb potential and the 1D Dirac-delta potential. First, we combine the Feshbach-Schur map with a refined perturbation argument to derive sharp convergence orders for FSM, yielding order $2s+2$ for eigenvalues and order $s+1$ for eigenfunctions in the $H^1$ norm. More importantly, the $H^1$ error with respect to the projected eigenfunction converges with a higher order $s+1+b$, where $b=\min\{s+2-d/2-\varepsilon,\; s+1,\; 2\}>0$ for arbitrarily small $\varepsilon>0$, revealing a super-convergence phenomenon. Second, in the presence of potentials with isolated point singularities, we develop an asymptotic-recovery (AR) technique to post-process the FSM solutions. The resulting method, dubbed AR-FSM, fully exploits the super-convergence property and achieves convergence orders $2s+2+2b$ for eigenvalues and $s+1+b$ for eigenfunctions in the $H^1$ norm, while the AR post-processing requires only a computational cost that is linear in the number of FSM degrees of freedom. The analysis introduces a rigorous definition of point singularities and develops a foundational framework for their study. It further establishes an asymptotic expansion of eigenfunctions consisting of a regular component in $H^{s+4}$ together with $d+1$ asymptotic functions associated with each singular point. Numerical experiments confirm the sharpness of these theoretical bounds.

Yibo Yang, Fei Lei, Yixuan Sun et al.

As AI-driven document understanding and processing tools become increasingly prevalent in real-world applications, the need for rigorous evaluation standards has grown increasingly urgent. Existing benchmarks and evaluations often focus on isolated capabilities or simplified scenarios, failing to capture the end-to-end task effectiveness required in practical settings. To address this gap, we introduce AIDABench, a comprehensive benchmark for evaluating AI systems on complex data analytics tasks in an end-to-end manner. AIDABench encompasses 600+ diverse document analysis tasks across three core capability dimensions: question answering, data visualization, and file generation. These tasks are grounded in realistic scenarios involving heterogeneous data types, including spreadsheets, databases, financial reports, and operational records, and reflect analytical demands across diverse industries and job functions. Notably, the tasks in AIDABench are sufficiently challenging that even human experts require 1-2 hours per question when assisted by AI tools, underscoring the benchmark's difficulty and real-world complexity. We evaluate 11 state-of-the-art models on AIDABench, spanning both proprietary (e.g., Claude Sonnet 4.5, Gemini 3 Pro Preview) and open-source (e.g., Qwen3-Max-2026-01-23-Thinking) families. Our results reveal that complex, real-world data analytics tasks remain a significant challenge for current AI systems, with the best-performing model achieving only 59.43% pass-at-1. We provide a detailed analysis of failure modes across each capability dimension and identify key challenges for future research. AIDABench offers a principled reference for enterprise procurement, tool selection, and model optimization, and is publicly available at https://github.com/MichaelYang-lyx/AIDABench.

Yanjie Li, Bin Xie, Songtao Guo et al.

Benefiting from the rapid development of deep learning, 2D and 3D computer vision applications are deployed in many safe-critical systems, such as autopilot and identity authentication. However, deep learning models are not trustworthy enough because of their limited robustness against adversarial attacks. The physically realizable adversarial attacks further pose fatal threats to the application and human safety. Lots of papers have emerged to investigate the robustness and safety of deep learning models against adversarial attacks. To lead to trustworthy AI, we first construct a general threat model from different perspectives and then comprehensively review the latest progress of both 2D and 3D adversarial attacks. We extend the concept of adversarial examples beyond imperceptive perturbations and collate over 170 papers to give an overview of deep learning model robustness against various adversarial attacks. To the best of our knowledge, we are the first to systematically investigate adversarial attacks for 3D models, a flourishing field applied to many real-world applications. In addition, we examine physical adversarial attacks that lead to safety violations. Last but not least, we summarize present popular topics, give insights on challenges, and shed light on future research on trustworthy AI.

Yanjie Li, Yiquan Li, Xuelong Dai et al.

2D face recognition has been proven insecure for physical adversarial attacks. However, few studies have investigated the possibility of attacking real-world 3D face recognition systems. 3D-printed attacks recently proposed cannot generate adversarial points in the air. In this paper, we attack 3D face recognition systems through elaborate optical noises. We took structured light 3D scanners as our attack target. End-to-end attack algorithms are designed to generate adversarial illumination for 3D faces through the inherent or an additional projector to produce adversarial points at arbitrary positions. Nevertheless, face reflectance is a complex procedure because the skin is translucent. To involve this projection-and-capture procedure in optimization loops, we model it by Lambertian rendering model and use SfSNet to estimate the albedo. Moreover, to improve the resistance to distance and angle changes while maintaining the perturbation unnoticeable, a 3D transform invariant loss and two kinds of sensitivity maps are introduced. Experiments are conducted in both simulated and physical worlds. We successfully attacked point-cloud-based and depth-image-based 3D face recognition algorithms while needing fewer perturbations than previous state-of-the-art physical-world 3D adversarial attacks.

Yanjie Li, Weijun Li, Lina Yu et al.

