Ran Cheng

Tao Jiang, Xinmeng Yu, Chenhao Yi et al.

Evolutionary model merging provides a powerful framework for the automated, training-free composition of LLMs through parameter-space search. However, existing methods predominantly rely on stochastic, hand-crafted operators that overlook the underlying performance landscape of the coefficient space. We propose Evolutionary Generative Merging (EvoGM), a framework that transcends manual heuristics by employing learnable generative modeling to optimize merging coefficients. Specifically, EvoGM features a dual-generator architecture with cycle-consistent learning to adaptively sample and refine promising merging candidates. By constructing winner-loser pairs from historical search trajectories, our framework effectively captures high-performance parameter distributions and maximizes data efficiency. This generative process is seamlessly integrated into a multi-round evolutionary pipeline, where elite merged models iteratively serve as new expert foundations. Extensive experiments across diverse benchmarks demonstrate that EvoGM significantly outperforms state-of-the-art baselines, exhibiting robust performance on both seen and unseen tasks. Code and data are available at https://github.com/JiangTao97/evogm.

Zhichao Lu, Ran Cheng, Yaochu Jin et al.

The ongoing advancements in network architecture design have led to remarkable achievements in deep learning across various challenging computer vision tasks. Meanwhile, the development of neural architecture search (NAS) has provided promising approaches to automating the design of network architectures for lower prediction error. Recently, the emerging application scenarios of deep learning have raised higher demands for network architectures considering multiple design criteria: number of parameters/floating-point operations, and inference latency, among others. From an optimization point of view, the NAS tasks involving multiple design criteria are intrinsically multiobjective optimization problems; hence, it is reasonable to adopt evolutionary multiobjective optimization (EMO) algorithms for tackling them. Nonetheless, there is still a clear gap confining the related research along this pathway: on the one hand, there is a lack of a general problem formulation of NAS tasks from an optimization point of view; on the other hand, there are challenges in conducting benchmark assessments of EMO algorithms on NAS tasks. To bridge the gap: (i) we formulate NAS tasks into general multi-objective optimization problems and analyze the complex characteristics from an optimization point of view; (ii) we present an end-to-end pipeline, dubbed $\texttt{EvoXBench}$, to generate benchmark test problems for EMO algorithms to run efficiently -- without the requirement of GPUs or Pytorch/Tensorflow; (iii) we instantiate two test suites comprehensively covering two datasets, seven search spaces, and three hardware devices, involving up to eight objectives. Based on the above, we validate the proposed test suites using six representative EMO algorithms and provide some empirical analyses. The code of $\texttt{EvoXBench}$ is available from $\href{https://github.com/EMI-Group/EvoXBench}{\rm{here}}$.

Jian Yao, Xiongcai Luo, Ran Cheng et al.

Recent advances in Large Reasoning Models have significantly improved chain-of-thought (CoT) capabilities via reinforcement learning (RL). However, generated reasoning chains frequently suffer from structural redundancy (i.e., \emph{overthinking}), incurring high computational overhead without improving answer correctness. Existing mitigation strategies typically rely on token-uniform length penalties, which provide coarse, segment-agnostic pressure toward shorter outputs and can inadvertently suppress useful reasoning alongside redundancy. To address this, we demonstrate that inefficiency concentrates in high-probability segments with low marginal utility. We derive a theoretical characterization of segment suboptimality under the correctness-length trade-off objective and propose \textsc{SLAT} (Segment-Level Adaptive Trimming), an RL framework that selectively suppresses redundant segments based on this criterion. Empirical results on standard benchmarks indicate that \textsc{SLAT} establishes a superior accuracy-efficiency Pareto frontier, reducing reasoning length by $50\%$ relative to uncompressed baselines while maintaining competitive accuracy. Overall, our results suggest that theoretically grounded, segment-aware trimming is a promising direction for efficient CoT reasoning in large language models.

Teng Wang, Yixiao Ge, Feng Zheng et al. · tencent-ai

The state of the arts in vision-language pretraining (VLP) achieves exemplary performance but suffers from high training costs resulting from slow convergence and long training time, especially on large-scale web datasets. An essential obstacle to training efficiency lies in the entangled prediction rate (percentage of tokens for reconstruction) and corruption rate (percentage of corrupted tokens) in masked language modeling (MLM), that is, a proper corruption rate is achieved at the cost of a large portion of output tokens being excluded from prediction loss. To accelerate the convergence of VLP, we propose a new pretraining task, namely, free language modeling (FLM), that enables a 100% prediction rate with arbitrary corruption rates. FLM successfully frees the prediction rate from the tie-up with the corruption rate while allowing the corruption spans to be customized for each token to be predicted. FLM-trained models are encouraged to learn better and faster given the same GPU time by exploiting bidirectional contexts more flexibly. Extensive experiments show FLM could achieve an impressive 2.5x pretraining time reduction in comparison to the MLM-based methods, while keeping competitive performance on both vision-language understanding and generation tasks. Code will be public at https://github.com/TencentARC/FLM.

Chuntao Ding, Zhichao Lu, Shangguang Wang et al.

Multi-task learning (MTL) seeks to learn a single model to accomplish multiple tasks by leveraging shared information among the tasks. Existing MTL models, however, have been known to suffer from negative interference among tasks. Efforts to mitigate task interference have focused on either loss/gradient balancing or implicit parameter partitioning with partial overlaps among the tasks. In this paper, we propose ETR-NLP to mitigate task interference through a synergistic combination of non-learnable primitives (NLPs) and explicit task routing (ETR). Our key idea is to employ non-learnable primitives to extract a diverse set of task-agnostic features and recombine them into a shared branch common to all tasks and explicit task-specific branches reserved for each task. The non-learnable primitives and the explicit decoupling of learnable parameters into shared and task-specific ones afford the flexibility needed for minimizing task interference. We evaluate the efficacy of ETR-NLP networks for both image-level classification and pixel-level dense prediction MTL problems. Experimental results indicate that ETR-NLP significantly outperforms state-of-the-art baselines with fewer learnable parameters and similar FLOPs across all datasets. Code is available at this \href{https://github.com/zhichao-lu/etr-nlp-mtl}.

Hui Bai, Ran Cheng, Yaochu Jin

Reinforcement learning (RL) is a machine learning approach that trains agents to maximize cumulative rewards through interactions with environments. The integration of RL with deep learning has recently resulted in impressive achievements in a wide range of challenging tasks, including board games, arcade games, and robot control. Despite these successes, there remain several crucial challenges, including brittle convergence properties caused by sensitive hyperparameters, difficulties in temporal credit assignment with long time horizons and sparse rewards, a lack of diverse exploration, especially in continuous search space scenarios, difficulties in credit assignment in multi-agent reinforcement learning, and conflicting objectives for rewards. Evolutionary computation (EC), which maintains a population of learning agents, has demonstrated promising performance in addressing these limitations. This article presents a comprehensive survey of state-of-the-art methods for integrating EC into RL, referred to as evolutionary reinforcement learning (EvoRL). We categorize EvoRL methods according to key research fields in RL, including hyperparameter optimization, policy search, exploration, reward shaping, meta-RL, and multi-objective RL. We then discuss future research directions in terms of efficient methods, benchmarks, and scalable platforms. This survey serves as a resource for researchers and practitioners interested in the field of EvoRL, highlighting the important challenges and opportunities for future research. With the help of this survey, researchers and practitioners can develop more efficient methods and tailored benchmarks for EvoRL, further advancing this promising cross-disciplinary research field.

