Yun Chen

Ze Yang, Yun Chen, Jingkang Wang et al. · utoronto

Rigorously testing autonomy systems is essential for making safe self-driving vehicles (SDV) a reality. It requires one to generate safety critical scenarios beyond what can be collected safely in the world, as many scenarios happen rarely on public roads. To accurately evaluate performance, we need to test the SDV on these scenarios in closed-loop, where the SDV and other actors interact with each other at each timestep. Previously recorded driving logs provide a rich resource to build these new scenarios from, but for closed loop evaluation, we need to modify the sensor data based on the new scene configuration and the SDV's decisions, as actors might be added or removed and the trajectories of existing actors and the SDV will differ from the original log. In this paper, we present UniSim, a neural sensor simulator that takes a single recorded log captured by a sensor-equipped vehicle and converts it into a realistic closed-loop multi-sensor simulation. UniSim builds neural feature grids to reconstruct both the static background and dynamic actors in the scene, and composites them together to simulate LiDAR and camera data at new viewpoints, with actors added or removed and at new placements. To better handle extrapolated views, we incorporate learnable priors for dynamic objects, and leverage a convolutional network to complete unseen regions. Our experiments show UniSim can simulate realistic sensor data with small domain gap on downstream tasks. With UniSim, we demonstrate closed-loop evaluation of an autonomy system on safety-critical scenarios as if it were in the real world.

Jingkang Wang, Sivabalan Manivasagam, Yun Chen et al. · utoronto

Realistic simulation is key to enabling safe and scalable development of % self-driving vehicles. A core component is simulating the sensors so that the entire autonomy system can be tested in simulation. Sensor simulation involves modeling traffic participants, such as vehicles, with high quality appearance and articulated geometry, and rendering them in real time. The self-driving industry has typically employed artists to build these assets. However, this is expensive, slow, and may not reflect reality. Instead, reconstructing assets automatically from sensor data collected in the wild would provide a better path to generating a diverse and large set with good real-world coverage. Nevertheless, current reconstruction approaches struggle on in-the-wild sensor data, due to its sparsity and noise. To tackle these issues, we present CADSim, which combines part-aware object-class priors via a small set of CAD models with differentiable rendering to automatically reconstruct vehicle geometry, including articulated wheels, with high-quality appearance. Our experiments show our method recovers more accurate shapes from sparse data compared to existing approaches. Importantly, it also trains and renders efficiently. We demonstrate our reconstructed vehicles in several applications, including accurate testing of autonomy perception systems.

Ze Yang, Sivabalan Manivasagam, Yun Chen et al. · utoronto

Reconstructing objects from real world data and rendering them at novel views is critical to bringing realism, diversity and scale to simulation for robotics training and testing. In this work, we present NeuSim, a novel approach that estimates accurate geometry and realistic appearance from sparse in-the-wild data captured at distance and at limited viewpoints. Towards this goal, we represent the object surface as a neural signed distance function and leverage both LiDAR and camera sensor data to reconstruct smooth and accurate geometry and normals. We model the object appearance with a robust physics-inspired reflectance representation effective for in-the-wild data. Our experiments show that NeuSim has strong view synthesis performance on challenging scenarios with sparse training views. Furthermore, we showcase composing NeuSim assets into a virtual world and generating realistic multi-sensor data for evaluating self-driving perception models.

Jeffrey Yunfan Liu, Yun Chen, Ze Yang et al.

We propose a new method for realistic real-time novel-view synthesis (NVS) of large scenes. Existing neural rendering methods generate realistic results, but primarily work for small scale scenes (<50 square meters) and have difficulty at large scale (>10000 square meters). Traditional graphics-based rasterization rendering is fast for large scenes but lacks realism and requires expensive manually created assets. Our approach combines the best of both worlds by taking a moderate-quality scaffold mesh as input and learning a neural texture field and shader to model view-dependant effects to enhance realism, while still using the standard graphics pipeline for real-time rendering. Our method outperforms existing neural rendering methods, providing at least 30x faster rendering with comparable or better realism for large self-driving and drone scenes. Our work is the first to enable real-time rendering of large real-world scenes.

Yun Chen, Jingkang Wang, Ze Yang et al.

Large scale 3D scene reconstruction is important for applications such as virtual reality and simulation. Existing neural rendering approaches (e.g., NeRF, 3DGS) have achieved realistic reconstructions on large scenes, but optimize per scene, which is expensive and slow, and exhibit noticeable artifacts under large view changes due to overfitting. Generalizable approaches or large reconstruction models are fast, but primarily work for small scenes/objects and often produce lower quality rendering results. In this work, we introduce G3R, a generalizable reconstruction approach that can efficiently predict high-quality 3D scene representations for large scenes. We propose to learn a reconstruction network that takes the gradient feedback signals from differentiable rendering to iteratively update a 3D scene representation, combining the benefits of high photorealism from per-scene optimization with data-driven priors from fast feed-forward prediction methods. Experiments on urban-driving and drone datasets show that G3R generalizes across diverse large scenes and accelerates the reconstruction process by at least 10x while achieving comparable or better realism compared to 3DGS, and also being more robust to large view changes.

Cunliang Kong, Xuezhi Fang, Liner Yang et al.

Dictionaries can help language learners to learn vocabulary by providing definitions of words. Since traditional dictionaries present word senses as discrete items in predefined inventories, they fall short of flexibility, which is required in providing specific meanings of words in particular contexts. In this paper, we introduce the LitMind Dictionary (https://dictionary.litmind.ink), an open-source online generative dictionary that takes a word and context containing the word as input and automatically generates a definition as output. Incorporating state-of-the-art definition generation models, it supports not only Chinese and English, but also Chinese-English cross-lingual queries. Moreover, it has a user-friendly front-end design that can help users understand the query words quickly and easily. All the code and data are available at https://github.com/blcuicall/litmind-dictionary.

Peng Chen, Zhimin Chen, Pu Miao et al.

Reconfigurable Intelligent Surfaces (RIS) emerge as promising technologies in future radar and wireless communication domains. This letter addresses the passive sensing issue utilizing wireless communication signals and RIS amidst interference from wireless access points (APs). We introduce an atomic norm minimization (ANM) approach to leverage spatial domain target sparsity and estimate the direction of arrival (DOA). However, the conventional semidefinite programming (SDP)-based solutions for the ANM problem are complex and lack efficient realization. Consequently, we propose a RIS-ADMM method, an innovative alternating direction method of multipliers (ADMM)-based iterative approach. This method yields closed-form expressions and effectively suppresses interference signals. Simulation outcomes affirm that our RIS-ADMM method surpasses existing techniques in DOA estimation accuracy while maintaining low computational complexity. The code for the proposed method is available online \url{https://github.com/chenpengseu/RIS-ADMM.git}.

