Peng Jiang

Yuecheng Li, Hengwei Ju, Zeyu Song et al.

Integrating large language model (LLM) representations into multimodal recommendation has shown promise, yet a fundamental challenge remains largely overlooked: the semantic heterogeneity between generative LM representations and the ID-based collaborative signals that recommendation systems rely on. Naively injecting LM features without alignment degrades recommendation performance rather than improving it. To resolve this, we propose RecGOAT, a dual-granularity semantic alignment framework built on graph neural networks and optimal transport theory. RecGOAT first enriches collaborative semantics through multimodal attentive graphs that capture item-item, user-item, and user-user relationships, initializing user representations via LLM-inferred behavioral preferences. It then aligns LM-derived modality representations with recommendation IDs at two complementary granularities: (1) instance-level alignment via cross-modal contrastive learning (CMCL), which produces discriminative per-sample representations; and (2) distribution-level alignment via optimal adaptive transport (OAT), which minimizes the 1-Wasserstein distance between ID distributions and LLM semantics to produce a unified, consistently aligned feature space. Theoretically, we prove that the unified representation achieves strictly lower target error than any single-modality representation, with the gap bounded by the Wasserstein distance and the InfoNCE loss, providing rigorous guarantees for both alignment consistency and fusion comprehensiveness. Extensive experiments on three public benchmarks demonstrate state-of-the-art performance. Deployment on a large-scale online advertising platform further validates RecGOAT's industrial scalability. Our code is available at https://github.com/6lyc/RecGOAT-LLM4Rec.

Yuecheng Li, Zeyu Song, Jing Yao et al.

Scaling recommender systems via large language models (LLMs) has become a prominent trend in the industry. However, aligning the LLM's semantic space with the recommender's ID space via post-training (e.g., SFT and RL) remains challenging. Existing LLM4Rec paradigms are bottlenecked by two main issues: (1) the difficulty of measuring and improving chain-of-thought (CoT) quality in open-domain recommendation during SFT, and (2) the neglect of the trade-off between LLM semantic rewards and recommendation preference rewards during RL alignment. Inspired by these challenges, we present Taiji, a novel LLM-as-Enhancer framework designed for industrial recommender systems. To overcome the SFT bottleneck, we utilize reverse-engineered reasoning and open-ended rejection sampling to generate high-quality, domain-specific CoT data. To resolve the RL alignment issue, we propose Pareto Optimal Policy Optimization (POPO), which adaptively adjusts cross-domain reward weights. Theoretically, it achieves an optimal trade-off between the semantic world knowledge of LLMs and the collaborative ID features representing online user preferences. Extensive offline evaluations and online A/B tests validate the effectiveness of Taiji. Deployed on Kuaishou's advertising platform since May 2026, Taiji currently serves over 400 million users daily, yielding significant commercial revenue and demonstrating its robust scalability in web-scale environments.

Zhihong Liu, Siqi Kou, Zheng Li et al.

Crafting a product display webpage from a source product image, along with layout and visual content instructions, holds significant practical value for domains such as marketing, advertising, and E-commerce. Intuitively, this task demands strict visual consistency across product displays and high-fidelity instruction following to jointly generate renderable HTML code. These requirements on controllability and instruction-following are closely aligned with the core features of advanced multimodal generative models, such as image editing models and unified models. To this end, this paper introduces ProductWebGen to systematically benchmark the product webpage generation capacities of these models. We organize ProductWebGen with 500 test samples covering 13 product categories; each sample consists of a source image, a visual content instruction, and a webpage instruction. The task is to generate a product showcase webpage including multiple consistent images in accordance with the source image and instructions. Given the mixed-modality input-output nature of the task, we design and systematically compare two workflows for evaluation -- one uses large language models and image editing models to separately generate HTML code and images (editing-based), while the other relies on a single UM to generate both, with image generation conditioned on the preceding multimodal context (UM-based). Empirical results show that editing-based approaches achieve leading results in webpage instruction following and content appeal, while UM-based ones may display more advantages in fulfilling visual content instructions. We also construct a supervised fine-tuning dataset, ProductWebGen-1k, with 1,000 groups of real product images and LLM-generated HTML code. We verify its effectiveness on the open-source UM BAGEL. The data and code are available at https://github.com/SJTU-DENG-Lab/ProductWebGen.

Jillur Rahman Saurav, Mohammad Sadegh Nasr, Paul Koomey et al.

Here we present a structural similarity index measure (SSIM) guided conditional Generative Adversarial Network (cGAN) that generatively performs image-to-image (i2i) synthesis to generate photo-accurate protein channels in multiplexed spatial proteomics images. This approach can be utilized to accurately generate missing spatial proteomics channels that were not included during experimental data collection either at the bench or the clinic. Experimental spatial proteomic data from the Human BioMolecular Atlas Program (HuBMAP) was used to generate spatial representations of missing proteins through a U-Net based image synthesis pipeline. HuBMAP channels were hierarchically clustered by the (SSIM) as a heuristic to obtain the minimal set needed to recapitulate the underlying biology represented by the spatial landscape of proteins. We subsequently prove that our SSIM based architecture allows for scaling of generative image synthesis to slides with up to 100 channels, which is better than current state of the art algorithms which are limited to data with 11 channels. We validate these claims by generating a new experimental spatial proteomics data set from human lung adenocarcinoma tissue sections and show that a model trained on HuBMAP can accurately synthesize channels from our new data set. The ability to recapitulate experimental data from sparsely stained multiplexed histological slides containing spatial proteomic will have tremendous impact on medical diagnostics and drug development, and also raises important questions on the medical ethics of utilizing data produced by generative image synthesis in the clinical setting. The algorithm that we present in this paper will allow researchers and clinicians to save time and costs in proteomics based histological staining while also increasing the amount of data that they can generate through their experiments.

Wanqi Xue, Qingpeng Cai, Zhenghai Xue et al.

Current advances in recommender systems have been remarkably successful in optimizing immediate engagement. However, long-term user engagement, a more desirable performance metric, remains difficult to improve. Meanwhile, recent reinforcement learning (RL) algorithms have shown their effectiveness in a variety of long-term goal optimization tasks. For this reason, RL is widely considered as a promising framework for optimizing long-term user engagement in recommendation. Though promising, the application of RL heavily relies on well-designed rewards, but designing rewards related to long-term user engagement is quite difficult. To mitigate the problem, we propose a novel paradigm, recommender systems with human preferences (or Preference-based Recommender systems), which allows RL recommender systems to learn from preferences about users historical behaviors rather than explicitly defined rewards. Such preferences are easily accessible through techniques such as crowdsourcing, as they do not require any expert knowledge. With PrefRec, we can fully exploit the advantages of RL in optimizing long-term goals, while avoiding complex reward engineering. PrefRec uses the preferences to automatically train a reward function in an end-to-end manner. The reward function is then used to generate learning signals to train the recommendation policy. Furthermore, we design an effective optimization method for PrefRec, which uses an additional value function, expectile regression and reward model pre-training to improve the performance. We conduct experiments on a variety of long-term user engagement optimization tasks. The results show that PrefRec significantly outperforms previous state-of-the-art methods in all the tasks.

Yue Feng, Shuchang Liu, Zhenghai Xue et al.