Mathematical formulas serve as the means of communication between humans and nature, encapsulating the operational laws governing natural phenomena. The concise formulation of these laws is a crucial objective in scientific research and an important challenge for artificial intelligence (AI). While traditional artificial neural networks (MLP) excel at data fitting, they often yield uninterpretable black box results that hinder our understanding of the relationship between variables x and predicted values y. Moreover, the fixed network architecture in MLP often gives rise to redundancy in both network structure and parameters. To address these issues, we propose MetaSymNet, a novel neural network that dynamically adjusts its structure in real-time, allowing for both expansion and contraction. This adaptive network employs the PANGU meta function as its activation function, which is a unique type capable of evolving into various basic functions during training to compose mathematical formulas tailored to specific needs. We then evolve the neural network into a concise, interpretable mathematical expression. To evaluate MetaSymNet's performance, we compare it with four state-of-the-art symbolic regression algorithms across more than 10 public datasets comprising 222 formulas. Our experimental results demonstrate that our algorithm outperforms others consistently regardless of noise presence or absence. Furthermore, we assess MetaSymNet against MLP and SVM regarding their fitting ability and extrapolation capability, these are two essential aspects of machine learning algorithms. The findings reveal that our algorithm excels in both areas. Finally, we compared MetaSymNet with MLP using iterative pruning in network structure complexity. The results show that MetaSymNet's network structure complexity is obviously less than MLP under the same goodness of fit.

Yanjie Li, Mingxing Duan, Xuelong Dai et al.

Adversarial patch attacks can fool the face recognition (FR) models via small patches. However, previous adversarial patch attacks often result in unnatural patterns that are easily noticeable. Generating transferable and stealthy adversarial patches that can efficiently deceive the black-box FR models while having good camouflage is challenging because of the huge stylistic difference between the source and target images. To generate transferable, natural-looking, and stealthy adversarial patches, we propose an innovative two-stage attack called Adv-Inpainting, which extracts style features and identity features from the attacker and target faces, respectively and then fills the patches with misleading and inconspicuous content guided by attention maps. In the first stage, we extract multi-scale style embeddings by a pyramid-like network and identity embeddings by a pretrained FR model and propose a novel Attention-guided Adaptive Instance Normalization layer (AAIN) to merge them via background-patch cross-attention maps. The proposed layer can adaptively fuse identity and style embeddings by fully exploiting priority contextual information. In the second stage, we design an Adversarial Patch Refinement Network (APR-Net) with a novel boundary variance loss, a spatial discounted reconstruction loss, and a perceptual loss to boost the stealthiness further. Experiments demonstrate that our attack can generate adversarial patches with improved visual quality, better stealthiness, and stronger transferability than state-of-the-art adversarial patch attacks and semantic attacks.

Yanjie Li, Jian Xu, Xueqing Chen et al.

Adsorption energy is a key descriptor of catalytic reactivity. It is fundamentally defined as the difference between the relaxed total energy of the adsorbate-surface system and that of an appropriate reference state; therefore, the accuracy of relaxed-energy prediction directly determines the reliability of machine-learning-driven catalyst screening. E(3)-equivariant graph neural networks (GNNs) can natively operate on three-dimensional atomic coordinates under periodic boundary conditions and have demonstrated strong performance on such tasks. In contrast, language-model-based approaches, while enabling human-readable textual descriptions and reducing reliance on explicit graph -- thereby broadening applicability -- remain insufficient in both adsorption-configuration energy prediction accuracy and in distinguishing ``the same system with different configurations,'' even with graph-assisted pretraining in the style of GAP-CATBERTa. To this end, we propose QE-Catalytic, a multimodal framework that deeply couples a large language model (\textbf{Q}wen) with an E(3)-equivariant graph Transformer (\textbf{E}quiformer-V2), enabling unified support for adsorption-configuration property prediction and inverse design on complex catalytic surfaces. During prediction, QE-Catalytic jointly leverages three-dimensional structures and structured configuration text, and injects ``3D geometric information'' into the language channel via graph-text alignment, allowing it to function as a high-performance text-based predictor when precise coordinates are unavailable, while also autoregressively generating CIF files for target-energy-driven structure design and information completion. On OC20, QE-Catalytic reduces the MAE of relaxed adsorption energy from 0.713~eV to 0.486~eV, and consistently outperforms baseline models such as CatBERTa and GAP-CATBERTa across multiple evaluation protocols.

Qi Liu, Jianqi Gao, Dongjie Zhu et al.

Multi-agent target assignment and path planning (TAPF) are two key problems in intelligent warehouse. However, most literature only addresses one of these two problems separately. In this study, we propose a method to simultaneously solve target assignment and path planning from a perspective of cooperative multi-agent deep reinforcement learning (RL). To the best of our knowledge, this is the first work to model the TAPF problem for intelligent warehouse to cooperative multi-agent deep RL, and the first to simultaneously address TAPF based on multi-agent deep RL. Furthermore, previous literature rarely considers the physical dynamics of agents. In this study, the physical dynamics of the agents is considered. Experimental results show that our method performs well in various task settings, which means that the target assignment is solved reasonably well and the planned path is almost shortest. Moreover, our method is more time-efficient than baselines.

Yanjie Li, Jingyi Liu, Weijun Li et al.

Mathematical formulas are the crystallization of human wisdom in exploring the laws of nature for thousands of years. Describing the complex laws of nature with a concise mathematical formula is a constant pursuit of scientists and a great challenge for artificial intelligence. This field is called symbolic regression (SR). Symbolic regression was originally formulated as a combinatorial optimization problem, and Genetic Programming (GP) and Reinforcement Learning algorithms were used to solve it. However, GP is sensitive to hyperparameters, and these two types of algorithms are inefficient. To solve this problem, researchers treat the mapping from data to expressions as a translation problem. And the corresponding large-scale pre-trained model is introduced. However, the data and expression skeletons do not have very clear word correspondences as the two languages do. Instead, they are more like two modalities (e.g., image and text). Therefore, in this paper, we proposed MMSR. The SR problem is solved as a pure multi-modal problem, and contrastive learning is also introduced in the training process for modal alignment to facilitate later modal feature fusion. It is worth noting that to better promote the modal feature fusion, we adopt the strategy of training contrastive learning loss and other losses at the same time, which only needs one-step training, instead of training contrastive learning loss first and then training other losses. Because our experiments prove training together can make the feature extraction module and feature fusion module wearing-in better. Experimental results show that compared with multiple large-scale pre-training baselines, MMSR achieves the most advanced results on multiple mainstream datasets including SRBench. Our code is open source at https://github.com/1716757342/MMSR

Yusong Deng, Min Wu, Lina Yu et al.