Teng Wang, Jinrui Zhang, Feng Zheng et al.

Joint video-language learning has received increasing attention in recent years. However, existing works mainly focus on single or multiple trimmed video clips (events), which makes human-annotated event boundaries necessary during inference. To break away from the ties, we propose a grounded vision-language learning framework for untrimmed videos, which automatically detects informative events and effectively excavates the alignments between multi-sentence descriptions and corresponding event segments. Instead of coarse-level video-language alignments, we present two dual pretext tasks to encourage fine-grained segment-level alignments, i.e., text-to-event grounding (TEG) and event-to-text generation (ETG). TEG learns to adaptively ground the possible event proposals given a set of sentences by estimating the cross-modal distance in a joint semantic space. Meanwhile, ETG aims to reconstruct (generate) the matched texts given event proposals, encouraging the event representation to retain meaningful semantic information. To encourage accurate label assignment between the event set and the text set, we propose a novel semantic-aware cost to mitigate the sub-optimal matching results caused by ambiguous boundary annotations. Our framework is easily extensible to tasks covering visually-grounded language understanding and generation. We achieve state-of-the-art dense video captioning performance on ActivityNet Captions, YouCook2 and YouMakeup, and competitive performance on several other language generation and understanding tasks. Our method also achieved 1st place in both the MTVG and MDVC tasks of the PIC 4th Challenge. Our code is publicly available at https://github.com/zjr2000/GVL.

Hui Bai, Ruimin Shen, Yue Lin et al.

Despite the emerging progress of integrating evolutionary computation into reinforcement learning, the absence of a high-performance platform endowing composability and massive parallelism causes non-trivial difficulties for research and applications related to asynchronous commercial games. Here we introduce Lamarckian - an open-source platform featuring support for evolutionary reinforcement learning scalable to distributed computing resources. To improve the training speed and data efficiency, Lamarckian adopts optimized communication methods and an asynchronous evolutionary reinforcement learning workflow. To meet the demand for an asynchronous interface by commercial games and various methods, Lamarckian tailors an asynchronous Markov Decision Process interface and designs an object-oriented software architecture with decoupled modules. In comparison with the state-of-the-art RLlib, we empirically demonstrate the unique advantages of Lamarckian on benchmark tests with up to 6000 CPU cores: i) both the sampling efficiency and training speed are doubled when running PPO on Google football game; ii) the training speed is 13 times faster when running PBT+PPO on Pong game. Moreover, we also present two use cases: i) how Lamarckian is applied to generating behavior-diverse game AI; ii) how Lamarckian is applied to game balancing tests for an asynchronous commercial game.

Jinrui Zhang, Teng Wang, Feng Zheng et al.

Generic Event Boundary Captioning (GEBC) aims to generate three sentences describing the status change for a given time boundary. Previous methods only process the information of a single boundary at a time, which lacks utilization of video context information. To tackle this issue, we design a model that directly takes the whole video as input and generates captions for all boundaries parallelly. The model could learn the context information for each time boundary by modeling the boundary-boundary interactions. Experiments demonstrate the effectiveness of context information. The proposed method achieved a 72.84 score on the test set, and we reached the $2^{nd}$ place in this challenge. Our code is available at: \url{https://github.com/zjr2000/Context-GEBC}

Zhichao Lu, Ran Cheng, Shihua Huang et al.

The architectural advancements in deep neural networks have led to remarkable leap-forwards across a broad array of computer vision tasks. Instead of relying on human expertise, neural architecture search (NAS) has emerged as a promising avenue toward automating the design of architectures. While recent achievements in image classification have suggested opportunities, the promises of NAS have yet to be thoroughly assessed on more challenging tasks of semantic segmentation. The main challenges of applying NAS to semantic segmentation arise from two aspects: (i) high-resolution images to be processed; (ii) additional requirement of real-time inference speed (i.e., real-time semantic segmentation) for applications such as autonomous driving. To meet such challenges, we propose a surrogate-assisted multi-objective method in this paper. Through a series of customized prediction models, our method effectively transforms the original NAS task into an ordinary multi-objective optimization problem. Followed by a hierarchical pre-screening criterion for in-fill selection, our method progressively achieves a set of efficient architectures trading-off between segmentation accuracy and inference speed. Empirical evaluations on three benchmark datasets together with an application using Huawei Atlas 200 DK suggest that our method can identify architectures significantly outperforming existing state-of-the-art architectures designed both manually by human experts and automatically by other NAS methods.

Jia Liu, Ran Cheng, Yaochu Jin

Deep neural networks have been found vulnerable to adversarial attacks, thus raising potentially concerns in security-sensitive contexts. To address this problem, recent research has investigated the adversarial robustness of deep neural networks from the architectural point of view. However, searching for architectures of deep neural networks is computationally expensive, particularly when coupled with adversarial training process. To meet the above challenge, this paper proposes a bi-fidelity multiobjective neural architecture search approach. First, we formulate the NAS problem for enhancing adversarial robustness of deep neural networks into a multiobjective optimization problem. Specifically, in addition to a low-fidelity performance predictor as the first objective, we leverage an auxiliary-objective -- the value of which is the output of a surrogate model trained with high-fidelity evaluations. Secondly, we reduce the computational cost by combining three performance estimation methods, i.e., parameter sharing, low-fidelity evaluation, and surrogate-based predictor. The effectiveness of the proposed approach is confirmed by extensive experiments conducted on CIFAR-10, CIFAR-100 and SVHN datasets.

Boyan Li, Luziwei Leng, Shuaijie Shen et al.

Advancements in adapting deep convolution architectures for Spiking Neural Networks (SNNs) have significantly enhanced image classification performance and reduced computational burdens. However, the inability of Multiplication-Free Inference (MFI) to align with attention and transformer mechanisms, which are critical to superior performance on high-resolution vision tasks, imposing limitations on these gains. To address this, our research explores a new pathway, drawing inspiration from the progress made in Multi-Layer Perceptrons (MLPs). We propose an innovative spiking MLP architecture that uses batch normalization to retain MFI compatibility and introducing a spiking patch encoding layer to enhance local feature extraction capabilities. As a result, we establish an efficient multi-stage spiking MLP network that blends effectively global receptive fields with local feature extraction for comprehensive spike-based computation. Without relying on pre-training or sophisticated SNN training techniques, our network secures a top-1 accuracy of 66.39% on the ImageNet-1K dataset, surpassing the directly trained spiking ResNet-34 by 2.67%. Furthermore, we curtail computational costs, model parameters, and simulation steps. An expanded version of our network compares with the performance of the spiking VGG-16 network with a 71.64% top-1 accuracy, all while operating with a model capacity 2.1 times smaller. Our findings highlight the potential of our deep SNN architecture in effectively integrating global and local learning abilities. Interestingly, the trained receptive field in our network mirrors the activity patterns of cortical cells. Source codes are publicly accessible at https://github.com/EMI-Group/mixer-snn.

Junjie Wen, Yichen Zhu, Jinming Li et al.