Conversational recommender systems (CRSs) aim to recommend high-quality items to users through a dialogue interface. It usually contains multiple sub-tasks, such as user preference elicitation, recommendation, explanation, and item information search. To develop effective CRSs, there are some challenges: 1) how to properly manage sub-tasks; 2) how to effectively solve different sub-tasks; and 3) how to correctly generate responses that interact with users. Recently, Large Language Models (LLMs) have exhibited an unprecedented ability to reason and generate, presenting a new opportunity to develop more powerful CRSs. In this work, we propose a new LLM-based CRS, referred to as LLMCRS, to address the above challenges. For sub-task management, we leverage the reasoning ability of LLM to effectively manage sub-task. For sub-task solving, we collaborate LLM with expert models of different sub-tasks to achieve the enhanced performance. For response generation, we utilize the generation ability of LLM as a language interface to better interact with users. Specifically, LLMCRS divides the workflow into four stages: sub-task detection, model matching, sub-task execution, and response generation. LLMCRS also designs schema-based instruction, demonstration-based instruction, dynamic sub-task and model matching, and summary-based generation to instruct LLM to generate desired results in the workflow. Finally, to adapt LLM to conversational recommendations, we also propose to fine-tune LLM with reinforcement learning from CRSs performance feedback, referred to as RLPF. Experimental results on benchmark datasets show that LLMCRS with RLPF outperforms the existing methods.

Zhenghai Xue, Qingpeng Cai, Shuchang Liu et al.

In many real-world scenarios, Reinforcement Learning (RL) algorithms are trained on data with dynamics shift, i.e., with different underlying environment dynamics. A majority of current methods address such issue by training context encoders to identify environment parameters. Data with dynamics shift are separated according to their environment parameters to train the corresponding policy. However, these methods can be sample inefficient as data are used \textit{ad hoc}, and policies trained for one dynamics cannot benefit from data collected in all other environments with different dynamics. In this paper, we find that in many environments with similar structures and different dynamics, optimal policies have similar stationary state distributions. We exploit such property and learn the stationary state distribution from data with dynamics shift for efficient data reuse. Such distribution is used to regularize the policy trained in a new environment, leading to the SRPO (\textbf{S}tate \textbf{R}egularized \textbf{P}olicy \textbf{O}ptimization) algorithm. To conduct theoretical analyses, the intuition of similar environment structures is characterized by the notion of homomorphous MDPs. We then demonstrate a lower-bound performance guarantee on policies regularized by the stationary state distribution. In practice, SRPO can be an add-on module to context-based algorithms in both online and offline RL settings. Experimental results show that SRPO can make several context-based algorithms far more data efficient and significantly improve their overall performance.

Qingpeng Cai, Shuchang Liu, Xueliang Wang et al.

Recently, short video platforms have achieved rapid user growth by recommending interesting content to users. The objective of the recommendation is to optimize user retention, thereby driving the growth of DAU (Daily Active Users). Retention is a long-term feedback after multiple interactions of users and the system, and it is hard to decompose retention reward to each item or a list of items. Thus traditional point-wise and list-wise models are not able to optimize retention. In this paper, we choose reinforcement learning methods to optimize the retention as they are designed to maximize the long-term performance. We formulate the problem as an infinite-horizon request-based Markov Decision Process, and our objective is to minimize the accumulated time interval of multiple sessions, which is equal to improving the app open frequency and user retention. However, current reinforcement learning algorithms can not be directly applied in this setting due to uncertainty, bias, and long delay time incurred by the properties of user retention. We propose a novel method, dubbed RLUR, to address the aforementioned challenges. Both offline and live experiments show that RLUR can significantly improve user retention. RLUR has been fully launched in Kuaishou app for a long time, and achieves consistent performance improvement on user retention and DAU.

Qingpeng Cai, Zhenghai Xue, Chi Zhang et al.

The wide popularity of short videos on social media poses new opportunities and challenges to optimize recommender systems on the video-sharing platforms. Users sequentially interact with the system and provide complex and multi-faceted responses, including watch time and various types of interactions with multiple videos. One the one hand, the platforms aims at optimizing the users' cumulative watch time (main goal) in long term, which can be effectively optimized by Reinforcement Learning. On the other hand, the platforms also needs to satisfy the constraint of accommodating the responses of multiple user interactions (auxiliary goals) such like, follow, share etc. In this paper, we formulate the problem of short video recommendation as a Constrained Markov Decision Process (CMDP). We find that traditional constrained reinforcement learning algorithms can not work well in this setting. We propose a novel two-stage constrained actor-critic method: At stage one, we learn individual policies to optimize each auxiliary signal. At stage two, we learn a policy to (i) optimize the main signal and (ii) stay close to policies learned at the first stage, which effectively guarantees the performance of this main policy on the auxiliaries. Through extensive offline evaluations, we demonstrate effectiveness of our method over alternatives in both optimizing the main goal as well as balancing the others. We further show the advantage of our method in live experiments of short video recommendations, where it significantly outperforms other baselines in terms of both watch time and interactions. Our approach has been fully launched in the production system to optimize user experiences on the platform.

Ziru Liu, Jiejie Tian, Qingpeng Cai et al.

In recent years, Multi-task Learning (MTL) has yielded immense success in Recommender System (RS) applications. However, current MTL-based recommendation models tend to disregard the session-wise patterns of user-item interactions because they are predominantly constructed based on item-wise datasets. Moreover, balancing multiple objectives has always been a challenge in this field, which is typically avoided via linear estimations in existing works. To address these issues, in this paper, we propose a Reinforcement Learning (RL) enhanced MTL framework, namely RMTL, to combine the losses of different recommendation tasks using dynamic weights. To be specific, the RMTL structure can address the two aforementioned issues by (i) constructing an MTL environment from session-wise interactions and (ii) training multi-task actor-critic network structure, which is compatible with most existing MTL-based recommendation models, and (iii) optimizing and fine-tuning the MTL loss function using the weights generated by critic networks. Experiments on two real-world public datasets demonstrate the effectiveness of RMTL with a higher AUC against state-of-the-art MTL-based recommendation models. Additionally, we evaluate and validate RMTL's compatibility and transferability across various MTL models.

Xiaowan Hu, Yiyi Chen, Yan Li et al.

With the rapid expansion of e-commerce, more consumers have become accustomed to making purchases via livestreaming. Accurately identifying the products being sold by salespeople, i.e., livestreaming product retrieval (LPR), poses a fundamental and daunting challenge. The LPR task encompasses three primary dilemmas in real-world scenarios: 1) the recognition of intended products from distractor products present in the background; 2) the video-image heterogeneity that the appearance of products showcased in live streams often deviates substantially from standardized product images in stores; 3) there are numerous confusing products with subtle visual nuances in the shop. To tackle these challenges, we propose the Spatiotemporal Graphing Multi-modal Network (SGMN). First, we employ a text-guided attention mechanism that leverages the spoken content of salespeople to guide the model to focus toward intended products, emphasizing their salience over cluttered background products. Second, a long-range spatiotemporal graph network is further designed to achieve both instance-level interaction and frame-level matching, solving the misalignment caused by video-image heterogeneity. Third, we propose a multi-modal hard example mining, assisting the model in distinguishing highly similar products with fine-grained features across the video-image-text domain. Through extensive quantitative and qualitative experiments, we demonstrate the superior performance of our proposed SGMN model, surpassing the state-of-the-art methods by a substantial margin. The code is available at https://github.com/Huxiaowan/SGMN.

Jia Huang, Peng Jiang, Alvika Gautam et al.

