arXiv subjects

Cheng Luo

Publications and source records attributed to Cheng Luo.

At least 37 records · Page 2Linked to original sources

arXiv2026

LegalOne: A Family of Foundation Models for Reliable Legal Reasoning

While Large Language Models (LLMs) have demonstrated impressive general capabilities, their direct application in the legal domain is often hindered by a lack of precise domain knowledge and complexity of performing rigorous multi-step judicial reasoning. To address this gap, we present LegalOne, a family of foundational models specifically tailored for the Chinese legal domain. LegalOne is developed through a comprehensive three-phase pipeline designed to master legal reasoning. First, during mid-training phase, we propose Plasticity-Adjusted Sampling (PAS) to address the challenge of domain adaptation. This perplexity-based scheduler strikes a balance between the acquisition of new knowledge and the retention of original capabilities, effectively establishing a robust legal foundation. Second, during supervised fine-tuning, we employ Legal Agentic CoT Distillation (LEAD) to distill explicit reasoning from raw legal texts. Unlike naive distillation, LEAD utilizes an agentic workflow to convert complex judicial processes into structured reasoning trajectories, thereby enforcing factual grounding and logical rigor. Finally, we implement a Curriculum Reinforcement Learning (RL) strategy. Through a progressive reinforcement process spanning memorization, understanding, and reasoning, LegalOne evolves from simple pattern matching to autonomous and reliable legal reasoning. Experimental results demonstrate that LegalOne achieves state-of-the-art performance across a wide range of legal tasks, surpassing general-purpose LLMs with vastly larger parameter counts through enhanced knowledge density and efficiency. We publicly release the LegalOne weights and the LegalKit evaluation framework to advance the field of Legal AI, paving the way for deploying trustworthy and interpretable foundation models in high-stakes judicial applications.

cs.CL↗

arXiv2026

R-KV: Redundancy-aware KV Cache Compression for Reasoning Models

Reasoning models have demonstrated impressive performance in self-reflection and chain-of-thought reasoning. However, they often produce excessively long outputs, leading to prohibitively large key-value (KV) caches during inference. While chain-of-thought inference significantly improves performance on complex reasoning tasks, it can also lead to reasoning failures when deployed with existing KV cache compression approaches. To address this, we propose Redundancy-aware KV Cache Compression for Reasoning models (R-KV), a novel method specifically targeting redundant tokens in reasoning models. Our method preserves nearly 100% of the full KV cache performance using only 10% of the KV cache, substantially outperforming existing KV cache baselines, which reach only 60% of the performance. Remarkably, R-KV even achieves 105% of full KV cache performance with 16% of the KV cache. This KV-cache reduction also leads to a 90% memory saving and a 6.6X throughput over standard chain-of-thought reasoning inference. Experimental results show that R-KV consistently outperforms existing KV cache compression baselines across two mathematical reasoning datasets.

cs.CL↗

arXiv2025

AutoRec: Accelerating Loss Recovery for Live Streaming in a Multi-Supplier Market

Due to the limited permissions for upgrading dualside (i.e., server-side and client-side) loss tolerance schemes from the perspective of CDN vendors in a multi-supplier market, modern large-scale live streaming services are still using the automatic-repeat-request (ARQ) based paradigm for loss recovery, which only requires server-side modifications. In this paper, we first conduct a large-scale measurement study with up to 50 million live streams. We find that loss shows dynamics and live streaming contains frequent on-off mode switching in the wild. We further find that the recovery latency, enlarged by the ubiquitous retransmission loss, is a critical factor affecting live streaming's client-side QoE (e.g., video freezing). We then propose an enhanced recovery mechanism called AutoRec, which can transform the disadvantages of on-off mode switching into an advantage for reducing loss recovery latency without any modifications on the client side. AutoRec allows users to customize overhead tolerance and recovery latency tolerance and adaptively adjusts strategies as the network environment changes to ensure that recovery latency meets user demands whenever possible while keeping overhead under control. We implement AutoRec upon QUIC and evaluate it via testbed and real-world commercial services deployments. The experimental results demonstrate the practicability and profitability of AutoRec.