Cunliang Kong, Yun Chen, Hengyuan Zhang et al.

The definition generation task can help language learners by providing explanations for unfamiliar words. This task has attracted much attention in recent years. We propose a novel task of Simple Definition Generation (SDG) to help language learners and low literacy readers. A significant challenge of this task is the lack of learner's dictionaries in many languages, and therefore the lack of data for supervised training. We explore this task and propose a multitasking framework SimpDefiner that only requires a standard dictionary with complex definitions and a corpus containing arbitrary simple texts. We disentangle the complexity factors from the text by carefully designing a parameter sharing scheme between two decoders. By jointly training these components, the framework can generate both complex and simple definitions simultaneously. We demonstrate that the framework can generate relevant, simple definitions for the target words through automatic and manual evaluations on English and Chinese datasets. Our method outperforms the baseline model by a 1.77 SARI score on the English dataset, and raises the proportion of the low level (HSK level 1-3) words in Chinese definitions by 3.87%.

Yun Chen, Jiaping Xiao

Collaborative heterogeneous robot systems can greatly improve the efficiency of target search and navigation tasks. In this paper, we design a heterogeneous robot system consisting of a UAV and a UGV for search and rescue missions in unknown environments. The system is able to search for targets and navigate to them in a maze-like mine environment with the policies learned through deep reinforcement learning algorithms. During the training process, if two robots are trained simultaneously, the rewards related to their collaboration may not be properly obtained. Hence, we introduce a multi-stage reinforcement learning framework and a curiosity module to encourage agents to explore unvisited environments. Experiments in simulation environments show that our framework can train the heterogeneous robot system to achieve the search and navigation with unknown target locations while existing baselines may not, and accelerate the training speed.

Jingkang Wang, Henry Che, Yun Chen et al.

Reconstructing large-scale dynamic scenes from visual observations is a fundamental challenge in computer vision, with critical implications for robotics and autonomous systems. While recent differentiable rendering methods such as Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have achieved impressive photorealistic reconstruction, they suffer from scalability limitations and require annotations to decouple actor motion. Existing self-supervised methods attempt to eliminate explicit annotations by leveraging motion cues and geometric priors, yet they remain constrained by per-scene optimization and sensitivity to hyperparameter tuning. In this paper, we introduce Flux4D, a simple and scalable framework for 4D reconstruction of large-scale dynamic scenes. Flux4D directly predicts 3D Gaussians and their motion dynamics to reconstruct sensor observations in a fully unsupervised manner. By adopting only photometric losses and enforcing an "as static as possible" regularization, Flux4D learns to decompose dynamic elements directly from raw data without requiring pre-trained supervised models or foundational priors simply by training across many scenes. Our approach enables efficient reconstruction of dynamic scenes within seconds, scales effectively to large datasets, and generalizes well to unseen environments, including rare and unknown objects. Experiments on outdoor driving datasets show Flux4D significantly outperforms existing methods in scalability, generalization, and reconstruction quality.

Weikang Wang, Guanhua Chen, Hanqing Wang et al.

Multilingual pretrained language models (mPLMs) have shown their effectiveness in multilingual word alignment induction. However, these methods usually start from mBERT or XLM-R. In this paper, we investigate whether multilingual sentence Transformer LaBSE is a strong multilingual word aligner. This idea is non-trivial as LaBSE is trained to learn language-agnostic sentence-level embeddings, while the alignment extraction task requires the more fine-grained word-level embeddings to be language-agnostic. We demonstrate that the vanilla LaBSE outperforms other mPLMs currently used in the alignment task, and then propose to finetune LaBSE on parallel corpus for further improvement. Experiment results on seven language pairs show that our best aligner outperforms previous state-of-the-art models of all varieties. In addition, our aligner supports different language pairs in a single model, and even achieves new state-of-the-art on zero-shot language pairs that does not appear in the finetuning process.

Xin Guo, Haotian Xia, Zhaowei Liu et al.

Large language models have demonstrated outstanding performance in various natural language processing tasks, but their security capabilities in the financial domain have not been explored, and their performance on complex tasks like financial agent remains unknown. This paper presents FinEval, a benchmark designed to evaluate LLMs' financial domain knowledge and practical abilities. The dataset contains 8,351 questions categorized into four different key areas: Financial Academic Knowledge, Financial Industry Knowledge, Financial Security Knowledge, and Financial Agent. Financial Academic Knowledge comprises 4,661 multiple-choice questions spanning 34 subjects such as finance and economics. Financial Industry Knowledge contains 1,434 questions covering practical scenarios like investment research. Financial Security Knowledge assesses models through 1,640 questions on topics like application security and cryptography. Financial Agent evaluates tool usage and complex reasoning with 616 questions. FinEval has multiple evaluation settings, including zero-shot, five-shot with chain-of-thought, and assesses model performance using objective and subjective criteria. Our results show that Claude 3.5-Sonnet achieves the highest weighted average score of 72.9 across all financial domain categories under zero-shot setting. Our work provides a comprehensive benchmark closely aligned with Chinese financial domain.

Yan Yang, Zeguan Xiao, Xin Lu et al.

The widespread applications of large language models (LLMs) have brought about concerns regarding their potential misuse. Although aligned with human preference data before release, LLMs remain vulnerable to various malicious attacks. In this paper, we adopt a red-teaming strategy to enhance LLM safety and introduce SeqAR, a simple yet effective framework to design jailbreak prompts automatically. The SeqAR framework generates and optimizes multiple jailbreak characters and then applies sequential jailbreak characters in a single query to bypass the guardrails of the target LLM. Different from previous work which relies on proprietary LLMs or seed jailbreak templates crafted by human expertise, SeqAR can generate and optimize the jailbreak prompt in a cold-start scenario using open-sourced LLMs without any seed jailbreak templates. Experimental results show that SeqAR achieves attack success rates of 88% and 60% in bypassing the safety alignment of GPT-3.5-1106 and GPT-4, respectively. Furthermore, we extensively evaluate the transferability of the generated templates across different LLMs and held-out malicious requests, while also exploring defense strategies against the jailbreak attack designed by SeqAR.

Wentao Gao, Jiuyong Li, Debo Cheng et al.