Predicting pedestrian behavior is the key to ensure safety and reliability of autonomous vehicles. While deep learning methods have been promising by learning from annotated video frame sequences, they often fail to fully grasp the dynamic interactions between pedestrians and traffic, crucial for accurate predictions. These models also lack nuanced common sense reasoning. Moreover, the manual annotation of datasets for these models is expensive and challenging to adapt to new situations. The advent of Vision Language Models (VLMs) introduces promising alternatives to these issues, thanks to their advanced visual and causal reasoning skills. To our knowledge, this research is the first to conduct both quantitative and qualitative evaluations of VLMs in the context of pedestrian behavior prediction for autonomous driving. We evaluate GPT-4V(ision) on publicly available pedestrian datasets: JAAD and WiDEVIEW. Our quantitative analysis focuses on GPT-4V's ability to predict pedestrian behavior in current and future frames. The model achieves a 57% accuracy in a zero-shot manner, which, while impressive, is still behind the state-of-the-art domain-specific models (70%) in predicting pedestrian crossing actions. Qualitatively, GPT-4V shows an impressive ability to process and interpret complex traffic scenarios, differentiate between various pedestrian behaviors, and detect and analyze groups. However, it faces challenges, such as difficulty in detecting smaller pedestrians and assessing the relative motion between pedestrians and the ego vehicle.

Haofeng Yuan, Peng Jiang, Shiji Song

Set covering problem is an important class of combinatorial optimization problems, which has been widely applied and studied in many fields. In this paper, we propose an improved column generation algorithm with neural prediction (CG-P) for solving graph-based set covering problems. We leverage a graph neural network based neural prediction model to predict the probability to be included in the final solution for each edge. Our CG-P algorithm constructs a reduced graph that only contains the edges with higher predicted probability, and this graph reduction process significantly speeds up the solution process. We evaluate the CG-P algorithm on railway crew scheduling problems and it outperforms the baseline column generation algorithm. We provide two solution modes for our CG-P algorithm. In the optimal mode, we can obtain a solution with an optimality guarantee while reducing the time cost to 63.12%. In the fast mode, we can obtain a sub-optimal solution with a 7.62% optimality gap in only 2.91% computation time.

Wanqi Xue, Qingpeng Cai, Ruohan Zhan et al.

Long-term engagement is preferred over immediate engagement in sequential recommendation as it directly affects product operational metrics such as daily active users (DAUs) and dwell time. Meanwhile, reinforcement learning (RL) is widely regarded as a promising framework for optimizing long-term engagement in sequential recommendation. However, due to expensive online interactions, it is very difficult for RL algorithms to perform state-action value estimation, exploration and feature extraction when optimizing long-term engagement. In this paper, we propose ResAct which seeks a policy that is close to, but better than, the online-serving policy. In this way, we can collect sufficient data near the learned policy so that state-action values can be properly estimated, and there is no need to perform online exploration. ResAct optimizes the policy by first reconstructing the online behaviors and then improving it via a Residual Actor. To extract long-term information, ResAct utilizes two information-theoretical regularizers to confirm the expressiveness and conciseness of features. We conduct experiments on a benchmark dataset and a large-scale industrial dataset which consists of tens of millions of recommendation requests. Experimental results show that our method significantly outperforms the state-of-the-art baselines in various long-term engagement optimization tasks.

Peng Jiang, Srikanth Saripalli

Image and Point Clouds provide different information for robots. Finding the correspondences between data from different sensors is crucial for various tasks such as localization, mapping, and navigation. Learning-based descriptors have been developed for single sensors; there is little work on cross-modal features. This work treats learning cross-modal features as a dense contrastive learning problem. We propose a Tuple-Circle loss function for cross-modality feature learning. Furthermore, to learn good features and not lose generality, we developed a variant of widely used PointNet++ architecture for point cloud and U-Net CNN architecture for images. Moreover, we conduct experiments on a real-world dataset to show the effectiveness of our loss function and network structure. We show that our models indeed learn information from both images as well as LiDAR by visualizing the features.

Jianan Li, Shenwang Jiang, Liqiang Song et al.

The Five-hundred-meter Aperture Spherical radio Telescope (FAST) is the world's largest single-dish radio telescope. Its large reflecting surface achieves unprecedented sensitivity but is prone to damage, such as dents and holes, caused by naturally-occurring falling objects. Hence, the timely and accurate detection of surface defects is crucial for FAST's stable operation. Conventional manual inspection involves human inspectors climbing up and examining the large surface visually, a time-consuming and potentially unreliable process. To accelerate the inspection process and increase its accuracy, this work makes the first step towards automating the inspection of FAST by integrating deep-learning techniques with drone technology. First, a drone flies over the surface along a predetermined route. Since surface defects significantly vary in scale and show high inter-class similarity, directly applying existing deep detectors to detect defects on the drone imagery is highly prone to missing and misidentifying defects. As a remedy, we introduce cross-fusion, a dedicated plug-in operation for deep detectors that enables the adaptive fusion of multi-level features in a point-wise selective fashion, depending on local defect patterns. Consequently, strong semantics and fine-grained details are dynamically fused at different positions to support the accurate detection of defects of various scales and types. Our AI-powered drone-based automated inspection is time-efficient, reliable, and has good accessibility, which guarantees the long-term and stable operation of FAST.

Qingpeng Cai, Ruohan Zhan, Chi Zhang et al.

The wide popularity of short videos on social media poses new opportunities and challenges to optimize recommender systems on the video-sharing platforms. Users provide complex and multi-faceted responses towards recommendations, including watch time and various types of interactions with videos. As a result, established recommendation algorithms that concern a single objective are not adequate to meet this new demand of optimizing comprehensive user experiences. In this paper, we formulate the problem of short video recommendation as a constrained Markov Decision Process (MDP), where platforms want to optimize the main goal of user watch time in long term, with the constraint of accommodating the auxiliary responses of user interactions such as sharing/downloading videos. To solve the constrained MDP, we propose a two-stage reinforcement learning approach based on actor-critic framework. At stage one, we learn individual policies to optimize each auxiliary response. At stage two, we learn a policy to (i) optimize the main response and (ii) stay close to policies learned at the first stage, which effectively guarantees the performance of this main policy on the auxiliaries. Through extensive simulations, we demonstrate effectiveness of our approach over alternatives in both optimizing the main goal as well as balancing the others. We further show the advantage of our approach in live experiments of short video recommendations, where it significantly outperforms other baselines in terms of watch time and interactions from video views. Our approach has been fully launched in the production system to optimize user experiences on the platform.

Peng Jiang, Shengtong Liang, Tiao Lu

The Allen-Cahn equation with Flory-Huggins potential is a fundamental and crucial model in phase field simulation for describing phase separation phenomena, which serves as a core tool in diverse branches of natural sciences. The numerical simulation of the Allen-Cahn equation is of great importance but poses significant challenges due to the strong nonlinearity and the presence of logarithmic singularities at $u=0,1$ in the Flory-Huggins potential. In this paper, we consider convex splitting schemes to %preserve this bound and guarantee unconditional unique solvability, which reduces the numerical simulation to solving a singular nonlinear system arising from spatial discretization at each time step. We propose an iterative solver that is specifically designed for such systems based on the alternating direction method of multipliers (ADMM) approach. The scheme possesses properties such as bound preserving and discrete energy stability. Building upon the recent unconditionally convergent ADMM framework for the Cahn-Hilliard equation (Li et al., 2026), our key theoretical contributions are twofold: (a) a proof of unconditional convergence when the multiplier update step size $α\in (0,\frac{\sqrt{5}+1}{2})$; (b) a rigorous establishment of the linear convergence for the embedded ADMM solver. This effectively liberates the solver from time-step constraints or strict separation conditions. Comprehensive numerical experiments validate our proposed ADMM framework, where its theoretical predictions are fully substantiated in practice, showcasing efficiency and robustness.

