Boyu Chen

Boyu Chen, Peixia Li, Lei Bai et al.

Exploiting a general-purpose neural architecture to replace hand-wired designs or inductive biases has recently drawn extensive interest. However, existing tracking approaches rely on customized sub-modules and need prior knowledge for architecture selection, hindering the tracking development in a more general system. This paper presents a Simplified Tracking architecture (SimTrack) by leveraging a transformer backbone for joint feature extraction and interaction. Unlike existing Siamese trackers, we serialize the input images and concatenate them directly before the one-branch backbone. Feature interaction in the backbone helps to remove well-designed interaction modules and produce a more efficient and effective framework. To reduce the information loss from down-sampling in vision transformers, we further propose a foveal window strategy, providing more diverse input patches with acceptable computational costs. Our SimTrack improves the baseline with 2.5%/2.6% AUC gains on LaSOT/TNL2K and gets results competitive with other specialized tracking algorithms without bells and whistles.

Jiashuo Yu, Yao Yao, Boyu Chen et al.

We address the challenge of generating high-fidelity, long-form soundtracks that remain coherent across scene transitions. Existing AI music systems are mainly designed for short, isolated clips and lack mechanisms to ensure narrative continuity. We present JenBridge, a modular and interpretable framework for adaptive long-form video soundtracking that ensures both high-fidelity audio generation and transition naturalness. The core architecture is a Transformer-based generative model trained with a flow-matching objective, following a two-stage paradigm: pretraining on large-scale text-audio corpora to establish robust musical priors, then adapting to the video domain with dual text-visual conditioning for precise cross-modal alignment. Crucially, to achieve long-form coherence across diverse scene changes, JenBridge incorporates a novel adaptive transition mechanism. This system features a versatile toolkit of transition styles, including a generative transition method, and uniquely employs a Large Language Model (LLM) Agent that acts as a director to select the most appropriate transition for each narrative shift intelligently. To rigorously assess this task, we propose the LVS Benchmark, a new benchmark that includes a curated dataset and novel evaluation metrics focusing on holistic and transition-aware assessment. Extensive experiments on the proposed benchmark demonstrate that JenBridge significantly outperforms existing methods in both objective and subjective metrics, particularly in terms of transition naturalness and overall narrative coherence. JenBridge represents a significant step towards fully automated, professional-quality video soundtracking.

Zikang Wang, Boyu Chen, Zhengrong Yue et al.

Recent advances in video understanding have been driven by MLLMs. But these MLLMs are good at analyzing short videos, while suffering from difficulties in understanding videos with a longer context. To address this difficulty, several agent methods have been proposed, using MLLMs as agents for retrieving extra contextual knowledge in a long video. However, most existing agents ignore the key fact that a long video is composed with multiple shots, i.e., to answer the user question from a long video, it is critical to deeply understand its relevant shots like human. Without such insight, these agents often mistakenly find redundant even noisy temporal context, restricting their capacity for long video understanding. To fill this gap, we propose VideoChat-A1, a novel long video agent paradigm. Different from the previous works, our VideoChat-A1 can deeply think with long videos, via a distinct chain-of-shot reasoning paradigm. More specifically, it can progressively select the relevant shots of user question, and look into these shots in a coarse-to-fine partition. By multi-modal reasoning along the shot chain, VideoChat-A1 can effectively mimic step-by-step human thinking process, allowing the interactive discovery of preferable temporal context for thoughtful understanding in long videos. Extensive experiments show that, VideoChat-A1 achieves the state-of-the-art performance on the mainstream long video QA benchmarks, e.g., it achieves 77.0 on VideoMME (w/ subs) and 70.1 on EgoSchema, outperforming its strong baselines (e.g., InternVL2.5-8B and InternVideo2.5-8B), by up to 10.1\% and 6.2\%. Compared to leading closed-source GPT-4o and Gemini 1.5 Pro, VideoChat-A1 offers competitive accuracy, but only with 7% input frames and 12% inference time on average. The code is available on https://github.com/SpXace/VideoChat-A1.

Xinjia Li, Boyu Chen, Wenlian Lu

The performance of federated learning in neural networks is generally influenced by the heterogeneity of the data distribution. For a well-performing global model, taking a weighted average of the local models, as done by most existing federated learning algorithms, may not guarantee consistency with local models in the space of neural network maps. In this paper, we propose a novel framework of federated learning equipped with the process of decentralized knowledge distillation (FedDKD) (i.e., without data on the server). The FedDKD introduces a module of decentralized knowledge distillation (DKD) to distill the knowledge of the local models to train the global model by approaching the neural network map average based on the metric of divergence defined in the loss function, other than only averaging parameters as done in literature. Numeric experiments on various heterogeneous datasets reveal that FedDKD outperforms the state-of-the-art methods with more efficient communication and training in a few DKD steps, especially on some extremely heterogeneous datasets.

Boyu Chen, Yu Qiao, Yali Wang

Tiny Actions Challenge focuses on understanding human activities in real-world surveillance. Basically, there are two main difficulties for activity recognition in this scenario. First, human activities are often recorded at a distance, and appear in a small resolution without much discriminative clue. Second, these activities are naturally distributed in a long-tailed way. It is hard to alleviate data bias for such heavy category imbalance. To tackle these problems, we propose a comprehensive recognition solution in this paper. First, we train video backbones with data balance, in order to alleviate overfitting in the challenge benchmark. Second, we design a dual-resolution distillation framework, which can effectively guide low-resolution action recognition by super-resolution knowledge. Finally, we apply model en-semble with post-processing, which can further boost per-formance on the long-tailed categories. Our solution ranks Top-1 on the leaderboard.