Vision-Language-Action (VLA) models have shown remarkable potential in visuomotor control and instruction comprehension through end-to-end learning processes. However, current VLA models face significant challenges: they are slow during inference and require extensive pre-training on large amounts of robotic data, making real-world deployment difficult. In this paper, we introduce a new family of compact vision-language-action models, called TinyVLA, which offers two key advantages over existing VLA models: (1) faster inference speeds, and (2) improved data efficiency, eliminating the need for pre-training stage. Our framework incorporates two essential components to build TinyVLA: (1) initializing the policy backbone with robust, high-speed multimodal models, and (2) integrating a diffusion policy decoder during fine-tuning to enable precise robot actions. We conducted extensive evaluations of TinyVLA in both simulation and on real robots, demonstrating that our approach significantly outperforms the state-of-the-art VLA model, OpenVLA, in terms of speed and data efficiency, while delivering comparable or superior performance. Additionally, TinyVLA exhibits strong generalization capabilities across various dimensions, including language instructions, novel objects, unseen positions, changes in object appearance, background variations, and environmental shifts, often matching or exceeding the performance of OpenVLA. We believe that \methodname offers an interesting perspective on utilizing pre-trained multimodal models for policy learning. Our project is at https://tiny-vla.github.io.

Hu Zhang, Yanchen Li, Luziwei Leng et al.

Event-based sensors, distinguished by their high temporal resolution of 1 $\mathrmμ\text{s}$ and a dynamic range of 120 $\text{dB}$, stand out as ideal tools for deployment in fast-paced settings like vehicles and drones. Traditional object detection techniques that utilize Artificial Neural Networks (ANNs) face challenges due to the sparse and asynchronous nature of the events these sensors capture. In contrast, Spiking Neural Networks (SNNs) offer a promising alternative, providing a temporal representation that is inherently aligned with event-based data. This paper explores the unique membrane potential dynamics of SNNs and their ability to modulate sparse events. We introduce an innovative spike-triggered adaptive threshold mechanism designed for stable training. Building on these insights, we present a specialized spiking feature pyramid network (SpikeFPN) optimized for automotive event-based object detection. Comprehensive evaluations demonstrate that SpikeFPN surpasses both traditional SNNs and advanced ANNs enhanced with attention mechanisms. Evidently, SpikeFPN achieves a mean Average Precision (mAP) of 0.477 on the GEN1 Automotive Detection (GAD) benchmark dataset, marking significant increases over the selected SNN baselines. Moreover, the efficient design of SpikeFPN ensures robust performance while optimizing computational resources, attributed to its innate sparse computation capabilities. Source codes are publicly accessible at https://github.com/EMI-Group/spikefpn.

Teng Wang, Wenhao Jiang, Zhichao Lu et al.

Existing vision-language pre-training (VLP) methods primarily rely on paired image-text datasets, which are either annotated by enormous human labors, or crawled from the internet followed by elaborate data cleaning techniques. To reduce the dependency on well-aligned image-text pairs, it is promising to directly leverage the large-scale text-only and image-only corpora. This paper proposes a data augmentation method, namely cross-modal CutMix (CMC), for implicit cross-modal alignment learning in unpaired VLP. Specifically, CMC transforms natural sentences from the textual view into a multi-modal view, where visually-grounded words in a sentence are randomly replaced by diverse image patches with similar semantics. There are several appealing proprieties of the proposed CMC. First, it enhances the data diversity while keeping the semantic meaning intact for tackling problems where the aligned data are scarce; Second, by attaching cross-modal noise on uni-modal data, it guides models to learn token-level interactions across modalities for better denoising. Furthermore, we present a new unpaired VLP method, dubbed as VLMixer, that integrates CMC with contrastive learning to pull together the uni-modal and multi-modal views for better instance-level alignments among different modalities. Extensive experiments on five downstream tasks show that VLMixer could surpass previous state-of-the-art unpaired VLP methods.

Ruihao Xia, Junhong Cai, Luziwei Leng et al.

Event cameras offer unique advantages for vision tasks in challenging environments, yet processing asynchronous event streams remains an open challenge. While existing methods rely on specialized architectures or resource-intensive training, the potential of leveraging modern Visual Foundation Models (VFMs) pretrained on image data remains under-explored for event-based vision. To address this, we propose Temporal-Guided VFM (TGVFM), a novel framework that integrates VFMs with our temporal context fusion block seamlessly to bridge this gap. Our temporal block introduces three key components: (1) Long-Range Temporal Attention to model global temporal dependencies, (2) Dual Spatiotemporal Attention for multi-scale frame correlation, and (3) Deep Feature Guidance Mechanism to fuse semantic-temporal features. By retraining event-to-video models on real-world data and leveraging transformer-based VFMs, TGVFM preserves spatiotemporal dynamics while harnessing pretrained representations. Experiments demonstrate SoTA performance across semantic segmentation, depth estimation, and object detection, with improvements of 16%, 21%, and 16% over existing methods, respectively. Overall, this work unlocks the cross-modality potential of image-based VFMs for event-based vision with temporal reasoning. Code is available at https://github.com/XiaRho/TGVFM.

Yanchen Li, Jiachun Li, Kebin Sun et al.

Drawing on the intricate structures of the brain, Spiking Neural Networks (SNNs) emerge as a transformative development in artificial intelligence, closely emulating the complex dynamics of biological neural networks. While SNNs show promising efficiency on specialized sparse-computational hardware, their practical training often relies on conventional GPUs. This reliance frequently leads to extended computation times when contrasted with traditional Artificial Neural Networks (ANNs), presenting significant hurdles for advancing SNN research. To navigate this challenge, we present a novel temporal fusion method, specifically designed to expedite the propagation dynamics of SNNs on GPU platforms, which serves as an enhancement to the current significant approaches for handling deep learning tasks with SNNs. This method underwent thorough validation through extensive experiments in both authentic training scenarios and idealized conditions, confirming its efficacy and adaptability for single and multi-GPU systems. Benchmarked against various existing SNN libraries/implementations, our method achieved accelerations ranging from $5\times$ to $40\times$ on NVIDIA A100 GPUs. Publicly available experimental codes can be found at https://github.com/EMI-Group/snn-temporal-fusion.

Haoyu Zhang, Zhihao Yu, Rui Wang et al.

Modern computer vision requires balancing predictive accuracy with real-time efficiency, yet the high inference cost of large vision models (LVMs) limits deployment on resource-constrained edge devices. Although Evolutionary Neural Architecture Search (ENAS) is well suited for multi-objective optimization, its practical use is hindered by two issues: expensive candidate evaluation and ranking inconsistency among subnetworks. To address them, we propose EvoNAS, an efficient distributed framework for multi-objective evolutionary architecture search. We build a hybrid supernet that integrates Vision State Space and Vision Transformer (VSS-ViT) modules, and optimize it with a Cross-Architecture Dual-Domain Knowledge Distillation (CA-DDKD) strategy. By coupling the computational efficiency of VSS blocks with the semantic expressiveness of ViT modules, CA-DDKD improves the representational capacity of the shared supernet and enhances ranking consistency, enabling reliable fitness estimation during evolution without extra fine-tuning. To reduce the cost of large-scale validation, we further introduce a Distributed Multi-Model Parallel Evaluation (DMMPE) framework based on GPU resource pooling and asynchronous scheduling. Compared with conventional data-parallel evaluation, DMMPE improves efficiency by over 70% through concurrent multi-GPU, multi-model execution. Experiments on COCO, ADE20K, KITTI, and NYU-Depth v2 show that the searched architectures, termed EvoNets, consistently achieve Pareto-optimal trade-offs between accuracy and efficiency. Compared with representative CNN-, ViT-, and Mamba-based models, EvoNets deliver lower inference latency and higher throughput under strict computational budgets while maintaining strong generalization on downstream tasks such as novel view synthesis. Code is available at https://github.com/EMI-Group/evonas

Yansong Huang, Zherui Zhang, Ao Jiao et al.