cs.NI↗

arXiv2025

ArcGen: Generalizing Neural Backdoor Detection Across Diverse Architectures

Backdoor attacks pose a significant threat to the security and reliability of deep learning models. To mitigate such attacks, one promising approach is to learn to extract features from the target model and use these features for backdoor detection. However, we discover that existing learning-based neural backdoor detection methods do not generalize well to new architectures not seen during the learning phase. In this paper, we analyze the root cause of this issue and propose a novel black-box neural backdoor detection method called ArcGen. Our method aims to obtain architecture-invariant model features, i.e., aligned features, for effective backdoor detection. Specifically, in contrast to existing methods directly using model outputs as model features, we introduce an additional alignment layer in the feature extraction function to further process these features. This reduces the direct influence of architecture information on the features. Then, we design two alignment losses to train the feature extraction function. These losses explicitly require that features from models with similar backdoor behaviors but different architectures are aligned at both the distribution and sample levels. With these techniques, our method demonstrates up to 42.5% improvements in detection performance (e.g., AUC) on unseen model architectures. This is based on a large-scale evaluation involving 16,896 models trained on diverse datasets, subjected to various backdoor attacks, and utilizing different model architectures. Our code is available at https://github.com/SeRAlab/ArcGen.

cs.LG↗

arXiv2025

An End-to-End Neural Network Transceiver Design for OFDM System with FPGA-Accelerated Implementation

The evolution toward sixth-generation (6G) wireless networks demands high-performance transceiver architectures capable of handling complex and dynamic environments. Conventional orthogonal frequency-division multiplexing (OFDM) receivers rely on cascaded discrete Fourier transform (DFT) and demodulation blocks, which are prone to inter-stage error propagation and suboptimal global performance. In this work, we propose two neural network (NN) models DFT-Net and Demodulation-Net (Demod-Net) to jointly replace the IDFT/DFT and demodulation modules in an OFDM transceiver. The models are trained end-to-end (E2E) to minimize bit error rate (BER) while preserving operator equivalence for hybrid deployment. A customized DFT-Demodulation Net Accelerator (DDNA) is further developed to efficiently map the proposed networks onto field-programmable gate array (FPGA) platforms. Leveraging fine-grained pipelining and block matrix operations, DDNA achieves high throughput and flexibility under stringent latency constraints. Experimental results show that the DL-based transceiver consistently outperforms the conventional OFDM system across multiple modulation schemes. With only a modest increase in hardware resource usage, it achieves approximately 1.5 dB BER gain and up to 66\% lower execution time.

eess.SY↗

arXiv2025

EcoSpa: Efficient Transformer Training with Coupled Sparsity

Transformers have become the backbone of modern AI, yet their high computational demands pose critical system challenges. While sparse training offers efficiency gains, existing methods fail to preserve critical structural relationships between weight matrices that interact multiplicatively in attention and feed-forward layers. This oversight leads to performance degradation at high sparsity levels. We introduce EcoSpa, an efficient structured sparse training method that jointly evaluates and sparsifies coupled weight matrix pairs, preserving their interaction patterns through aligned row/column removal. EcoSpa introduces a new granularity for calibrating structural component importance and performs coupled estimation and sparsification across both pre-training and fine-tuning scenarios. Evaluations demonstrate substantial improvements: EcoSpa enables efficient training of LLaMA-1B with 50\% memory reduction and 21\% faster training, achieves $2.2\times$ model compression on GPT-2-Medium with $2.4$ lower perplexity, and delivers $1.6\times$ inference speedup. The approach uses standard PyTorch operations, requiring no custom hardware or kernels, making efficient transformer training accessible on commodity hardware.

cs.LG↗

arXiv2025

OmniResponse: Online Multimodal Conversational Response Generation in Dyadic Interactions

In this paper, we introduce Online Multimodal Conversational Response Generation (OMCRG), a novel task designed to produce synchronized verbal and non-verbal listener feedback online, based on the speaker's multimodal inputs. OMCRG captures natural dyadic interactions and introduces new challenges in aligning generated audio with listeners' facial responses. To tackle these challenges, we incorporate text as an intermediate modality to connect audio and facial responses. We propose OmniResponse, a Multimodal Large Language Model (MLLM) that autoregressively generates accurate multimodal listener responses. OmniResponse leverages a pretrained LLM enhanced with two core components: Chrono-Text Markup, which precisely timestamps generated text tokens, and TempoVoice, a controllable online text-to-speech (TTS) module that outputs speech synchronized with facial responses. To advance OMCRG research, we offer ResponseNet, a dataset of 696 detailed dyadic interactions featuring synchronized split-screen videos, multichannel audio, transcripts, and annotated facial behaviors. Comprehensive evaluations on ResponseNet demonstrate that OmniResponse outperforms baseline models in terms of semantic speech content, audio-visual synchronization, and generation quality. Our dataset, code, and models are publicly available.