Peike Li, Boyu Chen, Yao Yao et al.

Music generation has attracted growing interest with the advancement of deep generative models. However, generating music conditioned on textual descriptions, known as text-to-music, remains challenging due to the complexity of musical structures and high sampling rate requirements. Despite the task's significance, prevailing generative models exhibit limitations in music quality, computational efficiency, and generalization. This paper introduces JEN-1, a universal high-fidelity model for text-to-music generation. JEN-1 is a diffusion model incorporating both autoregressive and non-autoregressive training. Through in-context learning, JEN-1 performs various generation tasks including text-guided music generation, music inpainting, and continuation. Evaluations demonstrate JEN-1's superior performance over state-of-the-art methods in text-music alignment and music quality while maintaining computational efficiency. Our demos are available at https://jenmusic.ai/audio-demos

Ailin Huang, Ang Li, Aobo Kong et al.

We introduce Step 3.5 Flash, a sparse Mixture-of-Experts (MoE) model that bridges frontier-level agentic intelligence and computational efficiency. We focus on what matters most when building agents: sharp reasoning and fast, reliable execution. Step 3.5 Flash pairs a 196B-parameter foundation with 11B active parameters for efficient inference. It is optimized with interleaved 3:1 sliding-window/full attention and Multi-Token Prediction (MTP-3) to reduce the latency and cost of multi-round agentic interactions. To reach frontier-level intelligence, we design a scalable reinforcement learning framework that combines verifiable signals with preference feedback, while remaining stable under large-scale off-policy training, enabling consistent self-improvement across mathematics, code, and tool use. Step 3.5 Flash demonstrates strong performance across agent, coding, and math tasks, achieving 85.4% on IMO-AnswerBench, 86.4% on LiveCodeBench-v6 (2024.08-2025.05), 88.2% on tau2-Bench, 69.0% on BrowseComp (with context management), and 51.0% on Terminal-Bench 2.0, comparable to frontier models such as GPT-5.2 xHigh and Gemini 3.0 Pro. By redefining the efficiency frontier, Step 3.5 Flash provides a high-density foundation for deploying sophisticated agents in real-world industrial environments.

Jiajun Zhang, Boyu Chen, Zhilong Ji et al.

This paper describes the approach we have taken in the challenge. We still adopted the two-stage scheme same as the last champion, that is, detection first and segmentation followed. We trained more powerful detector and segmentor separately. Besides, we also perform pseudo-label training on the test set, based on student-teacher framework and end-to-end transformer based object detection. The method ranks first on the 2nd Unidentified Video Objects (UVO) challenge, achieving AR@100 of 46.8, 64.7 and 32.2 in the limited data frame track, unlimited data frame track and video track respectively.

Yi Yu, Wenlian Lu, Boyu Chen

We propose theoretical analyses of a modified natural gradient descent method in the neural network function space based on the eigendecompositions of neural tangent kernel and Fisher information matrix. We firstly present analytical expression for the function learned by this modified natural gradient under the assumptions of Gaussian distribution and infinite width limit. Thus, we explicitly derive the generalization error of the learned neural network function using theoretical methods from eigendecomposition and statistics theory. By decomposing of the total generalization error attributed to different eigenspace of the kernel in function space, we propose a criterion for balancing the errors stemming from training set and the distribution discrepancy between the training set and the true data. Through this approach, we establish that modifying the training direction of the neural network in function space leads to a reduction in the total generalization error. Furthermore, We demonstrate that this theoretical framework is capable to explain many existing results of generalization enhancing methods. These theoretical results are also illustrated by numerical examples on synthetic data.

Boyu Chen, Hanxuan Chen, Jiao He et al.

We present a simple yet novel parameterized form of linear mapping to achieves remarkable network compression performance: a pseudo SVD called Ternary SVD (TSVD). Unlike vanilla SVD, TSVD limits the $U$ and $V$ matrices in SVD to ternary matrices form in $\{\pm 1, 0\}$. This means that instead of using the expensive multiplication instructions, TSVD only requires addition instructions when computing $U(\cdot)$ and $V(\cdot)$. We provide direct and training transition algorithms for TSVD like Post Training Quantization and Quantization Aware Training respectively. Additionally, we analyze the convergence of the direct transition algorithms in theory. In experiments, we demonstrate that TSVD can achieve state-of-the-art network compression performance in various types of networks and tasks, including current baseline models such as ConvNext, Swim, BERT, and large language model like OPT.

Boyu Chen, Zirui Guo, Zidan Yang et al.

Retrieval-augmented generation (RAG) improves the response quality of large language models (LLMs) by retrieving knowledge from external databases. Typical RAG approaches split the text database into chunks, organizing them in a flat structure for efficient searches. To better capture the inherent dependencies and structured relationships across the text database, researchers propose to organize textual information into an indexing graph, known asgraph-based RAG. However, we argue that the limitation of current graph-based RAG methods lies in the redundancy of the retrieved information, rather than its insufficiency. Moreover, previous methods use a flat structure to organize retrieved information within the prompts, leading to suboptimal performance. To overcome these limitations, we propose PathRAG, which retrieves key relational paths from the indexing graph, and converts these paths into textual form for prompting LLMs. Specifically, PathRAG effectively reduces redundant information with flow-based pruning, while guiding LLMs to generate more logical and coherent responses with path-based prompting. Experimental results show that PathRAG consistently outperforms state-of-the-art baselines across six datasets and five evaluation dimensions. The code is available at the following link: https://github.com/BUPT-GAMMA/PathRAG