Symbolic regression is a task aimed at identifying patterns in data and representing them through mathematical expressions, generally involving skeleton prediction and constant optimization. Many methods have achieved some success, however they treat variables and symbols merely as characters of natural language without considering their mathematical essence. This paper introduces the operator feature neural network (OF-Net) which employs operator representation for expressions and proposes an implicit feature encoding method for the intrinsic mathematical operational logic of operators. By substituting operator features for numeric loss, we can predict the combination of operators of target expressions. We evaluate the model on public datasets, and the results demonstrate that the model achieves superior recovery rates and high $R^2$ scores. With the discussion of the results, we analyze the merit and demerit of OF-Net and propose optimizing schemes.

Yanjie Li

Property prediction and inverse structural design of catalytic materials are typically modeled as two independent tasks: the former predicts target properties from given structures, whereas the latter generates candidate structures according to desired properties. Although the decoupled paradigm facilitates the implementation of a ``generation--evaluation--screening'' workflow, the inconsistency between the generative model and the property prediction model in terms of representation spaces and training objectives can readily introduce data distribution shifts and evaluator bias, thereby limiting the stability of closed-loop optimization. In this work, we propose QE-Catalytic-V2, a unified graph--text multimodal large language model for catalytic materials, which integrates property prediction and inverse design within the same model and shared representation space. Under this unified framework, QE-Catalytic-V2 can not only perform reliable property prediction by leveraging three-dimensional structures and textual information, but also generate and screen physically feasible CIF candidates conditioned on target properties, thereby forming a closed-loop optimization workflow of ``inverse design--prediction--screening--redesign.'' Experimental results demonstrate that this unified paradigm outperforms decoupled baselines on both catalytic relaxed-energy prediction and inverse design tasks, validating the effectiveness of jointly modeling property prediction and structure generation within a single multimodal model.

Yanjie Li, Kaisheng Liang, Bin Xiao

In recent research, adversarial attacks on person detectors using patches or static 3D model-based texture modifications have struggled with low success rates due to the flexible nature of human movement. Modeling the 3D deformations caused by various actions has been a major challenge. Fortunately, advancements in Neural Radiance Fields (NeRF) for dynamic human modeling offer new possibilities. In this paper, we introduce UV-Attack, a groundbreaking approach that achieves high success rates even with extensive and unseen human actions. We address the challenge above by leveraging dynamic-NeRF-based UV mapping. UV-Attack can generate human images across diverse actions and viewpoints, and even create novel actions by sampling from the SMPL parameter space. While dynamic NeRF models are capable of modeling human bodies, modifying clothing textures is challenging because they are embedded in neural network parameters. To tackle this, UV-Attack generates UV maps instead of RGB images and modifies the texture stacks. This approach enables real-time texture edits and makes the attack more practical. We also propose a novel Expectation over Pose Transformation loss (EoPT) to improve the evasion success rate on unseen poses and views. Our experiments show that UV-Attack achieves a 92.7% attack success rate against the FastRCNN model across varied poses in dynamic video settings, significantly outperforming the state-of-the-art AdvCamou attack, which only had a 28.5% ASR. Moreover, we achieve 49.5% ASR on the latest YOLOv8 detector in black-box settings. This work highlights the potential of dynamic NeRF-based UV mapping for creating more effective adversarial attacks on person detectors, addressing key challenges in modeling human movement and texture modification. The code is available at https://github.com/PolyLiYJ/UV-Attack.

Chengxuan Lu, Shukuan Wang, Yanjie Li et al.

Reinforcement learning (RL) for large-scale Vision-Language-Action (VLA) models faces significant challenges in computational efficiency and data acquisition. We propose AcceRL, a fully asynchronous and decoupled RL framework designed to eliminate synchronization barriers by physically isolating training, inference, and rollouts. Crucially, AcceRL is the first to integrate a plug-and-play, trainable world model into a distributed asynchronous RL pipeline to generate virtual experiences. Experiments on the LIBERO benchmark demonstrate that AcceRL achieves state-of-the-art (SOTA) performance. Systematically, it exhibits super-linear scaling in throughput and highly efficient hardware utilization. Algorithmically, the world-model-augmented variant delivers unprecedented sample efficiency and robust training stability in complex control tasks.

Yanjie Li, Liping Zhang, Min Wu et al.

Mathematical formulas serve as a language through which humans communicate with nature. Discovering mathematical laws from scientific data to describe natural phenomena has been a long-standing pursuit of humanity for centuries. In the field of artificial intelligence, this challenge is known as the symbolic regression problem. Among existing symbolic regression approaches, Genetic Programming (GP) based on evolutionary algorithms remains one of the most classical and widely adopted methods. GP simulates the evolutionary process across generations through genetic mutation and crossover. However, mutations and crossovers in GP are entirely random. While this randomness effectively mimics natural evolution, it inevitably produces both beneficial and detrimental variations. If there existed a metaphorical `God` capable of foreseeing which genetic mutations or crossovers would yield superior outcomes and performing targeted gene editing accordingly, the efficiency of evolution could be substantially improved. Motivated by this idea, we propose in this paper a symbolic regression approach based on gene editing, termed GESR. In GESR, we trained two "hands of God" (two BERT models). Among them, the first leverages the BERT's masked language modeling capability to guide the mutation of genes (expression symbols). The other BERT model guides the crossover of individual genes by predicting the crossover point. Experimental results demonstrate that GESR significantly improves computational efficiency compared with traditional GP algorithms and achieves strong overall performance across multiple symbolic regression tasks.