Yunshan Peng, Ji Wu, Wentao Bai et al.

Reach and Frequency (R&F) contract advertising is an important form of widely used brand advertising. Unlike performance advertising, R&F contracts emphasize controllable delivery of UV and PV under given targeting, scheduling, and frequency control constraints. In practical systems, advertisers typically need to view the UV, PV change curves at different budget levels in real time when creating an R&F contract. However, most existing publicly available advertising datasets are based on independent samples, lacking a characterization of the core structure of the "budget-performance curve" (including UV and PV) in R&F contracts.This paper proposes and releases a large-scale R&F contract inventory estimation dataset. This dataset uses the R&F contract context consisting of "targeting-scheduling-frequency control" as the basic context, providing observations of UV and PV corresponding to multiple budget points within the same context, thus forming a complete budget-performance curve. The dataset explicitly includes a time-window-based frequency control mechanism (e.g.,"no more than 3 times within 5 days") and naturally satisfies the monotonicity and diminishing marginal returns characteristics in the budget and scheduling dimensions. We further derive the theoretical maximum exposure ceiling and use it as a consistency check to evaluate data quality and the feasibility of model predictions. Using this data set, this paper defines two standardized benchmark tasks: single-point performance prediction and reconstruction of budget-performance curves, and provides a set of reproducible baseline methods and evaluation protocols. This dataset can support systematic research on problems such as structural constraint learning, monotonic regression, curve consistency modeling, and R&F contract planning.The code for our experiments can be found at https://github.com/pengyunshan/RF-Inventory.

Ziang Yan, Shusen Wang, Guorui Zhou et al.

Online platforms often incentivize consumers to improve user engagement and platform revenue. Since different consumers might respond differently to incentives, individual-level budget allocation is an essential task in marketing campaigns. Recent advances in this field often address the budget allocation problem using a two-stage paradigm: the first stage estimates the individual-level treatment effects using causal inference algorithms, and the second stage invokes integer programming techniques to find the optimal budget allocation solution. Since the objectives of these two stages might not be perfectly aligned, such a two-stage paradigm could hurt the overall marketing effectiveness. In this paper, we propose a novel end-to-end framework to directly optimize the business goal under budget constraints. Our core idea is to construct a regularizer to represent the marketing goal and optimize it efficiently using gradient estimation techniques. As such, the obtained models can learn to maximize the marketing goal directly and precisely. We extensively evaluate our proposed method in both offline and online experiments, and experimental results demonstrate that our method outperforms current state-of-the-art methods. Our proposed method is currently deployed to allocate marketing budgets for hundreds of millions of users on a short video platform and achieves significant business goal improvements. Our code will be publicly available.

Hongyu Li, Peng Jiang, Tiejun Wang

The coin-tap test is a convenient and primary method for non-destructive testing, while its manual on-site operation is tough and costly. With the help of the latest intelligent signal processing method, convolutional neural networks (CNN), we achieve an intelligent coin-tap test which exhibited superior performance in recognizing the defects. However, this success of CNNs relies on plenty of well-labeled data from the identical scenario, which could be difficult to get for many real industrial practices. This paper further develops transfer learning strategies for this issue, that is, to transfer the model trained on data of one scenario to another. In experiments, the result presents a notable improvement by using domain adaptation and pseudo label learning strategies. Hence, it becomes possible to apply the model into scenarios with none or little (less than 10\%) labeled data adopting the transfer learning strategies proposed herein. In addition, we used a benchmark dataset constructed ourselves throughout this study. This benchmark dataset for the coin-tap test containing around 100,000 sound signals is published at https://github.com/PPhub-hy/torch-tapnet.

Peng Jiang, Srikanth Saripalli

As the demand for autonomous navigation in off-road environments increases, the need for effective solutions to understand these surroundings becomes essential. In this study, we confront the inherent complexities of semantic segmentation in RADAR data for off-road scenarios. We present a novel pipeline that utilizes LIDAR data and an existing annotated off-road LIDAR dataset for generating RADAR labels, in which the RADAR data are represented as images. Validated with real-world datasets, our pragmatic approach underscores the potential of RADAR technology for navigation applications in off-road environments.

Hongyu Li, Ximeng Ye, Peng Jiang et al.

Artificial intelligence (AI) shows great potential to reduce the huge cost of solving partial differential equations (PDEs). However, it is not fully realized in practice as neural networks are defined and trained on fixed domains and boundaries. Herein, we propose local neural operator (LNO) for solving transient PDEs on varied domains. It comes together with a handy strategy including boundary treatments, enabling one pre-trained LNO to predict solutions on different domains. For demonstration, LNO learns Navier-Stokes equations from randomly generated data samples, and then the pre-trained LNO is used as an explicit numerical time-marching scheme to solve the flow of fluid on unseen domains, e.g., the flow in a lid-driven cavity and the flow across the cascade of airfoils. It is about 1000$\times$ faster than the conventional finite element method to calculate the flow across the cascade of airfoils. The solving process with pre-trained LNO achieves great efficiency, with significant potential to accelerate numerical calculations in practice.

Te Yang, Xiangyu Zhu, Bo Wang et al.

Long-form video understanding requires efficient navigation of extensive visual data to pinpoint sparse yet critical information. Current approaches to longform video understanding either suffer from severe computational overhead due to dense preprocessing, or fail to effectively balance exploration and exploitation, resulting in incomplete information coverage and inefficiency. In this work, we introduce EEA, a novel video agent framework that archives exploration-exploitation balance through semantic guidance with hierarchical tree search process. EEA autonomously discovers and dynamically updates task-relevant semantic queries, and collects video frames closely matched to these queries as semantic anchors. During the tree search process, instead of uniform expansion, EEA preferentially explores semantically relevant frames while ensuring sufficient coverage within unknown segments. Moreover, EEA adaptively combines intrinsic rewards from visionlanguage models (VLMs) with semantic priors by explicitly modeling uncertainty to achieve stable and precise evaluation of video segments. Experiments across various long-video benchmarks validate the superior performance and computational efficiency of our proposed method.

Siqi Kou, Jiachun Jin, Zetong Zhou et al.

Recent progress in text-to-image (T2I) diffusion models (DMs) has enabled high-quality visual synthesis from diverse textual prompts. Yet, most existing T2I DMs, even those equipped with large language model (LLM)-based text encoders, remain text-pixel mappers -- they employ LLMs merely as text encoders, without leveraging their inherent reasoning capabilities to infer what should be visually depicted given the textual prompt. To move beyond such literal generation, we propose the think-then-generate (T2G) paradigm, where the LLM-based text encoder is encouraged to reason about and rewrite raw user prompts; the states of the rewritten prompts then serve as diffusion conditioning. To achieve this, we first activate the think-then-rewrite pattern of the LLM encoder with a lightweight supervised fine-tuning process. Subsequently, the LLM encoder and diffusion backbone are co-optimized to ensure faithful reasoning about the context and accurate rendering of the semantics via Dual-GRPO. In particular, the text encoder is reinforced using image-grounded rewards to infer and recall world knowledge, while the diffusion backbone is pushed to produce semantically consistent and visually coherent images. Experiments show substantial improvements in factual consistency, semantic alignment, and visual realism across reasoning-based image generation and editing benchmarks, achieving 0.79 on WISE score, nearly on par with GPT-4. Our results constitute a promising step toward next-generation unified models with reasoning, expression, and demonstration capacities.