Boyu Chen, Zhengrong Yue, Siran Chen et al. · tsinghua

Existing MLLMs encounter significant challenges in modeling the temporal context within long videos. Currently, mainstream Agent-based methods use external tools to assist a single MLLM in answering long video questions. Despite such tool-based support, a solitary MLLM still offers only a partial understanding of long videos, resulting in limited performance. In order to better address long video tasks, we introduce LVAgent, the first framework enabling multi-round dynamic collaboration of MLLM agents in long video understanding. Our method consists of four key steps: 1) Selection: We pre-select appropriate agents from the model library to form optimal agent teams based on different tasks. 2) Perception: We design an effective retrieval scheme for long videos to improve the coverage of critical temporal segments while maintaining computational efficiency. 3) Action: Agents answer long video questions and exchange reasons. 4) Reflection: We evaluate each agent's performance in each round of discussion and optimize the agent team for dynamic collaboration. The agents iteratively refine their answers by multi-round dynamical collaboration of MLLM agents. LVAgent is the first agent system method that outperforms all closed-source models (like GPT-4o) and open-source models (like InternVL-2.5 and Qwen2-VL) in the long video understanding tasks. Our LVAgent achieves an accuracy of 80\% on four mainstream long video understanding tasks. Notably, LVAgent improves accuracy by 13.3\% on LongVideoBench. Code is available at https://github.com/64327069/LVAgent.

Yao Yao, Peike Li, Boyu Chen et al.

With rapid advances in generative artificial intelligence, the text-to-music synthesis task has emerged as a promising direction for music generation. Nevertheless, achieving precise control over multi-track generation remains an open challenge. While existing models excel in directly generating multi-track mix, their limitations become evident when it comes to composing individual tracks and integrating them in a controllable manner. This departure from the typical workflows of professional composers hinders the ability to refine details in specific tracks. To address this gap, we propose JEN-1 Composer, a unified framework designed to efficiently model marginal, conditional, and joint distributions over multi-track music using a single model. Building upon an audio latent diffusion model, JEN-1 Composer extends the versatility of multi-track music generation. We introduce a progressive curriculum training strategy, which gradually escalates the difficulty of training tasks while ensuring the model's generalization ability and facilitating smooth transitions between different scenarios. During inference, users can iteratively generate and select music tracks, thus incrementally composing entire musical pieces in accordance with the Human-AI co-composition workflow. Our approach demonstrates state-of-the-art performance in controllable and high-fidelity multi-track music synthesis, marking a significant advancement in interactive AI-assisted music creation. Our demo pages are available at www.jenmusic.ai/research.

Boyu Chen, Junjie Liu, Zhu Li et al.

Probability of necessity and sufficiency (PNS) measures the likelihood of a feature set being both necessary and sufficient for predicting an outcome. It has proven effective in guiding representation learning for unimodal data, enhancing both predictive performance and model robustness. Despite these benefits, extending PNS to multimodal settings remains unexplored. This extension presents unique challenges, as the conditions for PNS estimation, exogeneity and monotonicity, need to be reconsidered in a multimodal context. We address these challenges by first conceptualizing multimodal representations as comprising modality-invariant and modality-specific components. We then analyze how to compute PNS for each component while ensuring non-trivial PNS estimation. Based on these analyses, we formulate tractable optimization objectives that enable multimodal models to learn high-PNS representations. Experiments demonstrate the effectiveness of our method on both synthetic and real-world data.

Boyu Chen, Tai Guo, Weiyu Cui et al.

Multimodal retrieval models are becoming increasingly important in scenarios such as food delivery, where rich multimodal features can meet diverse user needs and enable precise retrieval. Mainstream approaches typically employ a dual-tower architecture between queries and items, and perform joint optimization of intra-tower and inter-tower tasks. However, we observe that joint optimization often leads to certain modalities dominating the training process, while other modalities are neglected. In addition, inconsistent training speeds across modalities can easily result in the one-epoch problem. To address these challenges, we propose a staged pretraining strategy, which guides the model to focus on specialized tasks at each stage, enabling it to effectively attend to and utilize multimodal features, and allowing flexible control over the training process at each stage to avoid the one-epoch problem. Furthermore, to better utilize the semantic IDs that compress high-dimensional multimodal embeddings, we design both generative and discriminative tasks to help the model understand the associations between SIDs, queries, and item features, thereby improving overall performance. Extensive experiments on large-scale real-world Meituan data demonstrate that our method achieves improvements of 3.80%, 2.64%, and 2.17% on R@5, R@10, and R@20, and 5.10%, 4.22%, and 2.09% on N@5, N@10, and N@20 compared to mainstream baselines. Online A/B testing on the Meituan platform shows that our approach achieves a 1.12% increase in revenue and a 1.02% increase in click-through rate, validating the effectiveness and superiority of our method in practical applications.

Boyu Chen, Ameenat L. Solebo, Daqian Shi et al.