Evolutionary multi-objective optimization (EMO) algorithms have been demonstrated to be effective in solving multi-criteria decision-making problems. In real-world applications, analysts often employ several algorithms concurrently and compare their solution sets to gain insight into the characteristics of different algorithms and explore a broader range of feasible solutions. However, EMO algorithms are typically treated as black boxes, leading to difficulties in performing detailed analysis and comparisons between the internal evolutionary processes. Inspired by the successful application of visual analytics tools in explainable AI, we argue that interactive visualization can significantly enhance the comparative analysis between multiple EMO algorithms. In this paper, we present a visual analytics framework that enables the exploration and comparison of evolutionary processes in EMO algorithms. Guided by a literature review and expert interviews, the proposed framework addresses various analytical tasks and establishes a multi-faceted visualization design to support the comparative analysis of intermediate generations in the evolution as well as solution sets. We demonstrate the effectiveness of our framework through case studies on benchmarking and real-world multi-objective optimization problems to elucidate how analysts can leverage our framework to inspect and compare diverse algorithms.

Zhichao Lu, Chuntao Ding, Shangguang Wang et al.

Deploying high-performance convolutional neural network (CNN) models on low-earth orbit (LEO) satellites for rapid remote sensing image processing has attracted significant interest from industry and academia. However, the limited resources available on LEO satellites contrast with the demands of resource-intensive CNN models, necessitating the adoption of ground-station server assistance for training and updating these models. Existing approaches often require large floating-point operations (FLOPs) and substantial model parameter transmissions, presenting considerable challenges. To address these issues, this paper introduces a ground-station server-assisted framework. With the proposed framework, each layer of the CNN model contains only one learnable feature map (called the seed feature map) from which other feature maps are generated based on specific rules. The hyperparameters of these rules are randomly generated instead of being trained, thus enabling the generation of multiple feature maps from the seed feature map and significantly reducing FLOPs. Furthermore, since the random hyperparameters can be saved using a few random seeds, the ground station server assistance can be facilitated in updating the CNN model deployed on the LEO satellite. Experimental results on the ISPRS Vaihingen, ISPRS Potsdam, UAVid, and LoveDA datasets for semantic segmentation services demonstrate that the proposed framework outperforms existing state-of-the-art approaches. In particular, the SineFM-based model achieves a higher mIoU than the UNetFormer on the UAVid dataset, with 3.3x fewer parameters and 2.2x fewer FLOPs.

Jiaqi Yang, Yucong Chen, Xiangting Meng et al.

Recently there has been a growing interest in category-level object pose and size estimation, and prevailing methods commonly rely on single view RGB-D images. However, one disadvantage of such methods is that they require accurate depth maps which cannot be produced by consumer-grade sensors. Furthermore, many practical real-world situations involve a moving camera that continuously observes its surroundings, and the temporal information of the input video streams is simply overlooked by single-view methods. We propose a novel solution that makes use of RGB video streams. Our framework consists of three modules: a scale-aware monocular dense SLAM solution, a lightweight object pose predictor, and an object-level pose graph optimizer. The SLAM module utilizes a video stream and additional scale-sensitive readings to estimate camera poses and metric depth. The object pose predictor then generates canonical object representations from RGB images. The object pose is estimated through geometric registration of these canonical object representations with estimated object depth points. All per-view estimates finally undergo optimization within a pose graph, culminating in the output of robust and accurate canonical object poses. Our experimental results demonstrate that when utilizing public dataset sequences with high-quality depth information, the proposed method exhibits comparable performance to state-of-the-art RGB-D methods. We also collect and evaluate on new datasets containing depth maps of varying quality to further quantitatively benchmark the proposed method alongside previous RGB-D based methods. We demonstrate a significant advantage in scenarios where depth input is absent or the quality of depth sensing is limited.

Teng Wang, Zhu Liu, Feng Zheng et al.

This report describes the details of our approach for the event dense-captioning task in ActivityNet Challenge 2021. We present a semantic-aware pretraining method for dense video captioning, which empowers the learned features to recognize high-level semantic concepts. Diverse video features of different modalities are fed into an event captioning module to generate accurate and meaningful sentences. Our final ensemble model achieves a 10.00 METEOR score on the test set.

Jiaxin Wei, Lige Liu, Ran Cheng et al.

Recent years have witnessed the surge of learned representations that directly build upon point clouds. Though becoming increasingly expressive, most existing representations still struggle to generate ordered point sets. Inspired by spherical multi-view scanners, we propose a novel sampling model called Spotlights to represent a 3D shape as a compact 1D array of depth values. It simulates the configuration of cameras evenly distributed on a sphere, where each virtual camera casts light rays from its principal point through sample points on a small concentric spherical cap to probe for the possible intersections with the object surrounded by the sphere. The structured point cloud is hence given implicitly as a function of depths. We provide a detailed geometric analysis of this new sampling scheme and prove its effectiveness in the context of the point cloud completion task. Experimental results on both synthetic and real data demonstrate that our method achieves competitive accuracy and consistency while having a significantly reduced computational cost. Furthermore, we show superior performance on the downstream point cloud registration task over state-of-the-art completion methods.

Rui Zhang, Luziwei Leng, Kaiwei Che et al.

Spiking neural networks (SNNs), known for their low-power, event-driven computation and intrinsic temporal dynamics, are emerging as promising solutions for processing dynamic, asynchronous signals from event-based sensors. Despite their potential, SNNs face challenges in training and architectural design, resulting in limited performance in challenging event-based dense prediction tasks compared to artificial neural networks (ANNs). In this work, we develop an efficient spiking encoder-decoder network (SpikingEDN) for large-scale event-based semantic segmentation tasks. To enhance the learning efficiency from dynamic event streams, we harness the adaptive threshold which improves network accuracy, sparsity and robustness in streaming inference. Moreover, we develop a dual-path Spiking Spatially-Adaptive Modulation module, which is specifically tailored to enhance the representation of sparse events and multi-modal inputs, thereby considerably improving network performance. Our SpikingEDN attains a mean intersection over union (MIoU) of 72.57\% on the DDD17 dataset and 58.32\% on the larger DSEC-Semantic dataset, showing competitive results to the state-of-the-art ANNs while requiring substantially fewer computational resources. Our results shed light on the untapped potential of SNNs in event-based vision applications. The source code will be made publicly available.

Zirui Ge, Pengxiang Ding, Baohua Yin et al.

Video action models are an appealing foundation for Vision--Language--Action systems because they can learn visual dynamics from large-scale video data and transfer this knowledge to downstream robot control. Yet current diffusion-based video predictors are trained with likelihood-surrogate objectives, which encourage globally plausible predictions without explicitly optimizing the precision-critical visual dynamics needed for manipulation. This objective mismatch often leads to subtle errors in object pose, spatial relations, and contact timing that can be amplified by downstream policies. We propose VAMPO, a post-training framework that directly improves visual dynamics in video action models through policy optimization. Our key idea is to formulate multi-step denoising as a sequential decision process and optimize the denoising policy with rewards defined over expert visual dynamics in latent space. To make this optimization practical, we introduce an Euler Hybrid sampler that injects stochasticity only at the first denoising step, enabling tractable low-variance policy-gradient estimation while preserving the coherence of the remaining denoising trajectory. We further combine this design with GRPO and a verifiable non-adversarial reward. Across diverse simulated and real-world manipulation tasks, VAMPO improves task-relevant visual dynamics, leading to better downstream action generation and stronger generalization. The homepage is https://vampo-robot.github.io/VAMPO/.

Yang Cheng, Qingyuan Shu, Albert Lee et al.