Changyu Hou, Jun Wang, Yixuan Qiao et al.

Large scale pre-training models have been widely used in named entity recognition (NER) tasks. However, model ensemble through parameter averaging or voting can not give full play to the differentiation advantages of different models, especially in the open domain. This paper describes our NER system in the SemEval 2022 task11: MultiCoNER. We proposed an effective system to adaptively ensemble pre-trained language models by a Transformer layer. By assigning different weights to each model for different inputs, we adopted the Transformer layer to integrate the advantages of diverse models effectively. Experimental results show that our method achieves superior performances in Farsi and Dutch.

Zijian Zhang, Shuchang Liu, Jiaao Yu et al.

Multi-domain recommendation and multi-task recommendation have demonstrated their effectiveness in leveraging common information from different domains and objectives for comprehensive user modeling. Nonetheless, the practical recommendation usually faces multiple domains and tasks simultaneously, which cannot be well-addressed by current methods. To this end, we introduce M3oE, an adaptive Multi-domain Multi-task Mixture-of-Experts recommendation framework. M3oE integrates multi-domain information, maps knowledge across domains and tasks, and optimizes multiple objectives. We leverage three mixture-of-experts modules to learn common, domain-aspect, and task-aspect user preferences respectively to address the complex dependencies among multiple domains and tasks in a disentangled manner. Additionally, we design a two-level fusion mechanism for precise control over feature extraction and fusion across diverse domains and tasks. The framework's adaptability is further enhanced by applying AutoML technique, which allows dynamic structure optimization. To the best of the authors' knowledge, our M3oE is the first effort to solve multi-domain multi-task recommendation self-adaptively. Extensive experiments on two benchmark datasets against diverse baselines demonstrate M3oE's superior performance. The implementation code is available to ensure reproducibility.

Zhenghai Xue, Qingpeng Cai, Bin Yang et al.

The field of Reinforcement Learning (RL) has garnered increasing attention for its ability of optimizing user retention in recommender systems. A primary obstacle in this optimization process is the environment non-stationarity stemming from the continual and complex evolution of user behavior patterns over time, such as variations in interaction rates and retention propensities. These changes pose significant challenges to existing RL algorithms for recommendations, leading to issues with dynamics and reward distribution shifts. This paper introduces a novel approach called \textbf{A}daptive \textbf{U}ser \textbf{R}etention \textbf{O}ptimization (AURO) to address this challenge. To navigate the recommendation policy in non-stationary environments, AURO introduces an state abstraction module in the policy network. The module is trained with a new value-based loss function, aligning its output with the estimated performance of the current policy. As the policy performance of RL is sensitive to environment drifts, the loss function enables the state abstraction to be reflective of environment changes and notify the recommendation policy to adapt accordingly. Additionally, the non-stationarity of the environment introduces the problem of implicit cold start, where the recommendation policy continuously interacts with users displaying novel behavior patterns. AURO encourages exploration guarded by performance-based rejection sampling to maintain a stable recommendation quality in the cost-sensitive online environment. Extensive empirical analysis are conducted in a user retention simulator, the MovieLens dataset, and a live short-video recommendation platform, demonstrating AURO's superior performance against all evaluated baseline algorithms.

Jingtong Gao, Zeyu Song, Chi Lu et al.

Recommender systems are critical for delivering personalized content across digital platforms, and recent advances in Large Language Models (LLMs) offer new opportunities to enhance them with richer world knowledge and explicit reasoning capabilities. With the help of reasoning knowledge, recommendations can better infer users' underlying intents, adapt to evolving preferences, and leverage semantic relationships for improved accuracy and interpretability. However, existing reasoning-based recommendation methods often fail to fully align the LLM's reasoning process with recommendation-specific objectives due to structural disruption during integration and difficulties in translating free-form generation into accurate item predictions. In this paper, we introduce RPORec, a reinforced preference optimization framework that unifies an LLM backbone's reasoning ability with a dedicated recommendation head (Rechead) for precise item retrieval. RPORec comprises two stages: (1) Reasoning-Augmented Recommendation Modeling, where high-quality Chain-of-Thought (CoT) reasoning is generated and used as auxiliary knowledge to guide the Rechead in learning recommendation-specific representations; and (2) Advanced Reasoning Refinement and Alignment, in which the trained Rechead produces verifiable rewards to fine-tune the LLM backbone via reinforcement learning, enhancing reasoning quality, structural consistency, and task relevance. Extensive experiments on public benchmarks and large-scale online deployments show that RPORec consistently outperforms state-of-the-art LLM-based recommendation methods, demonstrating the effectiveness of reasoning-augmented recommendation modeling in real-world systems.

Peng Jiang, Juan Liu, Lang Wang et al.

Labeling medical images depends on professional knowledge, making it difficult to acquire large amount of annotated medical images with high quality in a short time. Thus, making good use of limited labeled samples in a small dataset to build a high-performance model is the key to medical image classification problem. In this paper, we propose a deeply supervised Layer Selective Attention Network (LSANet), which comprehensively uses label information in feature-level and prediction-level supervision. For feature-level supervision, in order to better fuse the low-level features and high-level features, we propose a novel visual attention module, Layer Selective Attention (LSA), to focus on the feature selection of different layers. LSA introduces a weight allocation scheme which can dynamically adjust the weighting factor of each auxiliary branch during the whole training process to further enhance deeply supervised learning and ensure its generalization. For prediction-level supervision, we adopt the knowledge synergy strategy to promote hierarchical information interactions among all supervision branches via pairwise knowledge matching. Using the public dataset, MedMNIST, which is a large-scale benchmark for biomedical image classification covering diverse medical specialties, we evaluate LSANet on multiple mainstream CNN architectures and various visual attention modules. The experimental results show the substantial improvements of our proposed method over its corresponding counterparts, demonstrating that LSANet can provide a promising solution for label-efficient learning in the field of medical image classification.

Ben Xue, Dan Liu, Lixiang Wang et al.

Generative recommendation has recently attracted widespread attention in industry due to its potential for scaling and stronger model capacity. However, deploying real-time generative recommendation in large-scale advertising requires designs beyond large-language-model (LLM)-style training and serving recipes. We present a production-oriented generative recommender co-designed across architecture, learning, and serving, named GR4AD (Generative Recommendation for ADdvertising). As for tokenization, GR4AD proposes UA-SID (Unified Advertisement Semantic ID) to capture complicated business information. Furthermore, GR4AD introduces LazyAR, a lazy autoregressive decoder that relaxes layer-wise dependencies for short, multi-candidate generation, preserving effectiveness while reducing inference cost, which facilitates scaling under fixed serving budgets. To align optimization with business value, GR4AD employs VSL (Value-Aware Supervised Learning) and proposes RSPO (Ranking-Guided Softmax Preference Optimization), a ranking-aware, list-wise reinforcement learning algorithm that optimizes value-based rewards under list-level metrics for continual online updates. For online inference, we further propose dynamic beam serving, which adapts beam width across generation levels and online load to control compute. Large-scale online A/B tests show up to 4.2% ad revenue improvement over an existing DLRM-based stack, with consistent gains from both model scaling and inference-time scaling. GR4AD has been fully deployed in Kuaishou advertising system with over 400 million users and achieves high-throughput real-time serving.