Anterior Segment Optical Coherence Tomography (AS-OCT) is an emerging imaging technique with great potential for diagnosing anterior uveitis, a vision-threatening ocular inflammatory condition. A hallmark of this condition is the presence of inflammatory cells in the eye's anterior chamber, and detecting these cells using AS-OCT images has attracted research interest. While recent efforts aim to replace manual cell detection with automated computer vision approaches, detecting extremely small (minuscule) objects in high-resolution images, such as AS-OCT, poses substantial challenges: (1) each cell appears as a minuscule particle, representing less than 0.005\% of the image, making the detection difficult, and (2) OCT imaging introduces pixel-level noise that can be mistaken for cells, leading to false positive detections. To overcome these challenges, we propose a minuscule cell detection framework through a progressive field-of-view focusing strategy. This strategy systematically refines the detection scope from the whole image to a target region where cells are likely to be present, and further to minuscule regions potentially containing individual cells. Our framework consists of two modules. First, a Field-of-Focus module uses a vision foundation model to segment the target region. Subsequently, a Fine-grained Object Detection module introduces a specialized Minuscule Region Proposal followed by a Spatial Attention Network to distinguish individual cells from noise within the segmented region. Experimental results demonstrate that our framework outperforms state-of-the-art methods for cell detection, providing enhanced efficacy for clinical applications. Our code is publicly available at: https://github.com/joeybyc/MCD.

Boyu Chen, Ameenat L. Solebo, Paul Taylor

Anterior uveitis, a common form of eye inflammation, can lead to permanent vision loss if not promptly diagnosed. Monitoring this condition involves quantifying inflammatory cells in the anterior chamber (AC) of the eye, which can be captured using Anterior Segment Optical Coherence Tomography (AS-OCT). However, manually identifying cells in AS-OCT images is time-consuming and subjective. Moreover, existing automated approaches may have limitations in both the effectiveness of detecting cells and the reliability of their detection results. To address these challenges, we propose an automated framework to detect cells in the AS-OCT images. This framework consists of a zero-shot chamber segmentation module and a cell detection module. The first module segments the AC area in the image without requiring human-annotated training data. Subsequently, the second module identifies individual cells within the segmented AC region. Through experiments, our framework demonstrates superior performance compared to current state-of-the-art methods for both AC segmentation and cell detection tasks. Notably, we find that previous cell detection approaches could suffer from low recall, potentially overlooking a significant number of cells. In contrast, our framework offers an improved solution, which could benefit the diagnosis and study of anterior uveitis. Our code for cell detection is publicly available at: https://github.com/joeybyc/cell_detection.

Xiang Chen, Hao Li, Jiangxin Dong et al.

This paper presents a review for the LoViF Challenge on Real-World All-in-One Image Restoration. The challenge aimed to advance research on real-world all-in-one image restoration under diverse real-world degradation conditions, including blur, low-light, haze, rain, and snow. It provided a unified benchmark to evaluate the robustness and generalization ability of restoration models across multiple degradation categories within a common framework. The competition attracted 124 registered participants and received 9 valid final submissions with corresponding fact sheets, significantly contributing to the progress of real-world all-in-one image restoration. This report provides a detailed analysis of the submitted methods and corresponding results, emphasizing recent progress in unified real-world image restoration. The analysis highlights effective approaches and establishes a benchmark for future research in real-world low-level vision.

Boyu Chen, Yi Chen, Lu Qiu et al.

Scaling humanoid foundation models is bottlenecked by the scarcity of robotic data. While massive egocentric human data offers a scalable alternative, bridging the cross-embodiment chasm remains a fundamental challenge due to kinematic mismatches. We introduce UniT (Unified Latent Action Tokenizer via Visual Anchoring), a framework that establishes a unified physical language for human-to-humanoid transfer. Grounded in the philosophy that heterogeneous kinematics share universal visual consequences, UniT employs a tri-branch cross-reconstruction mechanism: actions predict vision to anchor kinematics to physical outcomes, while vision reconstructs actions to filter out irrelevant visual confounders. Concurrently, a fusion branch synergies these purified modalities into a shared discrete latent space of embodiment-agnostic physical intents. We validate UniT across two paradigms: 1) Policy Learning (VLA-UniT): By predicting these unified tokens, it effectively leverages diverse human data to achieve state-of-the-art data efficiency and robust out-of-distribution (OOD) generalization on both humanoid simulation benchmark and real-world deployments, notably demonstrating zero-shot task transfer. 2) World Modeling (WM-UniT): By aligning cross-embodiment dynamics via unified tokens as conditions, it realizes direct human-to-humanoid action transfer. This alignment ensures that human data seamlessly translates into enhanced action controllability for humanoid video generation. Ultimately, by inducing a highly aligned cross-embodiment representation (empirically verified by t-SNE visualizations revealing the convergence of human and humanoid features into a shared manifold), UniT offers a scalable path to distill vast human knowledge into general-purpose humanoid capabilities.

StepFun, Bin Wang, Bojun Wang et al.

Large language models (LLMs) face low hardware efficiency during decoding, especially for long-context reasoning tasks. This paper introduces Step-3, a 321B-parameter VLM with hardware-aware model-system co-design optimized for minimizing decoding costs. Step-3 innovates in two key dimensions: (1) A novel Multi-Matrix Factorization Attention (MFA) mechanism that significantly reduces both KV cache size and computation while maintaining high attention expressiveness, and (2) Attention-FFN Disaggregation (AFD), a distributed inference system that decouples attention and Feed-Forward Network (FFN) layers into specialized subsystems. This co-design achieves unprecedented cost efficiency: Step-3 significantly reduces theoretical decoding costs compared with models like DeepSeek-V3 and Qwen3 MoE 235B, with the gains widening at longer context. Step-3 achieves low cost while activating 38B parameters per token (more than DeepSeek-V3 and Qwen3 MoE 235B), demonstrating that hardware-aligned attention arithmetic intensity, MoE sparsity, and AFD are critical to cost-effectiveness. We perform a head-to-head comparison with DeepSeek-V3 in its favorable scenarios. Our implementation on Hopper GPUs achieves a decoding throughput of up to 4,039 tokens per second per GPU under 50ms TPOT SLA (4K context, FP8, no MTP). It is higher than DeepSeek-V3's 2,324 in the same setup and sets a new Pareto frontier for LLM decoding.