Yanjie Li, Wenxuan Zhang, Xinqi Lyu et al.

Recently, text-to-image diffusion models have been widely used for style mimicry and personalized customization through methods such as DreamBooth and Textual Inversion. This has raised concerns about intellectual property protection and the generation of deceptive content. Recent studies, such as Glaze and Anti-DreamBooth, have proposed using adversarial noise to protect images from these attacks. However, recent purification-based methods, such as DiffPure and Noise Upscaling, have successfully attacked these latest defenses, showing the vulnerabilities of these methods. Moreover, present methods show limited transferability across models, making them less effective against unknown text-to-image models. To address these issues, we propose a novel anti-mimicry method, StyleGuard. We propose a novel style loss that optimizes the style-related features in the latent space to make it deviate from the original image, which improves model-agnostic transferability. Additionally, to enhance the perturbation's ability to bypass diffusion-based purification, we designed a novel upscale loss that involves ensemble purifiers and upscalers during training. Extensive experiments on the WikiArt and CelebA datasets demonstrate that StyleGuard outperforms existing methods in robustness against various transformations and purifications, effectively countering style mimicry in various models. Moreover, StyleGuard is effective on different style mimicry methods, including DreamBooth and Textual Inversion. The code is available at https://github.com/PolyLiYJ/StyleGuard.

Yuanhao Chen, Qi Liu, Pengbin Chen et al.

Offline reinforcement learning (RL) aims to learn a policy that maximizes the expected return using a given static dataset of transitions. However, offline RL faces the distribution shift problem. The policy constraint offline RL method is proposed to solve the distribution shift problem. During the policy constraint offline RL training, it is important to ensure the difference between the learned policy and behavior policy within a given threshold. Thus, the learned policy heavily relies on the quality of the behavior policy. However, a problem exists in existing policy constraint methods: if the dataset contains many low-reward transitions, the learned will be contained with a suboptimal reference policy, leading to slow learning speed, low sample efficiency, and inferior performances. This paper shows that the sampling method in policy constraint offline RL that uses all the transitions in the dataset can be improved. A simple but efficient sample filtering method is proposed to improve the sample efficiency and the final performance. First, we evaluate the score of the transitions by average reward and average discounted reward of episodes in the dataset and extract the transition samples of high scores. Second, the high-score transition samples are used to train the offline RL algorithms. We verify the proposed method in a series of offline RL algorithms and benchmark tasks. Experimental results show that the proposed method outperforms baselines.

Jian Liu, Xiongtao Shi, Thai Duy Nguyen et al.

The rapid evolution of artificial intelligence (AI) has shifted from static, data-driven models to dynamic systems capable of perceiving and interacting with real-world environments. Despite advancements in pattern recognition and symbolic reasoning, current AI systems, such as large language models, remain disembodied, unable to physically engage with the world. This limitation has driven the rise of embodied AI, where autonomous agents, such as humanoid robots, must navigate and manipulate unstructured environments with human-like adaptability. At the core of this challenge lies the concept of Neural Brain, a central intelligence system designed to drive embodied agents with human-like adaptability. A Neural Brain must seamlessly integrate multimodal sensing and perception with cognitive capabilities. Achieving this also requires an adaptive memory system and energy-efficient hardware-software co-design, enabling real-time action in dynamic environments. This paper introduces a unified framework for the Neural Brain of embodied agents, addressing two fundamental challenges: (1) defining the core components of Neural Brain and (2) bridging the gap between static AI models and the dynamic adaptability required for real-world deployment. To this end, we propose a biologically inspired architecture that integrates multimodal active sensing, perception-cognition-action function, neuroplasticity-based memory storage and updating, and neuromorphic hardware/software optimization. Furthermore, we also review the latest research on embodied agents across these four aspects and analyze the gap between current AI systems and human intelligence. By synthesizing insights from neuroscience, we outline a roadmap towards the development of generalizable, autonomous agents capable of human-level intelligence in real-world scenarios.

Kaisheng Liang, Xuelong Dai, Yanjie Li et al.

Deep neural networks (DNNs) exhibit vulnerability to adversarial examples that can transfer across different DNN models. A particularly challenging problem is developing transferable targeted attacks that can mislead DNN models into predicting specific target classes. While various methods have been proposed to enhance attack transferability, they often incur substantial computational costs while yielding limited improvements. Recent clean feature mixup methods use random clean features to perturb the feature space but lack optimization for disrupting adversarial examples, overlooking the advantages of attack-specific perturbations. In this paper, we propose Feature Tuning Mixup (FTM), a novel method that enhances targeted attack transferability by combining both random and optimized noises in the feature space. FTM introduces learnable feature perturbations and employs an efficient stochastic update strategy for optimization. These learnable perturbations facilitate the generation of more robust adversarial examples with improved transferability. We further demonstrate that attack performance can be enhanced through an ensemble of multiple FTM-perturbed surrogate models. Extensive experiments on the ImageNet-compatible dataset across various DNN models demonstrate that our method achieves significant improvements over state-of-the-art methods while maintaining low computational cost.

Yanjie Li, Weijun Li, Lina Yu et al.