Xikai Zhang, Yongzhi Li, Likang Xiao et al.

Reinforcement learning has become a cornerstone for aligning and unlocking the reasoning capabilities of large-scale models. At its core, the training loop of GRPO and its variants alternates between rollout sampling and policy update. Unlike supervised learning, where each gradient step is anchored to an explicit ground-truth target, the optimal gradient direction for updating model parameters in this setting is not known a priori; the high-quality rollouts drawn during the sampling stage therefore act as the implicit "teacher" that guides every parameter update. However, GRPO adopt a simple sampling scheme that conditions all rollouts on the same original prompt. When a task lies beyond the policy model's current capability, this sampling scheme rarely yields a high-quality rollout, leaving the policy model without a meaningful gradient direction when updating its parameters, which causes training to stall. To address this issue, we propose FBOS-RL, a Feedback-Driven Bi-Objective Synergistic reinforcement learning framework. Specifically, we let the model perform Feedback-Guided Exploration Enhancement based on the feedback provided by the environment, and on top of this we design two mutually reinforcing training objectives: Exploitation-oriented Policy Alignment(EPA) and Exploration-oriented Capability Cultivation(ECC). Extensive experiments demonstrate that EPA and ECC can mutually reinforce each other, forming a positive flywheel effect that significantly improves both the training efficiency and the final performance ceiling of reinforcement learning. Specifically, under an identical number of rollouts, FBOS-RL learns substantially faster than GRPO and feedback-based baselines and ultimately attains a higher performance ceiling, while exhibiting higher policy entropy and lower gradient norms throughout training.

Kasi Viswanath, Peng Jiang, Sujit PB et al.

LiDAR is used in autonomous driving to provide 3D spatial information and enable accurate perception in off-road environments, aiding in obstacle detection, mapping, and path planning. Learning-based LiDAR semantic segmentation utilizes machine learning techniques to automatically classify objects and regions in LiDAR point clouds. Learning-based models struggle in off-road environments due to the presence of diverse objects with varying colors, textures, and undefined boundaries, which can lead to difficulties in accurately classifying and segmenting objects using traditional geometric-based features. In this paper, we address this problem by harnessing the LiDAR intensity parameter to enhance object segmentation in off-road environments. Our approach was evaluated in the RELLIS-3D data set and yielded promising results as a preliminary analysis with improved mIoU for classes "puddle" and "grass" compared to more complex deep learning-based benchmarks. The methodology was evaluated for compatibility across both Velodyne and Ouster LiDAR systems, assuring its cross-platform applicability. This analysis advocates for the incorporation of calibrated intensity as a supplementary input, aiming to enhance the prediction accuracy of learning based semantic segmentation frameworks. https://github.com/MOONLABIISERB/lidar-intensity-predictor/tree/main

Jingyu Liu, Minquan Wang, Ye Ma et al.

Videos showcasing specific products are increasingly important for E-commerce. Key moments naturally exist as the first appearance of a specific product, presentation of its distinctive features, the presence of a buying link, etc. Adding proper sound effects (SFX) to these key moments, or video decoration with SFX (VDSFX), is crucial for enhancing the user engaging experience. Previous studies about adding SFX to videos perform video to SFX matching at a holistic level, lacking the ability of adding SFX to a specific moment. Meanwhile, previous studies on video highlight detection or video moment retrieval consider only moment localization, leaving moment to SFX matching untouched. By contrast, we propose in this paper D&M, a unified method that accomplishes key moment detection and moment to SFX matching simultaneously. Moreover, for the new VDSFX task we build a large-scale dataset SFX-Moment from an E-commerce platform. For a fair comparison, we build competitive baselines by extending a number of current video moment detection methods to the new task. Extensive experiments on SFX-Moment show the superior performance of the proposed method over the baselines.

Yuhang Zhou, Zhibin Wang, Peng Jiang et al.

Training large language models faces frequent interruptions due to various faults, demanding robust fault-tolerance. Existing backup-free methods, such as redundant computation, dynamic parallelism, and data rerouting, each incur performance penalties, whether from ongoing overhead, lengthy reconfigurations, or post-recovery inefficiencies. We propose Chameleon, an adaptive fault-tolerant system that intelligently selects optimal recovery strategies when a failure occurs. Chameleon achieves this through a unified performance model, expedient execution plan search, accurate performance estimation, and efficient communication optimizations. Experiments on a 32-card cluster show that Chameleon maintains a performance gap of within 11.00% between post-recovery and failure-free training, while preserving model convergence and efficient memory usage. Compared to state-of-the-art methods, Chameleon achieves up to 1.229x and 1.355x higher average throughput than Oobleck and Recycle, respectively.

Siyun Yang, Shixiao Yang, Jian Wang et al.

In online advertising, marketing interventions such as coupons introduce significant confounding bias into Click-Through Rate (CTR) prediction. Observed clicks reflect a mixture of users' intrinsic preferences and the uplift induced by these interventions. This causes conventional models to miscalibrate base CTRs, which distorts downstream ranking and billing decisions. Furthermore, marketing interventions often operate as multi-valued treatments with varying magnitudes, introducing additional complexity to CTR prediction. To address these issues, we propose the \textbf{Uni}fied \textbf{M}ulti-\textbf{V}alued \textbf{T}reatment Network (UniMVT). Specifically, UniMVT disentangles confounding factors from treatment-sensitive representations, enabling a full-space counterfactual inference module to jointly reconstruct the debiased base CTR and intensity-response curves. To handle the complexity of multi-valued treatments, UniMVT employs an auxiliary intensity estimation task to capture treatment propensities and devise a unit uplift objective that normalizes the intervention effect. This ensures comparable estimation across the continuous coupon-value spectrum. UniMVT simultaneously achieves debiased CTR prediction for accurate system calibration and precise uplift estimation for incentive allocation. Extensive experiments on synthetic and industrial datasets demonstrate UniMVT's superiority in both predictive accuracy and calibration. Furthermore, real-world A/B tests confirm that UniMVT significantly improves business metrics through more effective coupon distribution.

Yuyang You, Yongzhi Li, Jiahui Li et al.

Video generation has recently emerged as a central task in the field of generative AI. However, the substantial computational cost inherent in video synthesis makes model distillation a critical technique for efficient deployment. Despite its significance, there is a scarcity of methods specifically designed for video diffusion models. Prevailing approaches often directly adapt image distillation techniques, which frequently lead to artifacts such as oversaturation, temporal inconsistency, and mode collapse. To address these challenges, we propose a novel distillation framework tailored specifically for video diffusion models. Its core innovations include: (1) an adaptive regression loss that dynamically adjusts spatial supervision weights to prevent artifacts arising from excessive distribution shifts; (2) a temporal regularization loss to counteract temporal collapse, promoting smooth and physically plausible sampling trajectories; and (3) an inference-time frame interpolation strategy that reduces sampling overhead while preserving perceptual quality. Extensive experiments and ablation studies on the VBench and VBench2 benchmarks demonstrate that our method achieves stable few-step video synthesis, significantly enhancing perceptual fidelity and motion realism. It consistently outperforms existing distillation baselines across multiple metrics.

Hao Jiang, Long Zhang, Guoquan Wang et al.