Global Climate Models (GCMs) are crucial for predicting future climate changes by simulating the Earth systems. However, the GCM Outputs exhibit systematic biases due to model uncertainties, parameterization simplifications, and inadequate representation of complex climate phenomena. Traditional bias correction methods, which rely on historical observation data and statistical techniques, often neglect unobserved confounders, leading to biased results. This paper proposes a novel bias correction approach to utilize both GCM and observational data to learn a factor model that captures multi-cause latent confounders. Inspired by recent advances in causality based time series deconfounding, our method first constructs a factor model to learn latent confounders from historical data and then applies them to enhance the bias correction process using advanced time series forecasting models. The experimental results demonstrate significant improvements in the accuracy of precipitation outputs. By addressing unobserved confounders, our approach offers a robust and theoretically grounded solution for climate model bias correction.

Wentao Gao, Ziqi Xu, Jiuyong Li et al.

As the growing demand for long sequence time-series forecasting in real-world applications, such as electricity consumption planning, the significance of time series forecasting becomes increasingly crucial across various domains. This is highlighted by recent advancements in representation learning within the field. This study introduces a novel multi-view approach for time series forecasting that innovatively integrates trend and seasonal representations with an Independent Component Analysis (ICA)-based representation. Recognizing the limitations of existing methods in representing complex and high-dimensional time series data, this research addresses the challenge by combining TS (trend and seasonality) and ICA (independent components) perspectives. This approach offers a holistic understanding of time series data, going beyond traditional models that often miss nuanced, nonlinear relationships. The efficacy of TSI model is demonstrated through comprehensive testing on various benchmark datasets, where it shows superior performance over current state-of-the-art models, particularly in multivariate forecasting. This method not only enhances the accuracy of forecasting but also contributes significantly to the field by providing a more in-depth understanding of time series data. The research which uses ICA for a view lays the groundwork for further exploration and methodological advancements in time series forecasting, opening new avenues for research and practical applications.

Moyu Zhang, Yun Chen, Yujun Jin et al.

Generative pre-training via discrete diffusion provides dense reconstruction supervision across all feature fields simultaneously, mitigating representation collapse from data sparsity in CTR prediction. However, all existing generative CTR methods share a fundamental limitation: the reconstruction objective assigns equal training weight to every feature field, ignoring the profound heterogeneity of reconstruction difficulty across high-cardinality ID fields, sparse categorical attributes, numerical values, and behavioral sequences. This causes easy fields to dominate training gradients while the hardest but most informative fields remain chronically underfit, a problem we term the generative difficulty imbalance.We propose HeteGenCTR, which resolves this imbalance through per-field learnable difficulty parameters jointly trained with the denoising network. This unified signal drives two coordinated components without additional hyperparameters: a self-balancing loss that automatically reallocates gradient budget toward harder fields with a provably stable equilibrium, and a difficulty-guided attention mechanism that suppresses the influence of already-converged easy fields while amplifying cross-field information flow toward hard fields. Both components share the same learned signal and remain mutually consistent throughout training. Experiments on five CTR benchmarks and a seven-day online A/B test demonstrate consistent, statistically significant improvements over state-of-the-art baselines, with disproportionate gains for cold-start and long-tail users.

Moyu Zhang, Yun Chen, Yujun Jin et al.

Scaling test-time compute has proven highly effective for language models, yet this opportunity remains largely unexplored for industrial Click-Through Rate (CTR) prediction. CTR models suffer from a fundamental asymmetry: feature combinations well-represented in training yield confident predictions, while sparsely observed ones produce unreliable outputs. Existing training-phase solutions such as adaptive gating learn a fixed selection function subject to the same sparsity, offering no per-instance recourse at deployment.We propose UTTSI (Uncertainty-Triggered Test-Time Selective Inference), a training-free model-agnostic framework that scales inference depth proportionally to per-instance uncertainty. A dual-signal estimator combining model logit confidence with a data-level frequency prior distinguishes epistemic uncertainty from aleatoric ambiguity. Every instance undergoes adaptive feature filtering to remove unreliable embeddings; uncertain instances additionally receive stochastic feature-path explorations whose predictions are aggregated via consistency-weighted ensembling. Confident instances bypass exploration entirely, keeping average overhead at approximately $2.8\times$ base model cost with worst-case latency unchanged.Experiments on four datasets with three backbone architectures demonstrate consistent, statistically significant gains over all training-phase baselines. A seven-day online A/B test further confirms a 5.3% relative CTR gain ($p < 0.01$), establishing selective test-time compute allocation as a practical complement to training-phase advances for CTR prediction.

Zeguan Xiao, Yan Yang, Guanhua Chen et al.

Extensive efforts have been made before the public release of Large language models (LLMs) to align their behaviors with human values. However, even meticulously aligned LLMs remain vulnerable to malicious manipulations such as jailbreaking, leading to unintended behaviors. In this work, we propose a novel black-box jailbreak framework for automated red teaming of LLMs. We designed malicious content concealing and memory reframing with an iterative optimization algorithm to jailbreak LLMs, motivated by the research about the distractibility and over-confidence phenomenon of LLMs. Extensive experiments of jailbreaking both open-source and proprietary LLMs demonstrate the superiority of our framework in terms of effectiveness, scalability and transferability. We also evaluate the effectiveness of existing jailbreak defense methods against our attack and highlight the crucial need to develop more effective and practical defense strategies.

Zhaowei Liu, Xin Guo, Fangqi Lou et al.

Reasoning large language models are rapidly evolving across various domains. However, their capabilities in handling complex financial tasks still require in-depth exploration. In this paper, we introduce Fin-R1, a reasoning large language model specifically designed for the financial sector. Fin-R1 is built using a two-stage architecture, leveraging a financial reasoning dataset distilled and processed based on DeepSeek-R1. Through supervised fine-tuning (SFT) and reinforcement learning (RL) training, it demonstrates performance close to DeepSeek-R1 with a parameter size of 7 billion across a range of financial reasoning tasks. It achieves the state-of-the-art (SOTA) in the FinQA and ConvFinQA tasks between those LLMs in our evaluation, surpassing larger models in other tasks as well. Fin-R1 showcases strong reasoning and decision-making capabilities, providing solutions to various problems encountered in the financial domain. Our code is available at https://github.com/SUFE-AIFLM-Lab/Fin-R1.

Zeguan Xiao, Siqing Li, Yong Wang et al.