The mathematical formula is the human language to describe nature and is the essence of scientific research. Finding mathematical formulas from observational data is a major demand of scientific research and a major challenge of artificial intelligence. This area is called symbolic regression. Originally symbolic regression was often formulated as a combinatorial optimization problem and solved using GP or reinforcement learning algorithms. These two kinds of algorithms have strong noise robustness ability and good Versatility. However, inference time usually takes a long time, so the search efficiency is relatively low. Later, based on large-scale pre-training data proposed, such methods use a large number of synthetic data points and expression pairs to train a Generative Pre-Trained Transformer(GPT). Then this GPT can only need to perform one forward propagation to obtain the results, the advantage is that the inference speed is very fast. However, its performance is very dependent on the training data and performs poorly on data outside the training set, which leads to poor noise robustness and Versatility of such methods. So, can we combine the advantages of the above two categories of SR algorithms? In this paper, we propose \textbf{FormulaGPT}, which trains a GPT using massive sparse reward learning histories of reinforcement learning-based SR algorithms as training data. After training, the SR algorithm based on reinforcement learning is distilled into a Transformer. When new test data comes, FormulaGPT can directly generate a "reinforcement learning process" and automatically update the learning policy in context. Tested on more than ten datasets including SRBench, formulaGPT achieves the state-of-the-art performance in fitting ability compared with four baselines. In addition, it achieves satisfactory results in noise robustness, versatility, and inference efficiency.

Yanjie Li, Weijun Li, Lina Yu et al.

Multilayer perception (MLP) has permeated various disciplinary domains, ranging from bioinformatics to financial analytics, where their application has become an indispensable facet of contemporary scientific research endeavors. However, MLP has obvious drawbacks. 1), The type of activation function is single and relatively fixed, which leads to poor `representation ability' of the network, and it is often to solve simple problems with complex networks; 2), the network structure is not adaptive, it is easy to cause network structure redundant or insufficient. In this work, we propose a novel neural network paradigm X-Net promising to replace MLPs. X-Net can dynamically learn activation functions individually based on derivative information during training to improve the network's representational ability for specific tasks. At the same time, X-Net can precisely adjust the network structure at the neuron level to accommodate tasks of varying complexity and reduce computational costs. We show that X-Net outperforms MLPs in terms of representational capability. X-Net can achieve comparable or even better performance than MLP with much smaller parameters on regression and classification tasks. Specifically, in terms of the number of parameters, X-Net is only 3% of MLP on average and only 1.1% under some tasks. We also demonstrate X-Net's ability to perform scientific discovery on data from various disciplines such as energy, environment, and aerospace, where X-Net is shown to help scientists discover new laws of mathematics or physics.

Qi Liu, Xiaopeng Zhang, Mingshan Tan et al.

This paper proposes a novel method to enhance locomotion for a single humanoid robot through cooperative-heterogeneous multi-agent deep reinforcement learning (MARL). While most existing methods typically employ single-agent reinforcement learning algorithms for a single humanoid robot or MARL algorithms for multi-robot system tasks, we propose a distinct paradigm: applying cooperative-heterogeneous MARL to optimize locomotion for a single humanoid robot. The proposed method, multi-agent reinforcement learning for single humanoid locomotion (MASH), treats each limb (legs and arms) as an independent agent that explores the robot's action space while sharing a global critic for cooperative learning. Experiments demonstrate that MASH accelerates training convergence and improves whole-body cooperation ability, outperforming conventional single-agent reinforcement learning methods. This work advances the integration of MARL into single-humanoid-robot control, offering new insights into efficient locomotion strategies.

Xinqi Lyu, Yihao Liu, Yanjie Li et al.

Text-to-Image (T2I) models have gained widespread adoption across various applications. Despite the success, the potential misuse of T2I models poses significant risks of generating Not-Safe-For-Work (NSFW) content. To investigate the vulnerability of T2I models, this paper delves into adversarial attacks to bypass the safety mechanisms under black-box settings. Most previous methods rely on word substitution to search adversarial prompts. Due to limited search space, this leads to suboptimal performance compared to gradient-based training. However, black-box settings present unique challenges to training gradient-driven attack methods, since there is no access to the internal architecture and parameters of T2I models. To facilitate the learning of adversarial prompts in black-box settings, we propose a novel prompt learning attack framework (PLA), where insightful gradient-based training tailored to black-box T2I models is designed by utilizing multimodal similarities. Experiments show that our new method can effectively attack the safety mechanisms of black-box T2I models including prompt filters and post-hoc safety checkers with a high success rate compared to state-of-the-art methods. Warning: This paper may contain offensive model-generated content.

Yanjie Li, Yiming Cao, Dong Wang et al.

Multimodal agents built on large vision-language models (LVLMs) are increasingly deployed in open-world settings but remain highly vulnerable to prompt injection, especially through visual inputs. We introduce AgentTypo, a black-box red-teaming framework that mounts adaptive typographic prompt injection by embedding optimized text into webpage images. Our automatic typographic prompt injection (ATPI) algorithm maximizes prompt reconstruction by substituting captioners while minimizing human detectability via a stealth loss, with a Tree-structured Parzen Estimator guiding black-box optimization over text placement, size, and color. To further enhance attack strength, we develop AgentTypo-pro, a multi-LLM system that iteratively refines injection prompts using evaluation feedback and retrieves successful past examples for continual learning. Effective prompts are abstracted into generalizable strategies and stored in a strategy repository, enabling progressive knowledge accumulation and reuse in future attacks. Experiments on the VWA-Adv benchmark across Classifieds, Shopping, and Reddit scenarios show that AgentTypo significantly outperforms the latest image-based attacks such as AgentAttack. On GPT-4o agents, our image-only attack raises the success rate from 0.23 to 0.45, with consistent results across GPT-4V, GPT-4o-mini, Gemini 1.5 Pro, and Claude 3 Opus. In image+text settings, AgentTypo achieves 0.68 ASR, also outperforming the latest baselines. Our findings reveal that AgentTypo poses a practical and potent threat to multimodal agents and highlight the urgent need for effective defense.