Local-life recommendation have witnessed rapid growth, providing users with convenient access to daily essentials. However, this domain faces two key challenges: (1) spatial constraints, driven by the requirements of the local-life scenario, where items are usually shown only to users within a limited geographic area, indirectly reducing their exposure probability; and (2) long-tail sparsity, where few popular items dominate user interactions, while many high-quality long-tail items are largely overlooked due to imbalanced interaction opportunities. Existing methods typically adopt a user-centric perspective, such as modeling spatial user preferences or enhancing long-tail representations with collaborative filtering signals. However, we argue that an item-centric perspective is more suitable for this domain, focusing on enhancing long-tail items representation that align with the spatially-constrained characteristics of local lifestyle services. To tackle this issue, we propose ReST, a Plug-And-Play Spatially-Constrained Representation Enhancement Framework for Long-Tail Local-Life Recommendation. Specifically, we first introduce a Meta ID Warm-up Network, which initializes fundamental ID representations by injecting their basic attribute-level semantic information. Subsequently, we propose a novel Spatially-Constrained ID Representation Enhancement Network (SIDENet) based on contrastive learning, which incorporates two efficient strategies: a spatially-constrained hard sampling strategy and a dynamic representation alignment strategy. This design adaptively identifies weak ID representations based on their attribute-level information during training. It additionally enhances them by capturing latent item relationships within the spatially-constrained characteristics of local lifestyle services, while preserving compatibility with popular items.

Ruixiang Zhao, Jian Jia, Yan Li et al.

E-commerce is increasingly multimedia-enriched, with products exhibited in a broad-domain manner as images, short videos, or live stream promotions. A unified and vectorized cross-domain production representation is essential. Due to large intra-product variance and high inter-product similarity in the broad-domain scenario, a visual-only representation is inadequate. While Automatic Speech Recognition (ASR) text derived from the short or live-stream videos is readily accessible, how to de-noise the excessively noisy text for multimodal representation learning is mostly untouched. We propose ASR-enhanced Multimodal Product Representation Learning (AMPere). In order to extract product-specific information from the raw ASR text, AMPere uses an easy-to-implement LLM-based ASR text summarizer. The LLM-summarized text, together with visual data, is then fed into a multi-branch network to generate compact multimodal embeddings. Extensive experiments on a large-scale tri-domain dataset verify the effectiveness of AMPere in obtaining a unified multimodal product representation that clearly improves cross-domain product retrieval.

Runpeng Cui, Zhipeng Sun, Chi Lu et al.

Continuous value prediction plays a crucial role in industrial-scale recommendation systems, including tasks such as predicting users' watch-time and estimating the gross merchandise value (GMV) in e-commerce transactions. However, it remains challenging due to the highly complex and long-tailed nature of the data distributions. Existing generative approaches rely on rigid parametric distribution assumptions, which fundamentally limits their performance when such assumptions misalign with real-world data. Overly simplified forms cannot adequately model real-world complexities, while more intricate assumptions often suffer from poor scalability and generalization. To address these challenges, we propose a residual quantization (RQ)-based sequence learning framework that represents target continuous values as a sum of ordered quantization codes, predicted recursively from coarse to fine granularity with diminishing quantization errors. We introduce a representation learning objective that aligns RQ code embedding space with the ordinal structure of target values, allowing the model to capture continuous representations for quantization codes and further improving prediction accuracy. We perform extensive evaluations on public benchmarks for lifetime value (LTV) and watch-time prediction, alongside a large-scale online experiment for GMV prediction on an industrial short-video recommendation platform. The results consistently show that our approach outperforms state-of-the-art methods, while demonstrating strong generalization across diverse continuous value prediction tasks in recommendation systems.

Jinren Ding, Xuejian Xu, Shen Jiang et al.

Decision Transformer (DT) shows promise for generative auto-bidding by capturing temporal dependencies, but suffers from two critical limitations: insufficient cross-correlation modeling among state, action, and return-to-go (RTG) sequences, and indiscriminate learning of optimal/suboptimal behaviors. To address these, we propose C2, a novel framework enhancing DT with two core innovations: (1) a Cross Learning Block (CLB) via cross-attention to strengthen inter-sequence correlation modeling; (2) a Constraint-aware Loss (CL) incorporating budget and Cost-Per-Acquisition (CPA) constraints for selective learning of optimal trajectories. Extensive offline evaluations on the AuctionNet dataset demonstrate consistent performance gains (up to 3.2% over state-of-the-art method) across diverse budget settings; ablation studies verify the complementary synergy of CLB and CL, confirming C2's superiority in auto-bidding. The code for reproducing our results is available at: https://github.com/Dingjinren/C2.

Wei Yang, Rui Zhong, Yiqun Chen et al.

Multimodal recommendation aims to integrate collaborative signals with heterogeneous content such as visual and textual information, but remains challenged by modality-specific noise, semantic inconsistency, and unstable propagation over user-item graphs. These issues are often exacerbated by naive fusion or shallow modeling strategies, leading to degraded generalization and poor robustness. While recent work has explored the frequency domain as a lens to separate stable from noisy signals, most methods rely on static filtering or reweighting, lacking the ability to reason over spectral structure or adapt to modality-specific reliability. To address these challenges, we propose a Structured Spectral Reasoning (SSR) framework for frequency-aware multimodal recommendation. Our method follows a four-stage pipeline: (i) Decompose graph-based multimodal signals into spectral bands via graph-guided transformations to isolate semantic granularity; (ii) Modulate band-level reliability with spectral band masking, a training-time masking with a prediction-consistency objective that suppresses brittle frequency components; (iii) Fuse complementary frequency cues using hyperspectral reasoning with low-rank cross-band interaction; and (iv) Align modality-specific spectral features via contrastive regularization to promote semantic and structural consistency. Experiments on three real-world benchmarks show consistent gains over strong baselines, particularly under sparse and cold-start settings. Additional analyses indicate that structured spectral modeling improves robustness and provides clearer diagnostics of how different bands contribute to performance.

Yanjie Zhu, Zhen Zhang, Yunli Wang et al.

Cascade ranking is a widely adopted paradigm in large-scale information retrieval systems for Top-K item selection. However, the Top-K operator is non-differentiable, hindering end-to-end training. Existing methods include Learning-to-Rank approaches (e.g., LambdaLoss), which optimize ranking metrics like NDCG and suffer from objective misalignment, and differentiable sorting-based methods (e.g., ARF, LCRON), which relax permutation matrices for direct Top-K optimization but introduce gradient conflicts through matrix aggregation. A promising alternative is to directly construct a differentiable approximation of the Top-K selection operator, bypassing the use of soft permutation matrices. However, even state-of-the-art differentiable Top-K operator (e.g., LapSum) require $O(n \log n)$ complexity due to their dependence on sorting for solving the threshold. Thus, we propose DFTopK, a novel differentiable Top-K operator achieving optimal $O(n)$ time complexity. By relaxing normalization constraints, DFTopK admits a closed-form solution and avoids sorting. DFTopK also avoids the gradient conflicts inherent in differentiable sorting-based methods. We evaluate DFTopK on both the public benchmark RecFLow and an industrial system. Experimental results show that DFTopK significantly improves training efficiency while achieving superior performance, which enables us to scale up training samples more efficiently. In the online A/B test, DFTopK yielded a +1.77% revenue lift with the same computational budget compared to the baseline. To the best of our knowledge, this work is the first to introduce differentiable Top-K operators into recommendation systems and the first to achieve theoretically optimal linear-time complexity for Top-K selection. We have open-sourced our implementation to facilitate future research in both academia and industry.

Hanzhou Liu, Peng Jiang, Jia Huang et al.