Machine unlearning for large language models (LLMs) aims to remove targeted knowledge while preserving general capability. In this paper, we recast LLM unlearning as an asymmetric two-task problem: retention is the primary objective and forgetting is an auxiliary. From this perspective, we propose a retention-prioritized gradient synthesis framework that decouples task-specific gradient extraction from conflict-aware combination. Instantiating the framework, we adapt established PCGrad to resolve gradient conflicts, and introduce SAGO, a novel retention-prioritized gradient synthesis method. Theoretically, both variants ensure non-negative cosine similarity with the retain gradient, while SAGO achieves strictly tighter alignment through constructive sign-constrained synthesis. Empirically, on WMDP Bio/Cyber and RWKU benchmarks, SAGO consistently pushes the Pareto frontier: e.g., on WMDP Bio (SimNPO+GD), recovery of target model MMLU performance progresses from 44.6% (naive) to 94.0% (+PCGrad) and further to 96.0% (+SAGO), while maintaining comparable forgetting strength. Our results show that re-shaping gradient geometry, rather than re-balancing losses, is the key to mitigating unlearning-retention trade-offs.

Tianci Xue, Ziqi Wang, Yixia Li et al.

Instruction tuning enhances the instruction following ability of large language models by finetuning with supervised instruction data. Previous work proposes in-context instruction tuning (ICIT) where specific positive or negative examples are incorporated into the prompt for better performance. In this work, we propose PACIT, a simple and effective in-context instruction tuning method, inspired by the pedagogical concept of desirable difficulty. The PACIT method unlocks the power of examples by encouraging the model to actively learn to grasp the distinctions between the positive and negative examples instead of merely reading. The model is expected to first verify the correctness of the provided example according to the task description, which is then set as the condition for generating a better response to the task instance. Our extensive experiments prove the effectiveness of PACIT, outperforming ICIT baseline on both in-domain and out-domain tasks up to 9.16 and 3.14 average ROUGE-L scores, respectively. Moreover, PACIT can notably enhance the performance of instruction tuning even when all positive and negative examples are generated with a self-instruct method.

Henry Che, Jingkang Wang, Yun Chen et al.

Urban scene reconstruction from real-world observations has emerged as a powerful tool for self-driving development and testing. While current neural rendering approaches achieve high-fidelity rendering along the recorded trajectories, their quality degrades significantly under large viewpoint shifts, limiting the applicability for closed-loop simulation. Recent works have shown promising results in using diffusion models to enhance quality at these challenging viewpoints and distill improvements back into 3D representations. However, they often require costly per-scene optimization, and the distilled representations remain fragile and fail to generalize beyond limited synthesized views. To address these limitations, we propose GenRe, a novel diffusion-guided generalizable enhancer for urban scene reconstruction. GenRe takes as input any pretrained 3D Gaussian representation and fixes the deficiencies within a few minutes. By learning to distill generative priors across diverse scenes, GenRe produces robust and high-fidelity representation efficiently that generalizes reliably to challenging unseen viewpoints (e.g., lane change). Experiments show that GenRe outperforms existing methods in both quality and efficiency and benefits various downstream tasks, enabling robust and scalable sensor simulation for autonomous driving.

Zeguan Xiao, Lang Mo, Yun Chen et al.

Large Language Models (LLMs) often memorize sensitive or harmful information, necessitating effective machine unlearning techniques. While existing parameter-efficient unlearning methods have shown promise, they still struggle with the forget-retain trade-off. This can be attributed to their reliance on parameter importance metrics to identify parameters that are important exclusively for the forget set, which is fundamentally limited by the superposition phenomenon. Due to the polysemantic nature of LLM parameters, such an importance metric may struggle to disentangle parameters associated with the forget and retain sets. In this work, we propose Representation-Guided Low-rank Unlearning (REGLU), a novel approach that leverages the geometric properties of representation spaces to achieve robust and precise unlearning. First, we develop a representation-guided initialization for LoRA that identifies the optimal subspace for selective forgetting. Second, we introduce a regularization loss that constrains the outputs of the LoRA update to lie in the orthogonal complement of the retain set's representation subspace, thereby minimizing interference with the model's performance on the retain set. We evaluate REGLU on the TOFU and WMDP benchmarks across multiple models. Our results demonstrate that REGLU consistently outperforms state-of-the-art baselines, achieving superior unlearning quality while maintaining higher model utility.

Jeonghyeok Do, Yun Chen, Geunhyuk Youk et al.

The landscape of skeleton-based action representation learning has evolved from Contrastive Learning (CL) to Masked Auto-Encoder (MAE) architectures. However, each paradigm faces inherent limitations: CL often overlooks fine-grained local details, while MAE is burdened by computationally heavy decoders. Moreover, MAE suffers from severe computational asymmetry -- benefiting from efficient masking during pre-training but requiring exhaustive full-sequence processing for downstream tasks. To resolve these bottlenecks, we propose SLiM (Skeleton Less is More), a novel unified framework that harmonizes masked modeling with contrastive learning via a shared encoder. By eschewing the reconstruction decoder, SLiM not only eliminates computational redundancy but also compels the encoder to capture discriminative features directly. SLiM is the first framework with decoder-free masked modeling of representative learning. Crucially, to prevent trivial reconstruction arising from high skeletal-temporal correlation, we introduce semantic tube masking, alongside skeletal-aware augmentations designed to ensure anatomical consistency across diverse temporal granularities. Extensive experiments demonstrate that SLiM consistently achieves state-of-the-art performance across all downstream protocols. Notably, our method delivers this superior accuracy with exceptional efficiency, reducing inference computational cost by 7.89x compared to existing MAE methods.

Jinran Nie, Liner Yang, Yun Chen et al.

Text generation rarely considers the control of lexical complexity, which limits its more comprehensive practical application. We introduce a novel task of lexical complexity controlled sentence generation, which aims at keywords to sentence generation with desired complexity levels. It has enormous potential in domains such as grade reading, language teaching and acquisition. The challenge of this task is to generate fluent sentences only using the words of given complexity levels. We propose a simple but effective approach for this task based on complexity embedding. Compared with potential solutions, our approach fuses the representations of the word complexity levels into the model to get better control of lexical complexity. And we demonstrate the feasibility of the approach for both training models from scratch and fine-tuning the pre-trained models. To facilitate the research, we develop two datasets in English and Chinese respectively, on which extensive experiments are conducted. Results show that our approach better controls lexical complexity and generates higher quality sentences than baseline methods.

Zhiwen Ruan, Yichao Du, Jianjie Zheng et al.

A promising paradigm for adapting instruction-tuned language models is to learn task-specific updates on a pretrained base model and subsequently merge them into the instruction-tuned model. However, existing approaches typically treat the instruction-tuned model as a passive target that is only involved at the final merging stage, without guiding the training process. We propose GIFT (Guided Fine-Tuning and Transfer), a simple and efficient framework that incorporates guidance from the instruction model into task adaptation. GIFT fine-tunes a low-rank adapter on the pretrained base model using confidence signals derived from the instruction-tuned model. The learned adapter is then merged into the instruction-tuned model, yielding task-specialized models that preserve general instruction-following behavior. We evaluate GIFT on mathematical and knowledge-intensive benchmarks across multiple model families and scales. Results show that GIFT consistently outperforms direct fine-tuning and representative transfer-based baselines, while maintaining robust generalization and favorable test-time scaling behavior.

