Searcharxiv⌕ Search

arXiv subjects

Xiang Cheng

Publications and source records attributed to Xiang Cheng.

At least 37 records · Page 2Linked to original sources
arXiv2026

Learning When to Denoise: Optimizing Asynchronous Schedules for Latent Diffusion

Multi-representation diffusion models can improve visual synthesis by denoising complementary views of an image, but their performance depends critically on the asynchronous schedule that determines when each representation is denoised. We propose to learn this schedule. Our method formulates asynchronous flow matching over multiple representation spaces and uses a schedule-corrected objective that keeps each representation's local noising-time weights fixed as the schedule changes. We instantiate the schedule with a flexible parametric class that is convex and monotone by construction, and learn it using a fast joint probe with less than 1% additional training compute. On ImageNet 256x256, the learned schedule substantially improves both convergence speed and final quality under a matched 675M-parameter XL backbone. With AutoGuidance, our 200-epoch model reaches FID 1.05, matching the 800-epoch SFD-XL baseline with 4x less training. Training to 600 epochs further improves to FID 1.02, outperforming the 1B-parameter SFD-XXL result of FID 1.04 while using a smaller model. In the unguided setting, our 200-epoch model reaches FID 2.37, already below the best 800-epoch SFD-XL result (2.54) at 4x less training, and improves to FID 2.14 at 600 epochs. Code is available at https://github.com/bsq532087/LWD

cs.CV↗
arXiv2026

Transformers Can Learn Posterior Predictive Distributions In-Context

Prior-data fitted networks (PFNs) have recently emerged as a powerful approach for Bayesian prediction tasks, approximating the posterior predictive distribution (PPD) through in-context learning. Despite their strong empirical performance and ability to go beyond point predictions, theoretical understandings of the algorithmic capability of transformers to learn distributions in context are still lacking. Focusing on Gaussian process regression problems, we show by construction that transformers can implement a gradient descent algorithm targeting the posterior predictive mean and variance, followed by nonlinear mappings that yield binned probabilities of PPD. We study the error bounds of the approximated PPD in terms of attention depth and bin resolution. Based on these results, we further demonstrate the key role of normalization and the choice of attention depth in enabling the extrapolation abilities of transformers beyond the pretraining sample size range. We conduct simulations that corroborate our findings, providing insight into the expressivity of PFNs targeting PPDs and how architectural choices may influence generalization capabilities.

stat.ML↗
arXiv2026

A Multi-Modal Intelligent U2V Channel Model for 6G Sensing-Communication Integration

This paper proposes a novel UAV-to-Vehicle (U2V) channel model for sixth-generation (6G) intelligent sensing-communication integration, based on three-dimensional (3D) scatterer prediction. To explore the mapping relationship between physical environment and electromagnetic space, a new high-fidelity mixed sensing-communication integration U2V simulation dataset under wide-lane scenarios with different vehicular traffic densities (VTDs) and UAV heights is constructed. Based on the constructed dataset, a novel 3D Scatterer Prediction and Distribution Estimation (3D-SPADE) algorithm is proposed, which leverages LiDAR point clouds to accurately predict the spatial distribution of scatterers. Furthermore, the clustering of scatterers and the subsequent classification into dynamic and static types are meticulously designed for highly dynamic U2V scenarios, while reducing computational complexity and improving modeling accuracy. As LiDAR point clouds vary over time, dynamic and static clusters evolve via 3D-SPADE, enabling precise modeling of channel non-stationarity and consistency. Simulation results demonstrate that, in the wide-lane scenario with varying VTDs and UAV heights, the proposed 3D-SPADE consistently achieves high scatterer occupancy detection performance within the voxel grid. In particular, under favorable configurations, recall reaches 93.26%, and precision reaches 95.74%, highlighting the reliability of 3D-SPADE. Key channel statistical characteristics are simulated and analyzed. These characteristics from the simulation experiments are highly consistent with ray-tracing results and exhibit better agreement than with the standardized model and inconsistent model, validating the necessity of exploring the mapping relationship and the effectiveness of the proposed model.