Stochastic diffusion processes are pervasive in nature, from the seemingly erratic Brownian motion to the complex interactions of synaptically-coupled spiking neurons. Recently, drawing inspiration from Langevin dynamics, neuromorphic diffusion models were proposed and have become one of the major breakthroughs in the field of generative artificial intelligence. Unlike discriminative models that have been well developed to tackle classification or regression tasks, diffusion models as well as other generative models such as ChatGPT aim at creating content based upon contexts learned. However, the more complex algorithms of these models result in high computational costs using today's technologies, creating a bottleneck in their efficiency, and impeding further development. Here, we develop a spintronic voltage-controlled magnetoelectric memory hardware for the neuromorphic diffusion process. The in-memory computing capability of our spintronic devices goes beyond current Von Neumann architecture, where memory and computing units are separated. Together with the non-volatility of magnetic memory, we can achieve high-speed and low-cost computing, which is desirable for the increasing scale of generative models in the current era. We experimentally demonstrate that the hardware-based true random diffusion process can be implemented for image generation and achieve comparable image quality to software-based training as measured by the Frechet inception distance (FID) score, achieving ~10^3 better energy-per-bit-per-area over traditional hardware.

Wei Wu, Fan Lu, Yunnan Wang et al.

Offering great potential in robotic manipulation, a capable Vision-Language-Action (VLA) foundation model is expected to faithfully generalize across tasks and platforms while ensuring cost efficiency (e.g., data and GPU hours required for adaptation). To this end, we develop LingBot-VLA with around 20,000 hours of real-world data from 9 popular dual-arm robot configurations. Through a systematic assessment on 3 robotic platforms, each completing 100 tasks with 130 post-training episodes per task, our model achieves clear superiority over competitors, showcasing its strong performance and broad generalizability. We have also built an efficient codebase, which delivers a throughput of 261 samples per second per GPU with an 8-GPU training setup, representing a 1.5~2.8$\times$ (depending on the relied VLM base model) speedup over existing VLA-oriented codebases. The above features ensure that our model is well-suited for real-world deployment. To advance the field of robot learning, we provide open access to the code, base model, and benchmark data, with a focus on enabling more challenging tasks and promoting sound evaluation standards.

Ran Cheng, Xinyu Jiang, Yuan Chen et al.

Camera relocalization is the key component of simultaneous localization and mapping (SLAM) systems. This paper proposes a learning-based approach, named Sparse Spatial Scene Embedding with Graph Neural Networks (S3E-GNN), as an end-to-end framework for efficient and robust camera relocalization. S3E-GNN consists of two modules. In the encoding module, a trained S3E network encodes RGB images into embedding codes to implicitly represent spatial and semantic embedding code. With embedding codes and the associated poses obtained from a SLAM system, each image is represented as a graph node in a pose graph. In the GNN query module, the pose graph is transformed to form a embedding-aggregated reference graph for camera relocalization. We collect various scene datasets in the challenging environments to perform experiments. Our results demonstrate that S3E-GNN method outperforms the traditional Bag-of-words (BoW) for camera relocalization due to learning-based embedding and GNN powered scene matching mechanism.

Junhong Cai, Guiqin Wang, Kejie Zhao et al.

Large Language Models (LLMs) excel across diverse domains but suffer from high energy costs due to quadratic attention and dense Feed-Forward Network (FFN) operations. To address these issues, we propose Module-aware Architecture Refinement (MAR), a two-stage framework that integrates State Space Models (SSMs) for linear-time sequence modeling and applies activation sparsification to reduce FFN costs. In addition, to mitigate low information density and temporal mismatch in integrating Spiking Neural Networks (SNNs) with SSMs, we design the Adaptive Ternary Multi-step Neuron (ATMN) and the Spike-aware Bidirectional Distillation Strategy (SBDS). Extensive experiments demonstrate that MAR effectively restores the performance of its dense counterpart under constrained resources while substantially reducing inference energy consumption. Furthermore, it outperforms efficient models of comparable or even larger scale, underscoring its potential for building efficient and practical LLMs.

Wenjia Hua, Kejie Zhao, Luziwei Leng et al.

Backpropagation algorithm has driven the remarkable success of deep neural networks, but its lack of biological plausibility and high computational costs have motivated the ongoing search for alternative training methods. Hebbian learning has attracted considerable interest as a biologically plausible alternative to backpropagation. Nevertheless, its exclusive reliance on local information, without consideration of global task objectives, fundamentally limits its scalability. Inspired by the biological synergy between neuromodulators and local plasticity, we introduce a novel model-agnostic Global-guided Hebbian Learning (GHL) framework, which seamlessly integrates local and global information to scale up across diverse networks and tasks. In specific, the local component employs Oja's rule with competitive learning to ensure stable and effective local updates. Meanwhile, the global component introduces a sign-based signal that guides the direction of local Hebbian plasticity updates. Extensive experiments demonstrate that our method consistently outperforms existing Hebbian approaches. Notably, on large-scale network and complex datasets like ImageNet, our framework achieves the competitive results and significantly narrows the gap with standard backpropagation.

Ziyu Wang, Chenyuan Liu, Yushun Xiang et al.

Recently, with the rapid development of robot learning and imitation learning, numerous datasets and methods have emerged. However, these datasets and their task designs often lack systematic consideration and principles. This raises important questions: Do the current datasets and task designs truly advance the capabilities of robotic agents? Do evaluations on a few common tasks accurately reflect the differentiated performance of various methods proposed by different teams and evaluated on different tasks? To address these issues, we introduce the Great March 100 (\textbf{GM-100}) as the first step towards a robot learning Olympics. GM-100 consists of 100 carefully designed tasks that cover a wide range of interactions and long-tail behaviors, aiming to provide a diverse and challenging set of tasks to comprehensively evaluate the capabilities of robotic agents and promote diversity and complexity in robot dataset task designs. These tasks are developed through systematic analysis and expansion of existing task designs, combined with insights from human-object interaction primitives and object affordances. We collect a large amount of trajectory data on different robotic platforms and evaluate several baseline models. Experimental results demonstrate that the GM-100 tasks are 1) feasible to execute and 2) sufficiently challenging to effectively differentiate the performance of current VLA models. Our data and code are available at https://rhos.ai/research/gm-100.

Bowen Zheng, Ran Cheng

In the history of knowledge distillation, the focus has once shifted over time from logit-based to feature-based approaches. However, this transition has been revisited with the advent of Decoupled Knowledge Distillation (DKD), which re-emphasizes the importance of logit knowledge through advanced decoupling and weighting strategies. While DKD marks a significant advancement, its underlying mechanisms merit deeper exploration. As a response, we rethink DKD from a predictive distribution perspective. First, we introduce an enhanced version, the Generalized Decoupled Knowledge Distillation (GDKD) loss, which offers a more versatile method for decoupling logits. Then we pay particular attention to the teacher model's predictive distribution and its impact on the gradients of GDKD loss, uncovering two critical insights often overlooked: (1) the partitioning by the top logit considerably improves the interrelationship of non-top logits, and (2) amplifying the focus on the distillation loss of non-top logits enhances the knowledge extraction among them. Utilizing these insights, we further propose a streamlined GDKD algorithm with an efficient partition strategy to handle the multimodality of teacher models' predictive distribution. Our comprehensive experiments conducted on a variety of benchmarks, including CIFAR-100, ImageNet, Tiny-ImageNet, CUB-200-2011, and Cityscapes, demonstrate GDKD's superior performance over both the original DKD and other leading knowledge distillation methods. The code is available at https://github.com/ZaberKo/GDKD.