Zhiwen Ruan, Yixia Li, Yefeng Liu et al.

Large Language Models (LLMs) have demonstrated exceptional performance across diverse natural language processing tasks. However, these models exhibit a critical limitation in output diversity, often generating highly similar content across multiple attempts. This limitation significantly affects tasks requiring diverse outputs, from creative writing to reasoning. Existing solutions, like temperature scaling, enhance diversity by modifying probability distributions but compromise output quality. We propose Guide-to-Generation (G2), a training-free plug-and-play method that enhances output diversity while preserving generation quality. G2 employs a base generator alongside dual Guides, which guide the generation process through decoding-based interventions to encourage more diverse outputs conditioned on the original query. Comprehensive experiments demonstrate that G2 effectively improves output diversity while maintaining an optimal balance between diversity and quality.

Hanqing Wang, Yajing Luo, Boya Xiong et al.

Stylistic headline generation is the task to generate a headline that not only summarizes the content of an article, but also reflects a desired style that attracts users. As style-specific article-headline pairs are scarce, previous researches focus on unsupervised approaches with a standard headline generation dataset and mono-style corpora. In this work, we follow this line and propose StyleBART, an unsupervised approach for stylistic headline generation. Our method decorates the pretrained BART model with adapters that are responsible for different styles and allows the generation of headlines with diverse styles by simply switching the adapters. Different from previous works, StyleBART separates the task of style learning and headline generation, making it possible to freely combine the base model and the style adapters during inference. We further propose an inverse paraphrasing task to enhance the style adapters. Extensive automatic and human evaluations show that StyleBART achieves new state-of-the-art performance in the unsupervised stylistic headline generation task, producing high-quality headlines with the desired style.

Yang Liu, Meng Xu, Shuo Wang et al.

Modern large language models (LLMs) should generally benefit individuals from various cultural backgrounds around the world. However, most recent advanced generative evaluation benchmarks tailed for LLMs mainly focus on English. To this end, we introduce OMGEval, the first Open-source Multilingual Generative test set that can assess the capability of LLMs in different languages. For each language, OMGEval provides 804 open-ended questions, covering a wide range of important capabilities of LLMs, such as general knowledge, logical reasoning, and so on. Each question is rigorously verified by human annotators. Notably, to sufficiently reflect the compatibility of LLMs in different cultural backgrounds, we perform localization for each non-English language. Specifically, the current version of OMGEval includes 5 languages (i.e., Zh, Ru, Fr, Es, Ar). Following AlpacaEval, we employ GPT-4 as the adjudicator to automatically score different model outputs, which is shown closely related to human evaluation. We evaluate several representative multilingual LLMs on the proposed OMGEval, which we believe will provide a valuable reference for the community to further understand and improve the multilingual capability of LLMs. OMGEval is available at https://github.com/blcuicall/OMGEval.

Yun Chen, Moyu Zhang, Jinxin Hu et al.

Neural Architecture Search (NAS) deployment in industrial production systems faces a fundamental validation bottleneck: verifying a single candidate architecture pi requires evaluating the deployed ensemble of M models, incurring prohibitive O(M) computational cost per candidate. This cost barrier severely limits architecture iteration frequency in real-world applications where ensembles (M=50-200) are standard for robustness. This work introduces Ensemble-Decoupled Architecture Search, a framework that leverages ensemble theory to predict system-level performance from single-learner evaluation. We establish the Ensemble-Decoupled Theory with a sufficient condition for monotonic ensemble improvement under homogeneity assumptions: a candidate architecture pi yields lower ensemble error than the current baseline if rho(pi) < rho(pi_old) - (M / (M - 1)) * (Delta E(pi) / sigma^2(pi)), where Delta E, rho, and sigma^2 are estimable from lightweight dual-learner training. This decouples architecture search from full ensemble training, reducing per-candidate search cost from O(M) to O(1) while maintaining O(M) deployment cost only for validated winners. We unify solution strategies across pipeline continuity: (1) closed-form optimization for tractable continuous pi (exemplified by feature bagging in CTR prediction), (2) constrained differentiable optimization for intractable continuous pi, and (3) LLM-driven search with iterative monotonic acceptance for discrete pi. The framework reveals two orthogonal improvement mechanisms -- base diversity gain and accuracy gain -- providing actionable design principles for industrial-scale NAS. All theoretical derivations are rigorous with detailed proofs deferred to the appendix. Comprehensive empirical validation will be included in the journal extension of this work.

Yutao Hou, Yajing Luo, Zhiwen Ruan et al.

Large language models (LLMs) demonstrate remarkable performance across various tasks, prompting researchers to develop diverse evaluation benchmarks. However, existing benchmarks typically measure the ability of LLMs to respond to individual questions, neglecting the complex interactions in real-world applications. In this paper, we introduce Compound Question Synthesis (CQ-Syn) to create the Compound-QA benchmark, focusing on compound questions with multiple sub-questions. This benchmark is derived from existing QA datasets, annotated with proprietary LLMs and verified by humans for accuracy. It encompasses five categories: Factual-Statement, Cause-and-Effect, Hypothetical-Analysis, Comparison-and-Selection, and Evaluation-and-Suggestion. It evaluates the LLM capability in terms of three dimensions including understanding, reasoning, and knowledge. Our assessment of eight open-source LLMs using Compound-QA reveals distinct patterns in their responses to compound questions, which are significantly poorer than those to non-compound questions. Additionally, we investigate various methods to enhance LLMs performance on compound questions. The results indicate that these approaches significantly improve the models' comprehension and reasoning abilities on compound questions.

Junyou Su, He Zhu, Xiao Luo et al.

Supervised fine-tuning (SFT) is fundamental to adapting large language models, yet training on complete datasets incurs prohibitive costs with diminishing returns. Existing data selection methods suffer from severe domain specificity: techniques optimized for general instruction-following fail on reasoning tasks, and vice versa. We observe that measuring entropy differences between base models and minimally instruction-tuned calibrated models reveals a pattern -- samples with the lowest differential entropy consistently yield optimal performance across domains, yet this principle manifests domain-adaptively: reasoning tasks favor entropy increase (cognitive expansion), while general tasks favor entropy decrease (cognitive compression). We introduce InstructDiff, a unified framework that operationalizes differential entropy as a domain-adaptive selection criterion through warmup calibration, bi-directional NLL filtering, and entropy-based ranking. Extensive experiments show that InstructDiff achieves 17\% relative improvement over full data training on mathematical reasoning and 52\% for general instruction-following, outperforming prior baselines while using only 10\% of the data.