Yiming Cao, Yanjie Li, Kaisheng Liang et al.

The growing deployment of Large Vision-Language Models (VLMs) raises safety concerns, as adversaries may exploit model vulnerabilities to induce harmful outputs, with targeted black-box adversarial attacks posing a particularly severe threat. However, existing methods primarily maximize encoder-level global similarity, which lacks the granularity for stealthy and practical fine-grained attacks, where only specific target should be altered (e.g., modifying a car while preserving its background). Moreover, they largely neglect the projector, a key semantic bridge in VLMs for multimodal alignment. To address these limitations, we propose a novel black-box targeted attack framework that leverages the projector. Specifically, we utilize the widely adopted Querying Transformer (Q-Former) which transforms global image embeddings into fine-grained query outputs, to enhance attack effectiveness and granularity. For standard global targeted attack scenarios, we propose the Intermediate Projector Guided Attack (IPGA), which aligns Q-Former fine-grained query outputs with the target to enhance attack strength and exploits the intermediate pretrained Q-Former that is not fine-tuned for any specific Large Language Model (LLM) to improve attack transferability. For fine-grained attack scenarios, we augment IPGA with the Residual Query Alignment (RQA) module, which preserves unrelated content by constraining non-target query outputs to enhance attack granularity. Extensive experiments demonstrate that IPGA significantly outperforms baselines in global targeted attacks, and IPGA with RQA (IPGA-R) attains superior success rates and unrelated content preservation over baselines in fine-grained attacks. Our method also transfers effectively to commercial VLMs such as Google Gemini and OpenAI GPT.

Jianqi Gao, Xizheng Pang, Qi Liu et al.

Reinforcement learning-based mapless navigation holds significant potential. However, it faces challenges in indoor environments with local minima area. This paper introduces a safe mapless navigation framework utilizing hierarchical reinforcement learning (HRL) to enhance navigation through such areas. The high-level policy creates a sub-goal to direct the navigation process. Notably, we have developed a sub-goal update mechanism that considers environment congestion, efficiently avoiding the entrapment of the robot in local minimum areas. The low-level motion planning policy, trained through safe reinforcement learning, outputs real-time control instructions based on acquired sub-goal. Specifically, to enhance the robot's environmental perception, we introduce a new obstacle encoding method that evaluates the impact of obstacles on the robot's motion planning. To validate the performance of our HRL-based navigation framework, we conduct simulations in office, home, and restaurant environments. The findings demonstrate that our HRL-based navigation framework excels in both static and dynamic scenarios. Finally, we implement the HRL-based navigation framework on a TurtleBot3 robot for physical validation experiments, which exhibits its strong generalization capabilities.

Jingyi Liu, Yanjie Li, Lina Yu et al.

Noise ubiquitously exists in signals due to numerous factors including physical, electronic, and environmental effects. Traditional methods of symbolic regression, such as genetic programming or deep learning models, aim to find the most fitting expressions for these signals. However, these methods often overlook the noise present in real-world data, leading to reduced fitting accuracy. To tackle this issue, we propose \textit{\textbf{D}eep Symbolic Regression against \textbf{N}oise via \textbf{C}ontrastive \textbf{L}earning (DN-CL)}. DN-CL employs two parameter-sharing encoders to embed data points from various data transformations into feature shields against noise. This model treats noisy data and clean data as different views of the ground-truth mathematical expressions. Distances between these features are minimized, utilizing contrastive learning to distinguish between 'positive' noise-corrected pairs and 'negative' contrasting pairs. Our experiments indicate that DN-CL demonstrates superior performance in handling both noisy and clean data, presenting a promising method of symbolic regression.

Zhen Xiang, Linzhi Zheng, Yanjie Li et al.

The rapid advancement of large language model (LLM) agents has raised new concerns regarding their safety and security. In this paper, we propose GuardAgent, the first guardrail agent to protect target agents by dynamically checking whether their actions satisfy given safety guard requests. Specifically, GuardAgent first analyzes the safety guard requests to generate a task plan, and then maps this plan into guardrail code for execution. By performing the code execution, GuardAgent can deterministically follow the safety guard request and safeguard target agents. In both steps, an LLM is utilized as the reasoning component, supplemented by in-context demonstrations retrieved from a memory module storing experiences from previous tasks. In addition, we propose two novel benchmarks: EICU-AC benchmark to assess the access control for healthcare agents and Mind2Web-SC benchmark to evaluate the safety policies for web agents. We show that GuardAgent effectively moderates the violation actions for different types of agents on these two benchmarks with over 98% and 83% guardrail accuracies, respectively. Project page: https://guardagent.github.io/

Shu Wei, Yanjie Li, Lina Yu et al.

The quest for analytical solutions to differential equations has traditionally been constrained by the need for extensive mathematical expertise. Machine learning methods like genetic algorithms have shown promise in this domain, but are hindered by significant computational time and the complexity of their derived solutions. This paper introduces SSDE (Symbolic Solver for Differential Equations), a novel reinforcement learning-based approach that derives symbolic closed-form solutions for various differential equations. Evaluations across a diverse set of ordinary and partial differential equations demonstrate that SSDE outperforms existing machine learning methods, delivering superior accuracy and efficiency in obtaining analytical solutions.

Yanjie Li, Lina Yu, Weijun Li et al.