Restoring 3D scenes captured under low-light con- ditions remains a fundamental yet challenging problem. Most existing approaches depend on precomputed camera poses and scene-specific optimization, which greatly restricts their scala- bility to dynamic real-world environments. To overcome these limitations, we introduce Lumos3D, a generalizable pose-free framework for 3D low-light scene restoration. Trained once on a single dataset, Lumos3D performs inference in a purely feed- forward manner, directly restoring illumination and structure from unposed, low-light multi-view images without any per- scene training or optimization. Built upon a geometry-grounded backbone, Lumos3D reconstructs a normal-light 3D Gaussian representation that restores illumination while faithfully pre- serving structural details. During training, a cross-illumination distillation scheme is employed, where the teacher network is distilled on normal-light ground truth to transfer accurate geometric information, such as depth, to the student model. A dedicated Lumos loss is further introduced to promote photomet- ric consistency within the reconstructed 3D space. Experiments on real-world datasets demonstrate that Lumos3D achieves high- fidelity low-light 3D scene restoration with accurate geometry and strong generalization to unseen cases. Furthermore, the framework naturally extends to handle over-exposure correction, highlighting its versatility for diverse lighting restoration tasks.

Milton Zhou, Sizhong Qin, Yongzhi Li et al.

Short-form videos have become a primary medium for digital advertising, requiring scalable and efficient content creation. However, current workflows and AI tools remain disjoint and modality-specific, leading to high production costs and low overall efficiency. To address this issue, we propose AutoCut, an end-to-end advertisement video editing framework based on multimodal discretization and controllable editing. AutoCut employs dedicated encoders to extract video and audio features, then applies residual vector quantization to discretize them into unified tokens aligned with textual representations, constructing a shared video-audio-text token space. Built upon a foundation model, we further develop a multimodal large language model for video editing through combined multimodal alignment and supervised fine-tuning, supporting tasks covering video selection and ordering, script generation, and background music selection within a unified editing framework. Finally, a complete production pipeline converts the predicted token sequences into deployable long video outputs. Experiments on real-world advertisement datasets show that AutoCut reduces production cost and iteration time while substantially improving consistency and controllability, paving the way for scalable video creation.

Yongxiang Tang, Yanhua Cheng, Xiaocheng Liu et al.

In many machine learning tasks, it is often necessary for the relationship between input and output variables to be monotonic, including both strictly monotonic and implicitly monotonic relationships. Traditional methods for maintaining monotonicity mainly rely on construction or regularization techniques, whereas this paper shows that the issue of strict monotonic probability can be viewed as a partial order between an observable revenue variable and a latent cost variable. This perspective enables us to reformulate the monotonicity challenge into modeling the latent cost variable. To tackle this, we introduce a generative network for the latent cost variable, termed the Generative Cost Model (GCM), which inherently addresses the strict monotonic problem, and propose the Implicit Generative Cost Model (IGCM) to address the implicit monotonic problem. We further validate our approach with a numerical simulation of quantile regression and conduct multiple experiments on public datasets, showing that our method significantly outperforms existing monotonic modeling techniques. The code for our experiments can be found at https://github.com/tyxaaron/GCM.

Peng Jiang, Philip Osteen, Maggie Wigness et al.

Semantic scene understanding is crucial for robust and safe autonomous navigation, particularly so in off-road environments. Recent deep learning advances for 3D semantic segmentation rely heavily on large sets of training data, however existing autonomy datasets either represent urban environments or lack multimodal off-road data. We fill this gap with RELLIS-3D, a multimodal dataset collected in an off-road environment, which contains annotations for 13,556 LiDAR scans and 6,235 images. The data was collected on the Rellis Campus of Texas A\&M University and presents challenges to existing algorithms related to class imbalance and environmental topography. Additionally, we evaluate the current state-of-the-art deep learning semantic segmentation models on this dataset. Experimental results show that RELLIS-3D presents challenges for algorithms designed for segmentation in urban environments. This novel dataset provides the resources needed by researchers to continue to develop more advanced algorithms and investigate new research directions to enhance autonomous navigation in off-road environments. RELLIS-3D is available at https://github.com/unmannedlab/RELLIS-3D

Zhiyi Pan, Peng Jiang, Changhe Tu

Scribble-supervised semantic segmentation has gained much attention recently for its promising performance without high-quality annotations. Many approaches have been proposed. Typically, they handle this problem to either introduce a well-labeled dataset from another related task, turn to iterative refinement and post-processing with the graphical model, or manipulate the scribble label. This work aims to achieve semantic segmentation supervised by scribble label directly without auxiliary information and other intermediate manipulation. Specifically, we impose diffusion on neural representation by random walk and consistency on neural eigenspace by self-supervision, which forces the neural network to produce dense and consistent predictions over the whole dataset. The random walk embedded in the network will compute a probabilistic transition matrix, with which the neural representation diffused to be uniform. Moreover, given the probabilistic transition matrix, we apply the self-supervision on its eigenspace for consistency in the image's main parts. In addition to comparing the common scribble dataset, we also conduct experiments on the modified datasets that randomly shrink and even drop the scribbles on image objects. The results demonstrate the superiority of the proposed method and are even comparable to some full-label supervised ones. The code and datasets are available at https://github.com/panzhiyi/RW-SS.

Shengtian Yang, Yu Li, Shuo He et al.

Reinforcement learning (RL) has equipped LLM agents with a strong ability to solve complex tasks. However, existing RL methods normally use a \emph{single} policy network, causing \emph{simplicity bias} where simple tasks occupy most parameters and dominate gradient updates, leaving insufficient capacity for complex tasks. A plausible remedy could be employing the Mixture-of-Experts (MoE) architecture in the policy network, as MoE allows different parameters (experts) to specialize in different tasks, preventing simple tasks from dominating all parameters. However, a key limitation of traditional MoE is its token-level routing, where the router assigns each token to specialized experts, which fragments phase-consistent patterns into scattered expert assignments and thus undermines expert specialization. In this paper, we propose \textbf{Phase-Aware Mixture of Experts (PA-MoE)}. It first features a lightweight \emph{phase router} that learns latent phase boundaries directly from the RL objective without pre-defining phase categories. Then, the phase router allocates temporally consistent assignments to the same expert, allowing experts to preserve phase-specific expertise. Experimental results demonstrate the effectiveness of our proposed PA-MoE.

Zijian Zhang, Shuchang Liu, Ziru Liu et al.

User simulators can rapidly generate a large volume of timely user behavior data, providing a testing platform for reinforcement learning-based recommender systems, thus accelerating their iteration and optimization. However, prevalent user simulators generally suffer from significant limitations, including the opacity of user preference modeling and the incapability of evaluating simulation accuracy. In this paper, we introduce an LLM-powered user simulator to simulate user engagement with items in an explicit manner, thereby enhancing the efficiency and effectiveness of reinforcement learning-based recommender systems training. Specifically, we identify the explicit logic of user preferences, leverage LLMs to analyze item characteristics and distill user sentiments, and design a logical model to imitate real human engagement. By integrating a statistical model, we further enhance the reliability of the simulation, proposing an ensemble model that synergizes logical and statistical insights for user interaction simulations. Capitalizing on the extensive knowledge and semantic generation capabilities of LLMs, our user simulator faithfully emulates user behaviors and preferences, yielding high-fidelity training data that enrich the training of recommendation algorithms. We establish quantifying and qualifying experiments on five datasets to validate the simulator's effectiveness and stability across various recommendation scenarios.