Zeguan Xiao, Xuanzhe Xu, Yun Chen et al.

Large language model (LLM) unlearning aims to remove specific data influences from pre-trained model without costly retraining, addressing privacy, copyright, and safety concerns. However, recent studies reveal a critical vulnerability: unlearned models rapidly recover "forgotten" knowledge through relearning attacks. This fragility raises serious security concerns, especially for open-weight models. In this work, we investigate the fundamental mechanism underlying this fragility from a representation geometry perspective. We discover that existing unlearning methods predominantly optimize along dominant components, leaving minor components largely unchanged. Critically, during relearning attacks, the modifications in these dominant components are easily reversed, enabling rapid knowledge recovery, whereas minor components exhibit stronger resistance to such reversal. We further provide a theoretical analysis that explains both observations from the spectral structure of representations. Building on this insight, we propose Minor Component Unlearning (MCU), a novel unlearning approach that explicitly targets minor components in representations. By concentrating unlearning effects in these inherently robust directions, our method achieves substantially improved resistance to relearning attacks. Extensive experiments on three datasets validate our approach, demonstrating significant improvements over state-of-the-art methods including sharpness-aware minimization.

Xiaowen Ma, Shuning Ge, Fan Yang et al.

Transformer-based architectures dominate time series modeling by enabling global attention over all timestamps, yet their rigid 'one-size-fits-all' context aggregation fails to address two critical challenges in real-world data: (1) inherent lag effects, where the relevance of historical timestamps to a query varies dynamically; (2) anomalous segments, which introduce noisy signals that degrade forecasting accuracy. To resolve these problems, we propose the Temporal Mix of Experts (TMOE), a novel attention-level mechanism that reimagines key-value (K-V) pairs as local experts (each specialized in a distinct temporal context) and performs adaptive expert selection for each query via localized filtering of irrelevant timestamps. Complementing this local adaptation, a shared global expert preserves the Transformer's strength in capturing long-range dependencies. We then replace the vanilla attention mechanism in popular time-series Transformer frameworks (i.e., PatchTST and Timer) with TMOE, without extra structural modifications, yielding our specific version TimeExpert and general version TimeExpert-G. Extensive experiments on seven real-world long-term forecasting benchmarks demonstrate that TimeExpert and TimeExpert-G outperform state-of-the-art methods. Code is available at https://github.com/xwmaxwma/TimeExpert.

Tianyi Wang, Long Li, Hongcan Guo et al.

Reinforcement Learning with Verifiable Rewards (RLVR) is increasingly viewed as a tree pruning mechanism. However, we identify a systemic pathology termed Recursive Space Contraction (RSC), an irreversible collapse driven by the combined dynamics of positive sharpening and negative squeezing, where the sampling probability of valid alternatives vanishes. While Kullback-Leibler (KL) regularization aims to mitigate this, it imposes a rigid Shape Matching constraint that forces the policy to mimic the reference model's full density, creating a gradient conflict with the sharpening required for correctness. We propose Anchored Policy Optimization (APO), shifting the paradigm from global Shape Matching to Support Coverage. By defining a Safe Manifold based on the reference model's high-confidence support, APO permits aggressive sharpening for efficiency while selectively invoking a restorative force during error correction to prevent collapse. We theoretically derive that APO serves as a gradient-aligned mechanism to maximize support coverage, enabling an Elastic Recovery that re-inflates valid branches. Empirical evaluations on mathematical benchmarks demonstrate that APO breaks the accuracy-diversity trade-off, significantly improving Pass@1 while restoring the Pass@K diversity typically lost by standard policy gradient methods.

Xin Ding, Yun Chen, Sen Zhang et al.

Continuous Conditional Diffusion Model (CCDM) is a diffusion-based framework designed to generate high-quality images conditioned on continuous regression labels. Although CCDM has demonstrated clear advantages over prior approaches across a range of datasets, it still exhibits notable limitations and has recently been surpassed by a GAN-based method, namely CcGAN-AVAR. These limitations mainly arise from its reliance on an outdated diffusion framework and its low sampling efficiency due to long sampling trajectories. To address these issues, we propose an improved CCDM framework, termed iCCDM, which incorporates the more advanced \textit{Elucidated Diffusion Model} (EDM) framework with substantial modifications to improve both generation quality and sampling efficiency. Specifically, iCCDM introduces a novel matrix-form EDM formulation together with an adaptive vicinal training strategy. Extensive experiments on four benchmark datasets, spanning image resolutions from $64\times64$ to $256\times256$, demonstrate that iCCDM consistently outperforms existing methods, including state-of-the-art large-scale text-to-image diffusion models (e.g., Stable Diffusion 3, FLUX.1, and Qwen-Image), achieving higher generation quality while significantly reducing sampling cost.

Ze Yang, Jingkang Wang, Haowei Zhang et al.

High-quality 3D assets for traffic participants are critical for multi-sensor simulation, which is essential for the safe end-to-end development of autonomy. Building assets from in-the-wild data is key for diversity and realism, but existing neural-rendering based reconstruction methods are slow and generate assets that render well only from viewpoints close to the original observations, limiting their usefulness in simulation. Recent diffusion-based generative models build complete and diverse assets, but perform poorly on in-the-wild driving scenes, where observed actors are captured under sparse and limited fields of view, and are partially occluded. In this work, we propose a 3D latent diffusion model that learns on in-the-wild LiDAR and camera data captured by a sensor platform and generates high-quality 3D assets with complete geometry and appearance. Key to our method is a "reconstruct-then-generate" approach that first leverages occlusion-aware neural rendering trained over multiple scenes to build a high-quality latent space for objects, and then trains a diffusion model that operates on the latent space. We show our method outperforms existing reconstruction and generation based methods, unlocking diverse and scalable content creation for simulation.

Yutao Hou, Yihan Jiang, Yuhan Xie et al.

Large language models (LLMs) are increasingly applied in financial scenarios. However, they may produce harmful outputs, including facilitating illegal activities or unethical behavior, posing serious compliance risks. To systematically evaluate LLM safety in finance, we propose FinSafetyBench, a bilingual (English-Chinese) red-teaming benchmark designed to test an LLM's refusal of requests that violate financial compliance. Grounded in real-world financial crime cases and ethics standards, the benchmark comprises 14 subcategories spanning financial crimes and ethical violations. Through extensive experiments on general-purpose and finance-specialized LLMs under three representative attack settings, we identify critical vulnerabilities that allow adversarial prompts to bypass compliance safeguards. Further analysis reveals stronger susceptibility in Chinese contexts and highlights the limitations of prompt-level defenses against sophisticated or implicit manipulation strategies.