Formulas are the language of communication between humans and nature. The discovery of formulas to describe natural laws from observational data is the purpose of scientific research. It is also an important research topic in artificial intelligence, which is called a symbolic regression problem. Most of the existing symbolic regression methods generate expressions directly from observed data. Although in some methods, we can inject some prior knowledge into the model by adding constraints or introducing some special character hints. However, these methods can only introduce a limited amount of prior knowledge specified in advance. Not to mention understanding natural language instructions. In this article, based on the powerful knowledge reserve and language understanding ability of multi-modal large language models, we present ChatSR, which acts like a knowledgeable human scientist, and we can tell it any prior knowledge through natural language to guide it in formula generation. By testing on 13 datasets, ChatSR not only shows state-of-the-art performance on traditional symbolic regression tasks. More notably, ChatSR can well understand the prior knowledge contained in natural language prompts and improve the quality of generated expressions. In addition, it is exciting that ChatSR has a good zero-shot capability to understand prior knowledge that is not present in the training data.

Min Wu, Weijun Li, Lina Yu et al.

Symbolic regression aims to derive interpretable symbolic expressions from data in order to better understand and interpret data. %which plays an important role in knowledge discovery and interpretable machine learning. In this study, a symbolic network called PruneSymNet is proposed for symbolic regression. This is a novel neural network whose activation function consists of common elementary functions and operators. The whole network is differentiable and can be trained by gradient descent method. Each subnetwork in the network corresponds to an expression, and our goal is to extract such subnetworks to get the desired symbolic expression. Therefore, a greedy pruning algorithm is proposed to prune the network into a subnetwork while ensuring the accuracy of data fitting. The proposed greedy pruning algorithm preserves the edge with the least loss in each pruning, but greedy algorithm often can not get the optimal solution. In order to alleviate this problem, we combine beam search during pruning to obtain multiple candidate expressions each time, and finally select the expression with the smallest loss as the final result. It was tested on the public data set and compared with the current popular algorithms. The results showed that the proposed algorithm had better accuracy.

Yanjie Li, Weijun Li, Lina Yu et al.

Finding a concise and interpretable mathematical formula that accurately describes the relationship between each variable and the predicted value in the data is a crucial task in scientific research, as well as a significant challenge in artificial intelligence. This problem is referred to as symbolic regression, which is an NP-hard problem. In the previous year, a novel symbolic regression methodology utilizing Monte Carlo Tree Search (MCTS) was advanced, achieving state-of-the-art results on a diverse range of datasets. although this algorithm has shown considerable improvement in recovering target expressions compared to previous methods, the lack of guidance during the MCTS process severely hampers its search efficiency. Recently, some algorithms have added a pre-trained policy network to guide the search of MCTS, but the pre-trained policy network generalizes poorly. To optimize the trade-off between efficiency and versatility, we introduce SR-GPT, a novel algorithm for symbolic regression that integrates Monte Carlo Tree Search (MCTS) with a Generative Pre-Trained Transformer (GPT). By using GPT to guide the MCTS, the search efficiency of MCTS is significantly improved. Next, we utilize the MCTS results to further refine the GPT, enhancing its capabilities and providing more accurate guidance for the MCTS. MCTS and GPT are coupled together and optimize each other until the target expression is successfully determined. We conducted extensive evaluations of SR-GPT using 222 expressions sourced from over 10 different symbolic regression datasets. The experimental results demonstrate that SR-GPT outperforms existing state-of-the-art algorithms in accurately recovering symbolic expressions both with and without added noise.

Sen Yang, Zhicheng Wang, Ze Chen et al.

This paper presents a new method to solve keypoint detection and instance association by using Transformer. For bottom-up multi-person pose estimation models, they need to detect keypoints and learn associative information between keypoints. We argue that these problems can be entirely solved by Transformer. Specifically, the self-attention in Transformer measures dependencies between any pair of locations, which can provide association information for keypoints grouping. However, the naive attention patterns are still not subjectively controlled, so there is no guarantee that the keypoints will always attend to the instances to which they belong. To address it we propose a novel approach of supervising self-attention for multi-person keypoint detection and instance association. By using instance masks to supervise self-attention to be instance-aware, we can assign the detected keypoints to their corresponding instances based on the pairwise attention scores, without using pre-defined offset vector fields or embedding like CNN-based bottom-up models. An additional benefit of our method is that the instance segmentation results of any number of people can be directly obtained from the supervised attention matrix, thereby simplifying the pixel assignment pipeline. The experiments on the COCO multi-person keypoint detection challenge and person instance segmentation task demonstrate the effectiveness and simplicity of the proposed method and show a promising way to control self-attention behavior for specific purposes.

Xuelong Dai, Yanjie Li, Hua Dai et al.

Utilizing 3D point cloud data has become an urgent need for the deployment of artificial intelligence in many areas like facial recognition and self-driving. However, deep learning for 3D point clouds is still vulnerable to adversarial attacks, e.g., iterative attacks, point transformation attacks, and generative attacks. These attacks need to restrict perturbations of adversarial examples within a strict bound, leading to the unrealistic adversarial 3D point clouds. In this paper, we propose an Adversarial Graph-Convolutional Generative Adversarial Network (AdvGCGAN) to generate visually realistic adversarial 3D point clouds from scratch. Specifically, we use a graph convolutional generator and a discriminator with an auxiliary classifier to generate realistic point clouds, which learn the latent distribution from the real 3D data. The unrestricted adversarial attack loss is incorporated in the special adversarial training of GAN, which enables the generator to generate the adversarial examples to spoof the target network. Compared with the existing state-of-art attack methods, the experiment results demonstrate the effectiveness of our unrestricted adversarial attack methods with a higher attack success rate and visual quality. Additionally, the proposed AdvGCGAN can achieve better performance against defense models and better transferability than existing attack methods with strong camouflage.

Yanjie Li, Sen Yang, Peidong Liu et al.