Yisong Zhang, Ran Cheng, Guoxing Yi et al.

Large Language Models (LLMs), with their strong understanding and reasoning capabilities, are increasingly being explored for tackling optimization problems, especially in synergy with evolutionary computation. Despite rapid progress, however, the field still lacks a unified synthesis and a systematic taxonomy. This survey addresses this gap by providing a comprehensive review of recent developments and organizing them within a structured framework. We classify existing research into two main stages: LLMs for optimization modeling and LLMs for optimization solving. The latter is further divided into three paradigms according to the role of LLMs in the optimization workflow: LLMs as stand-alone optimizers, low-level LLMs embedded within optimization algorithms, and high-level LLMs for algorithm selection and generation. For each category, we analyze representative methods, distill technical challenges, and examine their interplay with traditional approaches. We also review interdisciplinary applications spanning the natural sciences, engineering, and machine learning. By contrasting LLM-driven and conventional methods, we highlight key limitations and research gaps, and point toward future directions for developing self-evolving agentic ecosystems for optimization. An up-to-date collection of related literature is maintained at https://github.com/ishmael233/LLM4OPT.

Zeyuan Ma, Yue-Jiao Gong, Hongshu Guo et al.

Meta-Black-Box Optimization (MetaBBO) streamlines the automation of optimization algorithm design through meta-learning. It typically employs a bi-level structure: the meta-level policy undergoes meta-training to reduce the manual effort required in developing algorithms for low-level optimization tasks. The original MetaBox (2023) provided the first open-source framework for reinforcement learning-based single-objective MetaBBO. However, its relatively narrow scope no longer keep pace with the swift advancement in this field. In this paper, we introduce MetaBox-v2 (https://github.com/MetaEvo/MetaBox) as a milestone upgrade with four novel features: 1) a unified architecture supporting RL, evolutionary, and gradient-based approaches, by which we reproduce $23$ up-to-date baselines; 2) efficient parallelization schemes, which reduce the training/testing time by $10-40$x; 3) a comprehensive benchmark suite of $18$ synthetic/realistic tasks ($1900$+ instances) spanning single-objective, multi-objective, multi-model, and multi-task optimization scenarios; 4) plentiful and extensible interfaces for custom analysis/visualization and integrating to external optimization tools/benchmarks. To show the utility of MetaBox-v2, we carry out a systematic case study that evaluates the built-in baselines in terms of the optimization performance, generalization ability and learning efficiency. Valuable insights are concluded from thorough and detailed analysis for practitioners and those new to the field.

Yifan Zhao, Zhenyu Liang, Zhichao Lu et al.

As one of the emerging challenges in Automated Machine Learning, the Hardware-aware Neural Architecture Search (HW-NAS) tasks can be treated as black-box multi-objective optimization problems (MOPs). An important application of HW-NAS is real-time semantic segmentation, which plays a pivotal role in autonomous driving scenarios. The HW-NAS for real-time semantic segmentation inherently needs to balance multiple optimization objectives, including model accuracy, inference speed, and hardware-specific considerations. Despite its importance, benchmarks have yet to be developed to frame such a challenging task as multi-objective optimization. To bridge the gap, we introduce a tailored streamline to transform the task of HW-NAS for real-time semantic segmentation into standard MOPs. Building upon the streamline, we present a benchmark test suite, CitySeg/MOP, comprising fifteen MOPs derived from the Cityscapes dataset. The CitySeg/MOP test suite is integrated into the EvoXBench platform to provide seamless interfaces with various programming languages (e.g., Python and MATLAB) for instant fitness evaluations. We comprehensively assessed the CitySeg/MOP test suite on various multi-objective evolutionary algorithms, showcasing its versatility and practicality. Source codes are available at https://github.com/EMI-Group/evoxbench.

Zhihong Wu, Lishuang Wang, Kebin Sun et al.

Tree-based Genetic Programming (TGP) is a widely used evolutionary algorithm for tasks such as symbolic regression, classification, and robotic control. Due to the intensive computational demands of running TGP, GPU acceleration is crucial for achieving scalable performance. However, efficient GPU-based execution of TGP remains challenging, primarily due to three core issues: (1) the structural heterogeneity of program individuals, (2) the complexity of integrating multiple levels of parallelism, and (3) the incompatibility between high-performance CUDA execution and flexible Python-based environments. To address these issues, we propose EvoGP, a high-performance framework tailored for GPU acceleration of TGP via population-level parallel execution. First, EvoGP introduces a tensorized representation that encodes variable-sized trees into fixed-shape, memory-aligned arrays, enabling uniform memory access and parallel computation across diverse individuals. Second, EvoGP adopts an adaptive parallelism strategy that dynamically combines intra- and inter-individual parallelism based on dataset size, ensuring high GPU utilization across a broad spectrum of tasks. Third, EvoGP embeds custom CUDA kernels into the PyTorch runtime, achieving seamless integration with Python-based environments such as Gym, MuJoCo, Brax, and Genesis. Experiments show that EvoGP attains a peak throughput exceeding $10^{11}$ GPops/s, with speedups of up to $528\times$ over GPU-based TGP implementations and $18\times$ over the fastest CPU-based libraries, while maintaining comparable accuracy and improved scalability across large population sizes. EvoGP is open source and accessible at: https://github.com/EMI-Group/evogp.

Beichen Huang, Ran Cheng, Kay Chen Tan

We introduce EvoGit, a decentralized multi-agent framework for collaborative software development driven by autonomous code evolution. EvoGit deploys a population of independent coding agents, each proposing edits to a shared codebase without centralized coordination, explicit message passing, or shared memory. Instead, all coordination emerges through a Git-based phylogenetic graph that tracks the full version lineage and enables agents to asynchronously read from and write to the evolving code repository. This graph-based structure supports fine-grained branching, implicit concurrency, and scalable agent interaction while preserving a consistent historical record. Human involvement is minimal but strategic: users define high-level goals, periodically review the graph, and provide lightweight feedback to promote promising directions or prune unproductive ones. Experiments demonstrate EvoGit's ability to autonomously produce functional and modular software artifacts across two real-world tasks: (1) building a web application from scratch using modern frameworks, and (2) constructing a meta-level system that evolves its own language-model-guided solver for the bin-packing optimization problem. Our results underscore EvoGit's potential to establish a new paradigm for decentralized, automated, and continual software development. EvoGit is open-sourced at https://github.com/BillHuang2001/evogit.

Haoming Zhang, Ran Cheng

Neural Architecture Search (NAS) has emerged as a key tool in identifying optimal configurations of deep neural networks tailored to specific tasks. However, training and assessing numerous architectures introduces considerable computational overhead. One method to mitigating this is through performance predictors, which offer a means to estimate the potential of an architecture without exhaustive training. Given that neural architectures fundamentally resemble Directed Acyclic Graphs (DAGs), Graph Neural Networks (GNNs) become an apparent choice for such predictive tasks. Nevertheless, the scarcity of training data can impact the precision of GNN-based predictors. To address this, we introduce a novel GNN predictor for NAS. This predictor renders neural architectures into vector representations by combining both the conventional and inverse graph views. Additionally, we incorporate a customized training loss within the GNN predictor to ensure efficient utilization of both types of representations. We subsequently assessed our method through experiments on benchmark datasets including NAS-Bench-101, NAS-Bench-201, and the DARTS search space, with a training dataset ranging from 50 to 400 samples. Benchmarked against leading GNN predictors, the experimental results showcase a significant improvement in prediction accuracy, with a 3%--16% increase in Kendall-tau correlation. Source codes are available at https://github.com/EMI-Group/fr-nas.