Boyu Chen, Siran Chen, Kunchang Li et al.

Open-world video recognition is challenging since traditional networks are not generalized well on complex environment variations. Alternatively, foundation models with rich knowledge have recently shown their generalization power. However, how to apply such knowledge has not been fully explored for open-world video recognition. To this end, we propose a generic knowledge transfer pipeline, which progressively exploits and integrates external multimodal knowledge from foundation models to boost open-world video recognition. We name it PCA, based on three stages of Percept, Chat, and Adapt. First, we perform Percept process to reduce the video domain gap and obtain external visual knowledge. Second, we generate rich linguistic semantics as external textual knowledge in Chat stage. Finally, we blend external multimodal knowledge in Adapt stage, by inserting multimodal knowledge adaptation modules into networks. We conduct extensive experiments on three challenging open-world video benchmarks, i.e., TinyVIRAT, ARID, and QV-Pipe. Our approach achieves state-of-the-art performance on all three datasets.

Boyu Chen, Siran Chen, Kunchang Li et al.

Video understanding has been considered as one critical step towards world modeling, which is an important long-term problem in AI research. Recently, multi-modal foundation models have shown such potential via large-scale pretraining. However, these models simply align encoders of different modalities via contrastive learning, while lacking deeper multi-modal interactions, which is critical for understanding complex target movements with diversified video scenes. To fill this gap, we propose a unified Super Encoding Network (SEN) for video understanding, which builds up such distinct interactions through recursive association of multi-modal encoders in the foundation models. Specifically, we creatively treat those well-trained encoders as "super neurons" in our SEN. Via designing a Recursive Association (RA) block, we progressively fuse multi-modalities with the input video, based on knowledge integrating, distributing, and prompting of super neurons in a recursive manner. In this way, our SEN can effectively encode deeper multi-modal interactions, for prompting various video understanding tasks in downstream. Extensive experiments show that, our SEN can remarkably boost the four most representative video tasks, including tracking, recognition, chatting, and editing, e.g., for pixel-level tracking, the average jaccard index improves 2.7%, temporal coherence(TC) drops 8.8% compared to the popular CaDeX++ approach. For one-shot video editing, textual alignment improves 6.4%, and frame consistency increases 4.1% compared to the popular TuneA-Video approach.

Zhiying Li, Yeying Jin, Fan Shen et al.

Expressive human pose and shape estimation (EHPS) is crucial for digital human generation, especially in applications like live streaming. While existing research primarily focuses on reducing estimation errors, it largely neglects robustness and security aspects, leaving these systems vulnerable to adversarial attacks. To address this significant challenge, we propose the \textbf{Tangible Attack (TBA)}, a novel framework designed to generate adversarial examples capable of effectively compromising any digital human generation model. Our approach introduces a \textbf{Dual Heterogeneous Noise Generator (DHNG)}, which leverages Variational Autoencoders (VAE) and ControlNet to produce diverse, targeted noise tailored to the original image features. Additionally, we design a custom \textbf{adversarial loss function} to optimize the noise, ensuring both high controllability and potent disruption. By iteratively refining the adversarial sample through multi-gradient signals from both the noise and the state-of-the-art EHPS model, TBA substantially improves the effectiveness of adversarial attacks. Extensive experiments demonstrate TBA's superiority, achieving a remarkable 41.0\% increase in estimation error, with an average improvement of approximately 17.0\%. These findings expose significant security vulnerabilities in current EHPS models and highlight the need for stronger defenses in digital human generation systems.

Boyu Chen, Siran Chen, Zhengrong Yue et al.

User feedback is critical for refining recommendation systems, yet explicit feedback (e.g., likes or dislikes) remains scarce in practice. As a more feasible alternative, inferring user preferences from massive implicit feedback has shown great potential (e.g., a user quickly skipping a recommended video usually indicates disinterest). Unfortunately, implicit feedback is often noisy: a user might skip a video due to accidental clicks or other reasons, rather than disliking it. Such noise can easily misjudge user interests, thereby undermining recommendation performance. To address this issue, we propose a novel Group-aware User Behavior Simulation (G-UBS) paradigm, which leverages contextual guidance from relevant user groups, enabling robust and in-depth interpretation of implicit feedback for individual users. Specifically, G-UBS operates via two key agents. First, the User Group Manager (UGM) effectively clusters users to generate group profiles utilizing a ``summarize-cluster-reflect" workflow based on LLMs. Second, the User Feedback Modeler (UFM) employs an innovative group-aware reinforcement learning approach, where each user is guided by the associated group profiles during the reinforcement learning process, allowing UFM to robustly and deeply examine the reasons behind implicit feedback. To assess our G-UBS paradigm, we have constructed a Video Recommendation benchmark with Implicit Feedback (IF-VR). To the best of our knowledge, this is the first multi-modal benchmark for implicit feedback evaluation in video recommendation, encompassing 15k users, 25k videos, and 933k interaction records with implicit feedback. Extensive experiments on IF-VR demonstrate that G-UBS significantly outperforms mainstream LLMs and MLLMs, with a 4.0% higher proportion of videos achieving a play rate > 30% and 14.9% higher reasoning accuracy on IF-VR.