The 2D heatmap-based approaches have dominated Human Pose Estimation (HPE) for years due to high performance. However, the long-standing quantization error problem in the 2D heatmap-based methods leads to several well-known drawbacks: 1) The performance for the low-resolution inputs is limited; 2) To improve the feature map resolution for higher localization precision, multiple costly upsampling layers are required; 3) Extra post-processing is adopted to reduce the quantization error. To address these issues, we aim to explore a brand new scheme, called \textit{SimCC}, which reformulates HPE as two classification tasks for horizontal and vertical coordinates. The proposed SimCC uniformly divides each pixel into several bins, thus achieving \emph{sub-pixel} localization precision and low quantization error. Benefiting from that, SimCC can omit additional refinement post-processing and exclude upsampling layers under certain settings, resulting in a more simple and effective pipeline for HPE. Extensive experiments conducted over COCO, CrowdPose, and MPII datasets show that SimCC outperforms heatmap-based counterparts, especially in low-resolution settings by a large margin.

Yanjie Li, Shoukui Zhang, Zhicheng Wang et al.

Human pose estimation deeply relies on visual clues and anatomical constraints between parts to locate keypoints. Most existing CNN-based methods do well in visual representation, however, lacking in the ability to explicitly learn the constraint relationships between keypoints. In this paper, we propose a novel approach based on Token representation for human Pose estimation~(TokenPose). In detail, each keypoint is explicitly embedded as a token to simultaneously learn constraint relationships and appearance cues from images. Extensive experiments show that the small and large TokenPose models are on par with state-of-the-art CNN-based counterparts while being more lightweight. Specifically, our TokenPose-S and TokenPose-L achieve $72.5$ AP and $75.8$ AP on COCO validation dataset respectively, with significant reduction in parameters ($\downarrow80.6\%$; $\downarrow$ $56.8\%$) and GFLOPs ($\downarrow$ $75.3\%$; $\downarrow$ $24.7\%$). Code is publicly available.

Yiming Li, Yanjie Li, Yalei Lv et al.

Deep neural networks (DNNs) are vulnerable to the \emph{backdoor attack}, which intends to embed hidden backdoors in DNNs by poisoning training data. The attacked model behaves normally on benign samples, whereas its prediction will be changed to a particular target label if hidden backdoors are activated. So far, backdoor research has mostly been conducted towards classification tasks. In this paper, we reveal that this threat could also happen in semantic segmentation, which may further endanger many mission-critical applications ($e.g.$, autonomous driving). Except for extending the existing attack paradigm to maliciously manipulate the segmentation models from the image-level, we propose a novel attack paradigm, the \emph{fine-grained attack}, where we treat the target label ($i.e.$, annotation) from the object-level instead of the image-level to achieve more sophisticated manipulation. In the annotation of poisoned samples generated by the fine-grained attack, only pixels of specific objects will be labeled with the attacker-specified target class while others are still with their ground-truth ones. Experiments show that the proposed methods can successfully attack semantic segmentation models by poisoning only a small proportion of training data. Our method not only provides a new perspective for designing novel attacks but also serves as a strong baseline for improving the robustness of semantic segmentation methods.

Guangliang Du, Minmin Wang, Canlin Zhou et al.

Conventional dual-frequency fringe projection algorithm often suffers from phase unwrapping failure when the frequency ratio between the high frequency and the low one is too large. Zhang et.al. proposed an enhanced two-frequency phase-shifting method to use geometric constraints of digital fringe projection(DFP) to reduce the noise impact due to the large frequency ratio. However, this method needs to calibrate the DFP system and calculate the minimum phase map at the nearest position from the camera perspective, these procedures are are relatively complex and more time-cosuming. In this paper, we proposed an improved method, which eliminates the system calibration and determination in Zhang's method,meanwhile does not need to use the low frequency fringe pattern. In the proposed method,we only need a set of high frequency fringe patterns to measure the object after the high frequency is directly estimated by the experiment. Thus the proposed method can simplify the procedure and improve the speed. Finally, the experimental evaluation is conducted to prove the validity of the proposed method.The results demonstrate that the proposed method can overcome the main disadvantages encountered by Zhang's method.

Minmin Wang, Guangliang Du, Canlin Zhou et al.

It is a challenge for Phase Measurement Profilometry (PMP) to measure objects with a large range of reflectivity variation across the surface. Saturated or dark pixels in the deformed fringe patterns captured by the camera will lead to phase fluctuations and errors. Jiang et al. proposed a high dynamic range real-time 3D shape measurement method without changing camera exposures. Three inverted phase-shifted fringe patterns are used to complement three regular phase-shifted fringe patterns for phase retrieval when any of the regular fringe patterns are saturated. But Jiang's method still has some drawbacks: (1) The phases in saturated pixels are respectively estimated by different formulas for different cases. It is shortage of an universal formula; (2) it cannot be extended to four-step phase-shifting algorithm because inverted fringe patterns are the repetition of regular fringe patterns; (3) only three unsaturated intensity values at every pixel of fringe patterns are chosen for phase demodulation, lying idle the other unsaturated ones. We proposed a method for enhanced high dynamic range 3D shape measurement based on generalized phase-shifting algorithm, which combines the complementary technique of inverted and regular fringe patterns with generalized phase-shifting algorithm. Firstly, two sets of complementary phase-shifted fringe patterns, namely regular and inverted fringe patterns are projected and collected. Then all unsaturated intensity values at the same camera pixel from two sets of fringe patterns are selected, and employed to retrieve the phase by generalized phase-shifting algorithm. Finally, simulations and experiments are conducted to prove the validity of the proposed method. The results are analyzed and compared with Jiang's method, which demonstrate that the proposed method not only expands the scope of Jiang's method, but also improves the measurement accuracy.