Teng Wang, Ruimao Zhang, Zhichao Lu et al.

Dense video captioning aims to generate multiple associated captions with their temporal locations from the video. Previous methods follow a sophisticated "localize-then-describe" scheme, which heavily relies on numerous hand-crafted components. In this paper, we proposed a simple yet effective framework for end-to-end dense video captioning with parallel decoding (PDVC), by formulating the dense caption generation as a set prediction task. In practice, through stacking a newly proposed event counter on the top of a transformer decoder, the PDVC precisely segments the video into a number of event pieces under the holistic understanding of the video content, which effectively increases the coherence and readability of predicted captions. Compared with prior arts, the PDVC has several appealing advantages: (1) Without relying on heuristic non-maximum suppression or a recurrent event sequence selection network to remove redundancy, PDVC directly produces an event set with an appropriate size; (2) In contrast to adopting the two-stage scheme, we feed the enhanced representations of event queries into the localization head and caption head in parallel, making these two sub-tasks deeply interrelated and mutually promoted through the optimization; (3) Without bells and whistles, extensive experiments on ActivityNet Captions and YouCook2 show that PDVC is capable of producing high-quality captioning results, surpassing the state-of-the-art two-stage methods when its localization accuracy is on par with them. Code is available at https://github.com/ttengwang/PDVC.

Shihua Huang, Zhichao Lu, Ran Cheng et al.

Recent advancements in deep neural networks have made remarkable leap-forwards in dense image prediction. However, the issue of feature alignment remains as neglected by most existing approaches for simplicity. Direct pixel addition between upsampled and local features leads to feature maps with misaligned contexts that, in turn, translate to mis-classifications in prediction, especially on object boundaries. In this paper, we propose a feature alignment module that learns transformation offsets of pixels to contextually align upsampled higher-level features; and another feature selection module to emphasize the lower-level features with rich spatial details. We then integrate these two modules in a top-down pyramidal architecture and present the Feature-aligned Pyramid Network (FaPN). Extensive experimental evaluations on four dense prediction tasks and four datasets have demonstrated the efficacy of FaPN, yielding an overall improvement of 1.2 - 2.6 points in AP / mIoU over FPN when paired with Faster / Mask R-CNN. In particular, our FaPN achieves the state-of-the-art of 56.7% mIoU on ADE20K when integrated within Mask-Former. The code is available from https://github.com/EMI-Group/FaPN.

Ran Cheng, Christopher Agia, Yuan Ren et al.

With the increasing reliance of self-driving and similar robotic systems on robust 3D vision, the processing of LiDAR scans with deep convolutional neural networks has become a trend in academia and industry alike. Prior attempts on the challenging Semantic Scene Completion task - which entails the inference of dense 3D structure and associated semantic labels from "sparse" representations - have been, to a degree, successful in small indoor scenes when provided with dense point clouds or dense depth maps often fused with semantic segmentation maps from RGB images. However, the performance of these systems drop drastically when applied to large outdoor scenes characterized by dynamic and exponentially sparser conditions. Likewise, processing of the entire sparse volume becomes infeasible due to memory limitations and workarounds introduce computational inefficiency as practitioners are forced to divide the overall volume into multiple equal segments and infer on each individually, rendering real-time performance impossible. In this work, we formulate a method that subsumes the sparsity of large-scale environments and present S3CNet, a sparse convolution based neural network that predicts the semantically completed scene from a single, unified LiDAR point cloud. We show that our proposed method outperforms all counterparts on the 3D task, achieving state-of-the art results on the SemanticKITTI benchmark. Furthermore, we propose a 2D variant of S3CNet with a multi-view fusion strategy to complement our 3D network, providing robustness to occlusions and extreme sparsity in distant regions. We conduct experiments for the 2D semantic scene completion task and compare the results of our sparse 2D network against several leading LiDAR segmentation models adapted for bird's eye view segmentation on two open-source datasets.

Hao Tan, Ran Cheng, Shihua Huang et al.

Despite the remarkable successes of Convolutional Neural Networks (CNNs) in computer vision, it is time-consuming and error-prone to manually design a CNN. Among various Neural Architecture Search (NAS) methods that are motivated to automate designs of high-performance CNNs, the differentiable NAS and population-based NAS are attracting increasing interests due to their unique characters. To benefit from the merits while overcoming the deficiencies of both, this work proposes a novel NAS method, RelativeNAS. As the key to efficient search, RelativeNAS performs joint learning between fast-learners (i.e. networks with relatively higher accuracy) and slow-learners in a pairwise manner. Moreover, since RelativeNAS only requires low-fidelity performance estimation to distinguish each pair of fast-learner and slow-learner, it saves certain computation costs for training the candidate architectures. The proposed RelativeNAS brings several unique advantages: (1) it achieves state-of-the-art performance on ImageNet with top-1 error rate of 24.88%, i.e. outperforming DARTS and AmoebaNet-B by 1.82% and 1.12% respectively; (2) it spends only nine hours with a single 1080Ti GPU to obtain the discovered cells, i.e. 3.75x and 7875x faster than DARTS and AmoebaNet respectively; (3) it provides that the discovered cells obtained on CIFAR-10 can be directly transferred to object detection, semantic segmentation, and keypoint detection, yielding competitive results of 73.1% mAP on PASCAL VOC, 78.7% mIoU on Cityscapes, and 68.5% AP on MSCOCO, respectively. The implementation of RelativeNAS is available at https://github.com/EMI-Group/RelativeNAS

Ye Tian, Ran Cheng, Xingyi Zhang et al.

Over the last three decades, a large number of evolutionary algorithms have been developed for solving multiobjective optimization problems. However, there lacks an up-to-date and comprehensive software platform for researchers to properly benchmark existing algorithms and for practitioners to apply selected algorithms to solve their real-world problems. The demand of such a common tool becomes even more urgent, when the source code of many proposed algorithms has not been made publicly available. To address these issues, we have developed a MATLAB platform for evolutionary multi-objective optimization in this paper, called PlatEMO, which includes more than 50 multi-objective evolutionary algorithms and more than 100 multi-objective test problems, along with several widely used performance indicators. With a user-friendly graphical user interface, PlatEMO enables users to easily compare several evolutionary algorithms at one time and collect statistical results in Excel or LaTeX files. More importantly, PlatEMO is completely open source, such that users are able to develop new algorithms on the basis of it. This paper introduces the main features of PlatEMO and illustrates how to use it for performing comparative experiments, embedding new algorithms, creating new test problems, and developing performance indicators. Source code of PlatEMO is now available at: http://bimk.ahu.edu.cn/index.php?s=/Index/Software/index.html.

Zhongyi Zhou, Yichen Zhu, Minjie Zhu et al.

Humans possess a unified cognitive ability to perceive, comprehend, and interact with the physical world. Why can't large language models replicate this holistic understanding? Through a systematic analysis of existing training paradigms in vision-language-action models (VLA), we identify two key challenges: spurious forgetting, where robot training overwrites crucial visual-text alignments, and task interference, where competing control and understanding tasks degrade performance when trained jointly. To overcome these limitations, we propose ChatVLA, a novel framework featuring Phased Alignment Training, which incrementally integrates multimodal data after initial control mastery, and a Mixture-of-Experts architecture to minimize task interference. ChatVLA demonstrates competitive performance on visual question-answering datasets and significantly surpasses state-of-the-art vision-language-action (VLA) methods on multimodal understanding benchmarks. Notably, it achieves a six times higher performance on MMMU and scores 47.2% on MMStar with a more parameter-efficient design than ECoT. Furthermore, ChatVLA demonstrates superior performance on 25 real-world robot manipulation tasks compared to existing VLA methods like OpenVLA. Our findings highlight the potential of our unified framework for achieving both robust multimodal understanding and effective robot control.