Zhenkun Shi, Jun Zhu, Dehang Wang et al.

A key challenge in enzyme annotation is identifying the biochemical reactions catalyzed by proteins. Most existing methods rely on Enzyme Commission (EC) numbers as intermediaries: they first predict an EC number and then retrieve the associated reactions. This indirect strategy introduces ambiguity due to the complex many-to-many mappings among proteins, EC numbers, and reactions, and is further complicated by frequent updates to EC numbers and inconsistencies across databases. To address these challenges, we present RXNRECer, a transformer-based ensemble framework that directly predicts enzyme-catalyzed reactions without relying on EC numbers. It integrates protein language modeling and active learning to capture both high-level sequence semantics and fine-grained transformation patterns. Evaluations on curated cross-validation and temporal test sets demonstrate consistent improvements over six EC-based baselines, with gains of 16.54% in F1 score and 15.43% in accuracy. Beyond accuracy gains, the framework offers clear advantages for downstream applications, including scalable proteome-wide reaction annotation, enhanced specificity in refining generic reaction schemas, systematic annotation of previously uncurated proteins, and reliable identification of enzyme promiscuity. By incorporating large language models, it also provides interpretable rationales for predictions. These capabilities make RXNRECer a robust and versatile solution for EC-free, fine-grained enzyme function prediction, with potential applications across multiple areas of enzyme research and industrial applications.

Boyu Chen, Zikang Wang, Zhengrong Yue et al.

By leveraging tool-augmented Multimodal Large Language Models (MLLMs), multi-agent frameworks are driving progress in video understanding. However, most of them adopt static and non-learnable tool invocation mechanisms, which limit the discovery of diverse clues essential for robust perception and reasoning regarding temporally or spatially complex videos. To address this challenge, we propose a novel Multi-agent system for video understanding, namely VideoChat-M1. Instead of using a single or fixed policy, VideoChat-M1 adopts a distinct Collaborative Policy Planning (CPP) paradigm with multiple policy agents, which comprises three key processes. (1) Policy Generation: Each agent generates its unique tool invocation policy tailored to the user's query; (2) Policy Execution: Each agent sequentially invokes relevant tools to execute its policy and explore the video content; (3) Policy Communication: During the intermediate stages of policy execution, agents interact with one another to update their respective policies. Through this collaborative framework, all agents work in tandem, dynamically refining their preferred policies based on contextual insights from peers to effectively respond to the user's query. Moreover, we equip our CPP paradigm with a concise Multi-Agent Reinforcement Learning (MARL) method. Consequently, the team of policy agents can be jointly optimized to enhance VideoChat-M1's performance, guided by both the final answer reward and intermediate collaborative process feedback. Extensive experiments demonstrate that VideoChat-M1 achieves SOTA performance across eight benchmarks spanning four tasks. Notably, on LongVideoBench, our method outperforms the SOTA model Gemini 2.5 pro by 3.6% and GPT-4o by 15.6%.

Zhengrong Yue, Haiyu Zhang, Xiangyu Zeng et al.

Tokenizer is a crucial component for both visual understanding and generation. To advance toward the ultimate goal of universal modeling, recent research has focused on developing a unified tokenizer. However, existing tokenizers face a significant performance trade-off between understanding and generation, stemming from the inherent conflict between high-level semantic abstraction and low-level pixel reconstruction. To tackle this challenge, we propose a generic and unified tokenizer, namely UniFlow, by flexibly adapting any visual encoder with a concise reconstruction decoder. Specifically, we introduce layer-wise adaptive self-distillation applied to the well-pretrained visual encoders, which enables UniFlow to simultaneously inherit the strong semantic features for visual understanding and flexibly adapt to model fine-grained details for visual generation. Moreover, we propose a lightweight patch-wise pixel flow decoder, which efficiently achieves high-fidelity pixel reconstruction by modeling a conditional flow from the noisy state back to the patch-wise pixel domain. By leveraging the semantic features as visual conditions for the decoder, we effectively alleviate the training conflicts between understanding and generation. Furthermore, the patch-wise learning strategy simplifies the data distribution, thereby improving training efficiency. Extensive experiments across 13 challenging benchmarks spanning 7 widely studied visual understanding and generation tasks demonstrate that UniFlow achieves a win-win outcome. For instance, our 7B UniFlow-XL not only surpasses the 14B TokenFlow-XL by 7.75% on average understanding benchmarks, but also achieves competitive results in both visual reconstruction and generation, surpassing UniTok by 0.15 in rFID and 0.09 in gFID (without guidance), respectively.

Yi Wang, Lihai Yang, Boyu Chen et al.

Large language model agents increasingly depend on memory to sustain long horizon interaction, but existing frameworks remain limited. Most expose only a few basic primitives such as encode, retrieve, and delete, while higher order operations like merge, promote, demote, split, lock, and expire are missing or inconsistently supported. Moreover, there is no formal and executable specification for memory commands, leaving scope and lifecycle rules implicit and causing unpredictable behavior across systems. We introduce Text2Mem, a unified memory operation language that provides a standardized pathway from natural language to reliable execution. Text2Mem defines a compact yet expressive operation set aligned with encoding, storage, and retrieval. Each instruction is represented as a JSON based schema instance with required fields and semantic invariants, which a parser transforms into typed operation objects with normalized parameters. A validator ensures correctness before execution, while adapters map typed objects either to a SQL prototype backend or to real memory frameworks. Model based services such as embeddings or summarization are integrated when required. All results are returned through a unified execution contract. This design ensures safety, determinism, and portability across heterogeneous backends. We also outline Text2Mem Bench, a planned benchmark that separates schema generation from backend execution to enable systematic evaluation. Together, these components establish the first standardized foundation for memory control in agents.