Hanqing Wang, Bowen Ping, Shuo Wang et al.

LoRA employs lightweight modules to customize large language models (LLMs) for each downstream task or domain, where different learned additional modules represent diverse skills. Combining existing LoRAs to address new tasks can enhance the reusability of learned LoRAs, particularly beneficial for tasks with limited annotated data. Most prior works on LoRA combination primarily rely on task-level weights for each involved LoRA, making different examples and tokens share the same LoRA weights. However, in generative tasks, different tokens may necessitate diverse skills to manage. Taking the Chinese math task as an example, understanding the problem description may depend more on the Chinese LoRA, while the calculation part may rely more on the math LoRA. To this end, we propose LoRA-Flow, which utilizes dynamic weights to adjust the impact of different LoRAs. The weights at each step are determined by a fusion gate with extremely few parameters, which can be learned with only 200 training examples. Experiments across six generative tasks demonstrate that our method consistently outperforms baselines with task-level fusion weights. This underscores the necessity of introducing dynamic fusion weights for LoRA combination.

Peng Lai, Zhihao Ou, Yong Wang et al.

LLM-as-a-Judge has been widely adopted across various research and practical applications, yet the robustness and reliability of its evaluation remain a critical issue. A core challenge it faces is bias, which has primarily been studied in terms of known biases and their impact on evaluation outcomes, while automated and systematic exploration of potential unknown biases is still lacking. Nevertheless, such exploration is crucial for enhancing the robustness and reliability of evaluations. To bridge this gap, we propose BiasScope, a LLM-driven framework for automatically and at scale discovering potential biases that may arise during model evaluation. BiasScope can uncover potential biases across different model families and scales, with its generality and effectiveness validated on the JudgeBench dataset. It overcomes the limitations of existing approaches, transforming bias discovery from a passive process relying on manual effort and predefined bias lists into an active and comprehensive automated exploration. Moreover, based on BiasScope, we propose JudgeBench-Pro, an extended version of JudgeBench and a more challenging benchmark for evaluating the robustness of LLM-as-a-judge. Strikingly, even powerful LLMs as evaluators show error rates above 50\% on JudgeBench-Pro, underscoring the urgent need to strengthen evaluation robustness and to mitigate potential biases further.

Ava Pun, Gary Sun, Jingkang Wang et al. · cmu

Different outdoor illumination conditions drastically alter the appearance of urban scenes, and they can harm the performance of image-based robot perception systems if not seen during training. Camera simulation provides a cost-effective solution to create a large dataset of images captured under different lighting conditions. Towards this goal, we propose LightSim, a neural lighting camera simulation system that enables diverse, realistic, and controllable data generation. LightSim automatically builds lighting-aware digital twins at scale from collected raw sensor data and decomposes the scene into dynamic actors and static background with accurate geometry, appearance, and estimated scene lighting. These digital twins enable actor insertion, modification, removal, and rendering from a new viewpoint, all in a lighting-aware manner. LightSim then combines physically-based and learnable deferred rendering to perform realistic relighting of modified scenes, such as altering the sun location and modifying the shadows or changing the sun brightness, producing spatially- and temporally-consistent camera videos. Our experiments show that LightSim generates more realistic relighting results than prior work. Importantly, training perception models on data generated by LightSim can significantly improve their performance.

Rongyu Zhang, Yun Chen, Chenrui Wu et al.

Federated learning (FL) offers a privacy-centric distributed learning framework, enabling model training on individual clients and central aggregation without necessitating data exchange. Nonetheless, FL implementations often suffer from non-i.i.d. and long-tailed class distributions across mobile applications, e.g., autonomous vehicles, which leads models to overfitting as local training may converge to sub-optimal. In our study, we explore the impact of data heterogeneity on model bias and introduce an innovative personalized FL framework, Multi-level Personalized Federated Learning (MuPFL), which leverages the hierarchical architecture of FL to fully harness computational resources at various levels. This framework integrates three pivotal modules: Biased Activation Value Dropout (BAVD) to mitigate overfitting and accelerate training; Adaptive Cluster-based Model Update (ACMU) to refine local models ensuring coherent global aggregation; and Prior Knowledge-assisted Classifier Fine-tuning (PKCF) to bolster classification and personalize models in accord with skewed local data with shared knowledge. Extensive experiments on diverse real-world datasets for image classification and semantic segmentation validate that MuPFL consistently outperforms state-of-the-art baselines, even under extreme non-i.i.d. and long-tail conditions, which enhances accuracy by as much as 7.39% and accelerates training by up to 80% at most, marking significant advancements in both efficiency and effectiveness.

Yun Chen, Bowei Huang, Fan Guo et al.

Autonomous mobile manipulation in unstructured warehouses requires a balance between efficient large-scale navigation and high-precision object interaction. Traditional end-to-end learning approaches often struggle to handle the conflicting demands of these distinct phases. Navigation relies on robust decision-making over large spaces, while manipulation needs high sensitivity to fine local details. Forcing a single network to learn these different objectives simultaneously often causes optimization interference, where improving one task degrades the other. To address these limitations, we propose a Heterogeneous Multi-Expert Reinforcement Learning (HMER) framework tailored for autonomous forklifts. HMER decomposes long-horizon tasks into specialized sub-policies controlled by a Semantic Task Planner. This structure separates macro-level navigation from micro-level manipulation, allowing each expert to focus on its specific action space without interference. The planner coordinates the sequential execution of these experts, bridging the gap between task planning and continuous control. Furthermore, to solve the problem of sparse exploration, we introduce a Hybrid Imitation-Reinforcement Training Strategy. This method uses expert demonstrations to initialize the policy and Reinforcement Learning for fine-tuning. Experiments in Gazebo simulations show that HMER significantly outperforms sequential and end-to-end baselines. Our method achieves a task success rate of 94.2\% (compared to 62.5\% for baselines), reduces operation time by 21.4\%, and maintains placement error within 1.5 cm, validating its efficacy for precise material handling.

Tianyi Wang, Yixia Li, Long Li et al.