Ran Cheng, Lailai Zhu

Data assimilation (DA) is a fundamental component of modern weather prediction, yet it remains a major computational bottleneck in machine learning (ML)-based forecasting pipelines due to reliance on traditional variational methods. Recent generative ML-based DA methods offer a promising alternative but typically require many sampling steps and suffer from error accumulation under long-horizon auto-regressive rollouts with cycling assimilation. We propose FlowDA, a low-latency weather-scale generative DA framework based on flow matching. FlowDA conditions on observations through a SetConv-based embedding and fine-tunes the Aurora foundation model to deliver accurate, efficient, and robust analyses. Experiments across observation rates decreasing from $3.9\%$ to $0.1\%$ demonstrate superior performance of FlowDA over strong baselines with similar tunable-parameter size. FlowDA further shows robustness to observational noise and stable performance in long-horizon auto-regressive cycling DA. Overall, FlowDA points to an efficient and scalable direction for data-driven DA.

Jian Yao, Ran Cheng, Xingyu Wu et al.

The reasoning capabilities of large language models (LLMs) have advanced rapidly, particularly following the release of DeepSeek R1, which has inspired a surge of research into data quality and reinforcement learning (RL) algorithms. Despite the pivotal role diversity plays in RL, its influence on LLM reasoning remains largely underexplored. To bridge this gap, this work presents a systematic investigation into the impact of diversity in RL-based training for LLM reasoning, and proposes a novel diversity-aware policy optimization method. Across evaluations on 12 LLMs, we observe a strong positive correlation between the solution diversity and Potential at k (a novel metric quantifying an LLM's reasoning potential) in high-performing models. This finding motivates our method to explicitly promote diversity during RL training. Specifically, we design a token-level diversity and reformulate it into a practical objective, then we selectively apply it to positive samples. Integrated into the R1-zero training framework, our method achieves a 3.5 percent average improvement across four mathematical reasoning benchmarks, while generating more diverse and robust solutions.

Hui Bai, Ran Cheng

Hyperparameter optimization plays a key role in the machine learning domain. Its significance is especially pronounced in reinforcement learning (RL), where agents continuously interact with and adapt to their environments, requiring dynamic adjustments in their learning trajectories. To cater to this dynamicity, the Population-Based Training (PBT) was introduced, leveraging the collective intelligence of a population of agents learning simultaneously. However, PBT tends to favor high-performing agents, potentially neglecting the explorative potential of agents on the brink of significant advancements. To mitigate the limitations of PBT, we present the Generalized Population-Based Training (GPBT), a refined framework designed for enhanced granularity and flexibility in hyperparameter adaptation. Complementing GPBT, we further introduce Pairwise Learning (PL). Instead of merely focusing on elite agents, PL employs a comprehensive pairwise strategy to identify performance differentials and provide holistic guidance to underperforming agents. By integrating the capabilities of GPBT and PL, our approach significantly improves upon traditional PBT in terms of adaptability and computational efficiency. Rigorous empirical evaluations across a range of RL benchmarks confirm that our approach consistently outperforms not only the conventional PBT but also its Bayesian-optimized variant.

Ran Cheng

Catastrophic forgetting remains a central challenge in continual learning (CL), yet lacks a unified information-theoretic explanation for why some architectures forget catastrophically while others do not. We introduce \emph{Context Channel Capacity} ($C_\mathrm{ctx}$), the mutual information between a CL architecture's context signal and its generated parameters, and prove that zero forgetting requires $C_\mathrm{ctx} \geq H(T)$, where $H(T)$ is the task identity entropy. We establish an \emph{Impossibility Triangle} -- zero forgetting, online learning, and finite parameters cannot be simultaneously satisfied by sequential state-based learners -- and show that conditional regeneration architectures (HyperNetworks) bypass this triangle by redefining parameters as function values rather than states. We validate this framework across 8 CL methods on Split-MNIST (1,130+ experiments over 86 days, 4 seeds each), showing that $C_\mathrm{ctx}$ perfectly predicts forgetting behavior: methods with $C_\mathrm{ctx} = 0$ (NaiveSGD, EWC, SI, LwF, CFlow) exhibit catastrophic forgetting (6--97\%), while methods with $C_\mathrm{ctx} \approx 1$ (HyperNetwork) achieve zero forgetting (98.8\% ACC). We further propose \emph{Wrong-Context Probing} (P5), a practical diagnostic protocol for measuring $C_\mathrm{ctx}$, and extend the framework to CIFAR-10 via a novel \emph{Gradient Context Encoder} that closes the oracle gap from 23.3pp to 0.7pp. A systematic taxonomy of 15+ closed research directions -- including the Hebbian null result (frozen random features outperform learned features), CFlow's $θ_0$-memorizer phenomenon, and the $S_N$ symmetry barrier to column specialization -- provides the community with precisely diagnosed negative results. Our central design principle: \emph{architecture over algorithm} -- the context pathway must be structurally unbypassable.

Jian Yao, Ran Cheng, Kay Chen Tan

Recent advances in reinforcement learning (RL) have led to substantial improvements in the mathematical reasoning abilities of LLMs, as measured by standard benchmarks. Yet these gains often persist even when models are trained with flawed signals, such as random or inverted rewards. This raises a fundamental question: do such improvements reflect genuine reasoning, or are they merely artifacts of overfitting to benchmark-specific patterns? To answer this question, we adopt an evaluation-centric perspective and highlight two critical shortcomings in existing protocols. First, benchmark contamination arises because test problems are publicly available, thereby increasing the risk of data leakage. Second, evaluation fragility results from reliance on single-instance assessments, which are sensitive to stochastic outputs and fail to capture reasoning consistency. These limitations suggest the need for a new evaluation paradigm that can probe reasoning ability beyond memorization and one-off success. As response, we propose VAR-MATH, a symbolic evaluation framework that converts fixed numerical problems into parameterized templates and requires models to solve multiple instantiations of each. This design enforces consistency across structurally equivalent variants, mitigates contamination, and enhances robustness through bootstrapped metrics. We apply VAR-MATH to transform three popular benchmarks, AMC23, AIME24, and AIME25, into their symbolic counterparts, VAR-AMC23, VAR-AIME24, and VAR-AIME25. Experimental results show substantial performance drops for RL-trained models on these variabilized benchmarks, especially for smaller models, with average declines of 47.9\% on AMC23, 58.8\% on AIME24, and 72.9\% on AIME25. These findings indicate that some existing RL methods rely on superficial heuristics and fail to generalize beyond specific numerical forms.

Haoxi Zhang, Xinxu Zhang, Yuanxin Lin et al.

Automatic karyotype analysis is often defined as a visual perception task focused solely on chromosomal object-level modeling. This definition has led most existing methods to overlook componential and holistic information, significantly constraining model performance. Moreover, the lack of interpretability in current technologies hinders clinical adoption. In this paper, we introduce Tokensome, a novel vision-language model based on chromosome tokenization for explainable and cognitive karyotyping. Tokensome elevates the method from the conventional visual perception layer to the cognitive decision-making layer. This elevation enables the integration of domain knowledge and cognitive reasoning via knowledge graphs and LLMs, markedly enhancing model's explainability and facilitating abnormality detection.