Xinlong Zhai, Chunchen Wang, Ruijia Wang et al.

Drug-target interaction prediction (DTI) is essential in various applications including drug discovery and clinical application. There are two perspectives of input data widely used in DTI prediction: Intrinsic data represents how drugs or targets are constructed, and extrinsic data represents how drugs or targets are related to other biological entities. However, any of the two perspectives of input data can be scarce for some drugs or targets, especially for those unpopular or newly discovered. Furthermore, ground-truth labels for specific interaction types can also be scarce. Therefore, we propose the first method to tackle DTI prediction under input data and/or label scarcity. To make our model functional when only one perspective of input data is available, we design two separate experts to process intrinsic and extrinsic data respectively and fuse them adaptively according to different samples. Furthermore, to make the two perspectives complement each other and remedy label scarcity, two experts synergize with each other in a mutually supervised way to exploit the enormous unlabeled data. Extensive experiments on 3 real-world datasets under different extents of input data scarcity and/or label scarcity demonstrate our model outperforms states of the art significantly and steadily, with a maximum improvement of 53.53%. We also test our model without any data scarcity and it still outperforms current methods.

Boyu Chen, Peike Li, Yao Yao et al.

Large models for text-to-music generation have achieved significant progress, facilitating the creation of high-quality and varied musical compositions from provided text prompts. However, input text prompts may not precisely capture user requirements, particularly when the objective is to generate music that embodies a specific concept derived from a designated reference collection. In this paper, we propose a novel method for customized text-to-music generation, which can capture the concept from a two-minute reference music and generate a new piece of music conforming to the concept. We achieve this by fine-tuning a pretrained text-to-music model using the reference music. However, directly fine-tuning all parameters leads to overfitting issues. To address this problem, we propose a Pivotal Parameters Tuning method that enables the model to assimilate the new concept while preserving its original generative capabilities. Additionally, we identify a potential concept conflict when introducing multiple concepts into the pretrained model. We present a concept enhancement strategy to distinguish multiple concepts, enabling the fine-tuned model to generate music incorporating either individual or multiple concepts simultaneously. Since we are the first to work on the customized music generation task, we also introduce a new dataset and evaluation protocol for the new task. Our proposed Jen1-DreamStyler outperforms several baselines in both qualitative and quantitative evaluations. Demos will be available at https://www.jenmusic.ai/research#DreamStyler.

Ru Peng, Nankai Lin, Yi Fang et al.

Syntax knowledge contributes its powerful strength in Neural machine translation (NMT) tasks. Early NMT works supposed that syntax details can be automatically learned from numerous texts via attention networks. However, succeeding researches pointed out that limited by the uncontrolled nature of attention computation, the NMT model requires an external syntax to capture the deep syntactic awareness. Although existing syntax-aware NMT methods have born great fruits in combining syntax, the additional workloads they introduced render the model heavy and slow. Particularly, these efforts scarcely involve the Transformer-based NMT and modify its core self-attention network (SAN). To this end, we propose a parameter-free, Dependency-scaled Self-Attention Network (Deps-SAN) for syntax-aware Transformer-based NMT. A quantified matrix of dependency closeness between tokens is constructed to impose explicit syntactic constraints into the SAN for learning syntactic details and dispelling the dispersion of attention distributions. Two knowledge sparsing techniques are further integrated to avoid the model overfitting the dependency noises introduced by the external parser. Experiments and analyses on IWSLT14 German-to-English and WMT16 German-to-English benchmark NMT tasks verify the effectiveness of our approach.

Boyu Chen, Peixia Li, Baopu Li et al.

We present BN-NAS, neural architecture search with Batch Normalization (BN-NAS), to accelerate neural architecture search (NAS). BN-NAS can significantly reduce the time required by model training and evaluation in NAS. Specifically, for fast evaluation, we propose a BN-based indicator for predicting subnet performance at a very early training stage. The BN-based indicator further facilitates us to improve the training efficiency by only training the BN parameters during the supernet training. This is based on our observation that training the whole supernet is not necessary while training only BN parameters accelerates network convergence for network architecture search. Extensive experiments show that our method can significantly shorten the time of training supernet by more than 10 times and shorten the time of evaluating subnets by more than 600,000 times without losing accuracy.

Boyu Chen, Peixia Li, Baopu Li et al.

In this paper, we observe two levels of redundancies when applying vision transformers (ViT) for image recognition. First, fixing the number of tokens through the whole network produces redundant features at the spatial level. Second, the attention maps among different transformer layers are redundant. Based on the observations above, we propose a PSViT: a ViT with token Pooling and attention Sharing to reduce the redundancy, effectively enhancing the feature representation ability, and achieving a better speed-accuracy trade-off. Specifically, in our PSViT, token pooling can be defined as the operation that decreases the number of tokens at the spatial level. Besides, attention sharing will be built between the neighboring transformer layers for reusing the attention maps having a strong correlation among adjacent layers. Then, a compact set of the possible combinations for different token pooling and attention sharing mechanisms are constructed. Based on the proposed compact set, the number of tokens in each layer and the choices of layers sharing attention can be treated as hyper-parameters that are learned from data automatically. Experimental results show that the proposed scheme can achieve up to 6.6% accuracy improvement in ImageNet classification compared with the DeiT.