cs.CV↗

arXiv2025

Optimize Replica Server Placement in a Satellite Network

Satellite communication offers Internet connectivity to remote locations, such as villages, deserts, mountains, and at sea. However, transmitting content over satellite networks is significantly more expensive than traditional Internet. To address this issue, we propose placing content replica servers within satellite networks and optimizing replica placement for important performance metrics, such as latency, transmission, and storage cost. Our approach can support different types of satellite networks, including Low Earth Orbit (LEO), Medium Earth Orbit (MEO), Geostationary Orbit (GEO), and their combinations. An important challenge for supporting content replicas in such networks is that LEO and MEO satellites are constantly moving. We address this challenge by explicitly considering their moving trajectories and strategically optimizing not only client performance, but also the cost of transferring content from one satellite to another as needed. We demonstrate the effectiveness of our approach using both simulated traffic traces and a prototype system.

cs.NI↗

arXiv2025

First principles simulation of the collective rovibronic ground state in a cavity

Strong light-matter coupling in Fabry-Perot cavities can modify ground-state molecular reactivity, charge and energy transport, while modifications to single-molecule properties have not been observed experimentally. The mechanisms and reproducibility of such effects remain contested, with conflicting theoretical predictions driven by differences in Hamiltonian choice and quantum state representation. Here, we resolve these ambiguities with numerically exact quantum simulations of cavity-coupled molecular ensembles based on the ab initio light-matter Hamiltonian, treating electrons, nuclei, and cavity photons on equal footing. We investigate ensembles of the rotational-vibrational-electronic Shin-Metiu model using variational tree-tensor-network quantum dynamics, capturing rovibronic couplings and anharmonicity. Embedding ensembles in a cavity induces local modifications of rotational, nuclear, and --more weakly-- electronic observables in individual molecules. The extent of these modifications depends only on the per-molecule coupling strength up until each molecule reaches the ultrastrong coupling regime, which remains unattainable in practical Fabry-Perot setups. Increasing ensemble size toward the thermodynamic limit causes these local modifications to vanish regardless of dipole self-energy inclusion. Nonetheless, global ground-state observables, such as light-matter coupling energy contributions (affecting overall polarizability) and cavity field displacement fluctuations, depend crucially on proper treatment of intermolecular and light-matter correlations, whereas intramolecular observables remain largely insensitive. These insights are crucial for guiding future investigations using approximate quantum treatments, and for the interpretation of experimental results in polaritonic chemistry.

physics.chem-ph↗

arXiv2025

DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs

As large language models continue to scale, computational costs and resource consumption have emerged as significant challenges. While existing sparsification methods like pruning reduce computational overhead, they risk losing model knowledge through parameter removal. This paper proposes DSMoE (Dynamic Sparse Mixture-of-Experts), a novel approach that achieves sparsification by partitioning pre-trained FFN layers into computational blocks. We implement adaptive expert routing using sigmoid activation and straight-through estimators, enabling tokens to flexibly access different aspects of model knowledge based on input complexity. Additionally, we introduce a sparsity loss term to balance performance and computational efficiency. Extensive experiments on LLaMA models demonstrate that under equivalent computational constraints, DSMoE achieves superior performance compared to existing pruning and MoE approaches across language modeling and downstream tasks, particularly excelling in generation tasks. Analysis reveals that DSMoE learns distinctive layerwise activation patterns, providing new insights for future MoE architecture design.

cs.CL↗

arXiv2025

Low-Complexity Null-Space-Based Simultaneous Wireless Information and Power Transfer Scheme