Proximal Policy Optimization (PPO) is central to aligning Large Language Models (LLMs) in reasoning tasks with verifiable rewards. However, standard token-level PPO struggles in this setting due to the instability of temporal credit assignment over long Chain-of-Thought (CoT) horizons and the prohibitive memory cost of the value model. While critic-free alternatives like GRPO mitigate these issues, they incur significant computational overhead by requiring multiple samples for baseline estimation, severely limiting training throughput. In this paper, we introduce Sequence-Level PPO (SPPO), a scalable algorithm that harmonizes the sample efficiency of PPO with the stability of outcome-based updates. SPPO reformulates the reasoning process as a Sequence-Level Contextual Bandit problem, employing a decoupled scalar value function to derive low-variance advantage signals without multi-sampling. Extensive experiments on mathematical benchmarks demonstrate that SPPO significantly surpasses standard PPO and matches the performance of computation-heavy group-based methods, offering a resource-efficient framework for aligning reasoning LLMs.

Peihao Yan, Yun Chen, Jie Lu et al.

Emerging delay-critical edge AI applications, such as VR perception and real-time video analytics, impose stringent latency and reliability requirements on 5G networks. However, existing mobility management mechanisms are largely reactive and fail to adapt to dynamic network conditions, resulting in suboptimal handover decisions and degraded performance. In this paper, we present TARMM, a 5G Open Radio Access Network (O-RAN) system that optimizes user mobility management for delay-critical edge AI offloading. The core of TARMM is a temporal graph model that captures the spatiotemporal dynamics of the RAN across users and cells, enabling near real-time handover decisions. Building on this representation, we design a multi-agent reinforcement learning (MARL) framework with rule-based action masking and proactive resource preparation to ensure safe, stable, and efficient handovers. We implement TARMM on a multi-cell indoor 5G O-RAN testbed and evaluate it using diverse VR workloads. Extensive experiments show that TARMM reduces tail latency by up to 44% and packet loss by up to 56% compared to state-of-the-art approaches.

Yun Chen, Lingxiao Yang, Qi Chen et al.

Emotional Voice Conversion aims to manipulate a speech according to a given emotion while preserving non-emotion components. Existing approaches cannot well express fine-grained emotional attributes. In this paper, we propose an Attention-based Interactive diseNtangling Network (AINN) that leverages instance-wise emotional knowledge for voice conversion. We introduce a two-stage pipeline to effectively train our network: Stage I utilizes inter-speech contrastive learning to model fine-grained emotion and intra-speech disentanglement learning to better separate emotion and content. In Stage II, we propose to regularize the conversion with a multi-view consistency mechanism. This technique helps us transfer fine-grained emotion and maintain speech content. Extensive experiments show that our AINN outperforms state-of-the-arts in both objective and subjective metrics.

Zhiwen Ruan, Yixia Li, He Zhu et al.

Despite being pretrained on multilingual corpora, large language models (LLMs) exhibit suboptimal performance on low-resource languages. Recent approaches have leveraged multilingual encoders alongside LLMs by introducing trainable parameters connecting the two models. However, these methods typically focus on the encoder's output, overlooking valuable information from other layers. We propose \aname (\mname), a framework that integrates representations from all encoder layers, coupled with the \attaname mechanism to enable layer-wise interaction between the LLM and the multilingual encoder. Extensive experiments on multilingual reasoning tasks, along with analyses of learned representations, show that our approach consistently outperforms existing baselines.

Yutao Hou, Zeguan Xiao, Fei Yu et al.

Large language models (LLMs) have achieved distinguished performance on various reasoning-intensive tasks. However, LLMs might still face the challenges of robustness issues and fail unexpectedly in some simple reasoning tasks. Previous works evaluate the LLM robustness with hand-crafted templates or a limited set of perturbation rules, indicating potential data contamination in pre-training or fine-tuning datasets. In this work, inspired by stress testing in software engineering, we propose a novel framework, Automatic Robustness Checker (AR-Checker), to generate mathematical problem variants that maintain the semantic meanings of the original one but might fail the LLMs. The AR-Checker framework generates mathematical problem variants through multi-round parallel streams of LLM-based rewriting and verification. Our framework can generate benchmark variants dynamically for each LLM, thus minimizing the risk of data contamination. Experiments on GSM8K and MATH-500 demonstrate the strong performance of AR-Checker on mathematical tasks. We also evaluate AR-Checker on benchmarks beyond mathematics, including MMLU, MMLU-Pro, and CommonsenseQA, where it also achieves strong performance, further proving the effectiveness of AR-Checker.

Jinouwen Zhang, Rongkun Xue, Yazhe Niu et al.

Generative models, particularly diffusion models, have achieved remarkable success in density estimation for multimodal data, drawing significant interest from the reinforcement learning (RL) community, especially in policy modeling in continuous action spaces. However, existing works exhibit significant variations in training schemes and RL optimization objectives, and some methods are only applicable to diffusion models. In this study, we compare and analyze various generative policy training and deployment techniques, identifying and validating effective designs for generative policy algorithms. Specifically, we revisit existing training objectives and classify them into two categories, each linked to a simpler approach. The first approach, Generative Model Policy Optimization (GMPO), employs a native advantage-weighted regression formulation as the training objective, which is significantly simpler than previous methods. The second approach, Generative Model Policy Gradient (GMPG), offers a numerically stable implementation of the native policy gradient method. We introduce a standardized experimental framework named GenerativeRL. Our experiments demonstrate that the proposed methods achieve state-of-the-art performance on various offline-RL datasets, offering a unified and practical guideline for training and deploying generative policies.

Xu Li, Ruiqi Sun, Jiameng Lv et al.

Strong gravitational lensing is a powerful tool for investigating dark matter and dark energy properties. With the advent of large-scale sky surveys, we can discover strong lensing systems on an unprecedented scale, which requires efficient tools to extract them from billions of astronomical objects. The existing mainstream lens-finding tools are based on machine learning algorithms and applied to cut-out-centered galaxies. However, according to the design and survey strategy of optical surveys by CSST, preparing cutouts with multiple bands requires considerable efforts. To overcome these challenges, we have developed a framework based on a hierarchical visual Transformer with a sliding window technique to search for strong lensing systems within entire images. Moreover, given that multi-color images of strong lensing systems can provide insights into their physical characteristics, our framework is specifically crafted to identify strong lensing systems in images with any number of channels. As evaluated using CSST mock data based on an Semi-Analytic Model named CosmoDC2, our framework achieves precision and recall rates of 0.98 and 0.90, respectively. To evaluate the effectiveness of our method in real observations, we have applied it to a subset of images from the DESI Legacy Imaging Surveys and media images from Euclid Early Release Observations. 61 new strong lensing system candidates are discovered by our method. However, we also identified false positives arising primarily from the simplified galaxy morphology assumptions within the simulation. This underscores the practical limitations of our approach while simultaneously highlighting potential avenues for future improvements.