Boyu Chen, Peixia Li, Chuming Li et al.

We introduce the first Neural Architecture Search (NAS) method to find a better transformer architecture for image recognition. Recently, transformers without CNN-based backbones are found to achieve impressive performance for image recognition. However, the transformer is designed for NLP tasks and thus could be sub-optimal when directly used for image recognition. In order to improve the visual representation ability for transformers, we propose a new search space and searching algorithm. Specifically, we introduce a locality module that models the local correlations in images explicitly with fewer computational cost. With the locality module, our search space is defined to let the search algorithm freely trade off between global and local information as well as optimizing the low-level design choice in each module. To tackle the problem caused by huge search space, a hierarchical neural architecture search method is proposed to search the optimal vision transformer from two levels separately with the evolutionary algorithm. Extensive experiments on the ImageNet dataset demonstrate that our method can find more discriminative and efficient transformer variants than the ResNet family (e.g., ResNet101) and the baseline ViT for image classification.

Xin Cai, Boyu Chen, Jiabei Zeng et al.

This paper presents a method for gaze estimation according to face images. We train several gaze estimators adopting four different network architectures, including an architecture designed for gaze estimation (i.e.,iTracker-MHSA) and three originally designed for general computer vision tasks(i.e., BoTNet, HRNet, ResNeSt). Then, we select the best six estimators and ensemble their predictions through a linear combination. The method ranks the first on the leader-board of ETH-XGaze Competition, achieving an average angular error of $3.11^{\circ}$ on the ETH-XGaze test set.

Yuanzheng Ci, Chen Lin, Ming Sun et al.

The automation of neural architecture design has been a coveted alternative to human experts. Recent works have small search space, which is easier to optimize but has a limited upper bound of the optimal solution. Extra human design is needed for those methods to propose a more suitable space with respect to the specific task and algorithm capacity. To further enhance the degree of automation for neural architecture search, we present a Neural Search-space Evolution (NSE) scheme that iteratively amplifies the results from the previous effort by maintaining an optimized search space subset. This design minimizes the necessity of a well-designed search space. We further extend the flexibility of obtainable architectures by introducing a learnable multi-branch setting. By employing the proposed method, a consistent performance gain is achieved during a progressive search over upcoming search spaces. We achieve 77.3% top-1 retrain accuracy on ImageNet with 333M FLOPs, which yielded a state-of-the-art performance among previous auto-generated architectures that do not involve knowledge distillation or weight pruning. When the latency constraint is adopted, our result also performs better than the previous best-performing mobile models with a 77.9% Top-1 retrain accuracy.

Peixia Li, Boyu Chen, Wanli Ouyang et al.

The fully-convolutional siamese network based on template matching has shown great potentials in visual tracking. During testing, the template is fixed with the initial target feature and the performance totally relies on the general matching ability of the siamese network. However, this manner cannot capture the temporal variations of targets or background clutter. In this work, we propose a novel gradient-guided network to exploit the discriminative information in gradients and update the template in the siamese network through feed-forward and backward operations. Our algorithm performs feed-forward and backward operations to exploit the discriminative informaiton in gradients and capture the core attention of the target. To be specific, the algorithm can utilize the information from the gradient to update the template in the current frame. In addition, a template generalization training method is proposed to better use gradient information and avoid overfitting. To our knowledge, this work is the first attempt to exploit the information in the gradient for template update in siamese-based trackers. Extensive experiments on recent benchmarks demonstrate that our method achieves better performance than other state-of-the-art trackers.

Yaoting Huang, Boyu Chen, Wenlian Lu et al.

Due to the complication of the presented problem, intelligent algorithms show great power to solve the parts logistics optimization problem related to the vehicle routing problem (VRP). However, most of the existing research to VRP are incomprehensive and failed to solve a real-work parts logistics problem. In this work, towards SAIC logistics problem, we propose a systematic solution to this 2-Dimensional Loading Capacitated Multi-Depot Heterogeneous VRP with Time Windows by integrating diverse types of intelligent algorithms, including, a heuristic algorithm to initialize feasible logistics planning schemes by imitating manual planning, the core Tabu Search algorithm for global optimization, accelerated by a novel bundle technique, heuristically algorithms for routing, packing and queuing associated, and a heuristic post-optimization process to promote the optimal solution. Based on these algorithms, the SAIC Motor has successfully established an intelligent management system to give a systematic solution for the parts logistics planning, superior than manual planning in its performance, customizability and expandability.

Boyu Chen, Wenlian Lu, Ernest Fokoue

Meta-learning is a promising method to achieve efficient training method towards deep neural net and has been attracting increases interests in recent years. But most of the current methods are still not capable to train complex neuron net model with long-time training process. In this paper, a novel second-order meta-optimizer, named Meta-learning with Hessian-Free(MLHF) approach, is proposed based on the Hessian-Free approach. Two recurrent neural networks are established to generate the damping and the precondition matrix of this Hessian-Free framework. A series of techniques to meta-train the MLHF towards stable and reinforce the meta-training of this optimizer, including the gradient calculation of $H$. Numerical experiments on deep convolution neural nets, including CUDA-convnet and ResNet18(v2), with datasets of CIFAR10 and ILSVRC2012, indicate that the MLHF shows good and continuous training performance during the whole long-time training process, i.e., both the rapid-decreasing early stage and the steadily-deceasing later stage, and so is a promising meta-learning framework towards elevating the training efficiency in real-world deep neural nets.