Simultaneous wireless information and power transfer (SWIPT) has attracted sustained interest. We propose a null-space-based transmission scheme for multiuser SWIPT serving both energy users (EUs) and information users (IUs). Under a practical nonlinear energy-harvesting (EH) model and multiple waveform options, we revisit the role of dedicated energy beams (EBs). We show that, in general, dedicated EBs are unnecessary because information beams (IBs) with Gaussian signaling can simultaneously support wireless energy transfer (WET) and wireless information transfer (WIT), unless special energy-centric waveforms (e.g., deterministic sinusoidal waveforms) are employed and provide sufficient gains. Guided by these insights, we formulate an optimization problem for EB design to enable dedicated waveform transmission for WET, and we develop a low-complexity algorithm that reduces computation by ignoring the WET contribution of IBs during optimization. Numerical results corroborate that deterministic sinusoidal waveforms outperform Gaussian signaling when the received RF power lies in the EH high-efficiency region, making dedicated EBs beneficial. The proposed scheme achieves computational complexity reductions of 91.43\% and 98.54\% for the cases $M=8,,K^I=K^E=2$ and $M=16,,K^I=K^E=4$, respectively, with negligible performance loss, thereby validating the efficiency of the low-complexity algorithm.

eess.SP↗

arXiv2025

LabelGS: Label-Aware 3D Gaussian Splatting for 3D Scene Segmentation

3D Gaussian Splatting (3DGS) has emerged as a novel explicit representation for 3D scenes, offering both high-fidelity reconstruction and efficient rendering. However, 3DGS lacks 3D segmentation ability, which limits its applicability in tasks that require scene understanding. The identification and isolating of specific object components is crucial. To address this limitation, we propose Label-aware 3D Gaussian Splatting (LabelGS), a method that augments the Gaussian representation with object label.LabelGS introduces cross-view consistent semantic masks for 3D Gaussians and employs a novel Occlusion Analysis Model to avoid overfitting occlusion during optimization, Main Gaussian Labeling model to lift 2D semantic prior to 3D Gaussian and Gaussian Projection Filter to avoid Gaussian label conflict. Our approach achieves effective decoupling of Gaussian representations and refines the 3DGS optimization process through a random region sampling strategy, significantly improving efficiency. Extensive experiments demonstrate that LabelGS outperforms previous state-of-the-art methods, including Feature-3DGS, in the 3D scene segmentation task. Notably, LabelGS achieves a remarkable 22X speedup in training compared to Feature-3DGS, at a resolution of 1440X1080. Our code will be at https://github.com/garrisonz/LabelGS.

cs.CV↗

arXiv2025

Chinese Court Simulation with LLM-Based Agent System

Mock trial has long served as an important platform for legal professional training and education. It not only helps students learn about realistic trial procedures, but also provides practical value for case analysis and judgment prediction. Traditional mock trials are difficult to access by the public because they rely on professional tutors and human participants. Fortunately, the rise of large language models (LLMs) provides new opportunities for creating more accessible and scalable court simulations. While promising, existing research mainly focuses on agent construction while ignoring the systematic design and evaluation of court simulations, which are actually more important for the credibility and usage of court simulation in practice. To this end, we present the first court simulation framework -- SimCourt -- based on the real-world procedure structure of Chinese courts. Our framework replicates all 5 core stages of a Chinese trial and incorporates 5 courtroom roles, faithfully following the procedural definitions in China. To simulate trial participants with different roles, we propose and craft legal agents equipped with memory, planning, and reflection abilities. Experiment on legal judgment prediction show that our framework can generate simulated trials that better guide the system to predict the imprisonment, probation, and fine of each case. Further annotations by human experts show that agents' responses under our simulation framework even outperformed judges and lawyers from the real trials in many scenarios. These further demonstrate the potential of LLM-based court simulation.

cs.CY↗

arXiv2025

Algorithm Design and Prototype Validation for Reconfigurable Intelligent Sensing Surface: Forward-Only Transmission

Sensing-assisted communication schemes have recently garnered significant research attention. In this work, we design a dual-function reconfigurable intelligent surface (RIS), integrating both active and passive elements, referred to as the reconfigurable intelligent sensing surface (RISS), to enhance communication. By leveraging sensing results from the active elements, we propose communication enhancement and robust interference suppression schemes for both near-field and far-field models, implemented through the passive elements. These schemes remove the need for base station (BS) feedback for RISS control, simplifying the communication process by replacing traditional channel state information (CSI) feedback with real-time sensing from the active elements. The proposed schemes are theoretically analyzed and then validated using software-defined radio (SDR). Experimental results demonstrate the effectiveness of the sensing algorithms in real-world scenarios, such as direction of arrival (DOA) estimation and radio frequency (RF) identification recognition. Moreover, the RISS-assisted communication system shows strong performance in communication enhancement and interference suppression, particularly in near-field models.

eess.SP↗

arXiv2025

SynFER: Towards Boosting Facial Expression Recognition with Synthetic Data

Facial expression datasets remain limited in scale due to the subjectivity of annotations and the labor-intensive nature of data collection. This limitation poses a significant challenge for developing modern deep learning-based facial expression analysis models, particularly foundation models, that rely on large-scale data for optimal performance. To tackle the overarching and complex challenge, instead of introducing a new large-scale dataset, we introduce SynFER (Synthesis of Facial Expressions with Refined Control), a novel synthetic framework for synthesizing facial expression image data based on high-level textual descriptions as well as more fine-grained and precise control through facial action units. To ensure the quality and reliability of the synthetic data, we propose a semantic guidance technique to steer the generation process and a pseudo-label generator to help rectify the facial expression labels for the synthetic images. To demonstrate the generation fidelity and the effectiveness of the synthetic data from SynFER, we conduct extensive experiments on representation learning using both synthetic data and real-world data. Results validate the efficacy of our approach and the synthetic data. Notably, our approach achieves a 67.23% classification accuracy on AffectNet when training solely with synthetic data equivalent to the AffectNet training set size, which increases to 69.84% when scaling up to five times the original size. Code is available here.

cs.CV↗

arXiv2025

Why Generate When You Can Transform? Unleashing Generative Attention for Dynamic Recommendation

Sequential Recommendation (SR) focuses on personalizing user experiences by predicting future preferences based on historical interactions. Transformer models, with their attention mechanisms, have become the dominant architecture in SR tasks due to their ability to capture dependencies in user behavior sequences. However, traditional attention mechanisms, where attention weights are computed through query-key transformations, are inherently linear and deterministic. This fixed approach limits their ability to account for the dynamic and non-linear nature of user preferences, leading to challenges in capturing evolving interests and subtle behavioral patterns. Given that generative models excel at capturing non-linearity and probabilistic variability, we argue that generating attention distributions offers a more flexible and expressive alternative compared to traditional attention mechanisms. To support this claim, we present a theoretical proof demonstrating that generative attention mechanisms offer greater expressiveness and stochasticity than traditional deterministic approaches. Building upon this theoretical foundation, we introduce two generative attention models for SR, each grounded in the principles of Variational Autoencoders (VAE) and Diffusion Models (DMs), respectively. These models are designed specifically to generate adaptive attention distributions that better align with variable user preferences. Extensive experiments on real-world datasets show our models significantly outperform state-of-the-art in both accuracy and diversity.

cs.IR↗

arXiv2025

Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference

We present Seed Diffusion Preview, a large-scale language model based on discrete-state diffusion, offering remarkably fast inference speed. Thanks to non-sequential, parallel generation, discrete diffusion models provide a notable speedup to mitigate the inherent latency of token-by-token decoding, as demonstrated recently (e.g., Mercury Coder, Gemini Diffusion). Seed Diffusion Preview achieves an inference speed of 2,146 token/s over H20 GPUs while maintaining competitive performance across a sweep of standard code evaluation benchmarks, significantly faster than contemporary Mercury and Gemini Diffusion, establishing new state of the art on the speed-quality Pareto frontier for code models.

cs.CL↗

arXiv2025

Local Dispersive and Strichartz estimates for the Schrödinger equation associated to the Ornstein-Uhlenbeck operator

In this paper we study the linear and nonlinear Schrödinger equations associated with the Ornstein-Uhlenbeck (OU) operator endowed with the Gaussian measure. While classical Strichartz estimates are well-developed for the free Schrödinger operator on Euclidean spaces, extending them to non-translation-invariant operators like the OU operator presents significant challenges due to the lack of global dispersive decay. In this work, we overcome these difficulties by deriving localized $L^1 \to L^\infty$ dispersive estimates for the OU Schrödinger propagator using Mehler kernel techniques. We then establish a family of weighted Strichartz estimates in Gaussian $L^p$ spaces via interpolation and the abstract $TT^*$-method. As an application, we prove local well-posedness results for the nonlinear Schrödinger equation with power-type nonlinearity in both subcritical and critical regimes. Our framework reveals new dispersive phenomena in the context of the OU semigroup and provides the first comprehensive Strichartz theory in this setting.

math.FA↗