eess.SP↗
arXiv2026

Sensing-Assisted LoS/NLoS Identification in Dynamic UAV Positioning Systems

In this paper, a sensing-assisted non-line-of-sight (NLoS) identification method for dynamic uncrewed aerial vehicle (UAV) positioning is proposed for the first time. For urban UAV-to-ground scenarios, a new multi-modal sensing-communication integrated dataset is constructed to support line-of-sight (LoS)/NLoS identification, covering two typical urban scenarios and a wide range of flight altitudes. Based on the constructed dataset, a novel dual-input feature fusion network is proposed, which addresses the challenge of heterogeneous representations between RGB images and channel impulse response (CIR) data to enable the joint extraction and fusion of sensing and communication features for LoS/NLoS identification. Simulation results show that the identification accuracy can reach up to 97.69%, while achieving an improvement of at least 3.59% compared to traditional CIR-only and RGB-only methods. Moreover, strong few-shot generalization is observed, as the proposed method stabilizes and approaches full-sample performance with fewer than 200 target samples and exceeds traditional CIR-only and RGB-only methods with fewer than 100 target samples in all cross-scenario and cross-altitude experiments. Even under Gaussian noise with a variance of 0.35 applied to RGB images, the accuracy degradation remains approximately 0.5%. By utilizing the proposed LoS/NLoS identification method, the error of trilateration positioning can be reduced by approximately 70% in a crossroad scenario, verifying the utility of the proposed method.

eess.SP↗
arXiv2026

To Err Is Human; To Annotate, SILICON? Toward Robust Reproducibility in LLM Annotation

Unstructured text data annotation is foundational to management research. LLMs offer a cost-effective and scalable alternative to human annotation, but they introduce a novel challenge: the annotator itself can be retired. Proprietary models undergo regular deprecation cycles, threatening long-term reproducibility. Hence, the ability to reproduce annotation results when the original model becomes unavailable, i.e., robust reproducibility, is a central methodological challenge for LLM-based annotation. Achieving robust reproducibility requires first controlling measurement error. We develop an analytical framework that decomposes measurement error into four sources: guideline-induced error from inconsistent annotation criteria, baseline-induced error from unreliable human references, prompt-induced error from suboptimal meta-instruction, and model-induced error from architectural differences across LLMs. We develop the SILICON workflow that instantiates the analytical framework, prescribing targeted interventions at each error source. Empirical validation across nine management research tasks confirms that these interventions reduce measurement error, and simulations show that the resulting error reduction yields more accurate downstream statistical estimates. With measurement error controlled, we address two further aspects of robust reproducibility. First, we propose a regression-based methodology to establish backup open-weight models, which are permanently accessible. Every tested task has at least one open-weight model with no statistically detectable performance difference. Second, we quantify the upper bound of annotation quality attainable from the current set of available models by proposing a routing procedure that selectively sends low-confidence items to auxiliary models, revealing when model aggregation improves performance and when that may adversely affect labeling quality.

cs.CL↗
arXiv2026

Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information

On-policy self-distillation, where a student is pulled toward a copy of itself conditioned on privileged context (e.g., a verified solution or feedback), offers a promising direction for advancing reasoning capability without a stronger external teacher. Yet in math reasoning the gains are inconsistent, even when the same approach succeeds elsewhere. A pointwise mutual information analysis traces the failure to the privileged context itself: it inflates the teacher's confidence on tokens already implied by the solution (structural connectives, verifiable claims) and deflates it on deliberation tokens ("Wait", "Let", "Maybe") that drive multi-step search. We propose Anti-Self-Distillation (AntiSD), which ascends a divergence between student and teacher rather than descending it: this reverses the per-token sign and yields a naturally bounded advantage in one step. An entropy-triggered gate disables the term once the teacher entropy collapses, completing a drop-in replacement for default self-distillation. Across five models from 4B to 30B parameters on math reasoning benchmarks, AntiSD reaches the GRPO baseline's accuracy in 2 to 10x fewer training steps and improves final accuracy by up to 11.5 points. AntiSD opens a path to scalable self-improvement, where a language model bootstraps its own reasoning through its training signal.

cs.LG↗
arXiv2026

From Generic Correlation to Input-Specific Credit in On-Policy Self Distillation

On-policy self-distillation has emerged as a promising paradigm for post-training language models, in which the model conditions on environment feedback to serve as its own teacher, providing dense token-level rewards without external teacher models or step-level annotations. Despite its empirical success, what this reward actually measures and what kind of credit it assigns remain unclear. Under a posterior-compatibility interpretation of feedback conditioning, standard in the implicit-reward literature, we show that the self-distillation token reward is a Bayesian filtering increment whose trajectory sum is exactly the pointwise mutual information between the response and the feedback given the input. This pMI can be raised by input-specific reasoning or by input-generic shortcuts, so we further decompose the teacher log-probability along the input axis. Based on this analysis, we propose CREDIT (Contrastive REward from DIsTillation), which isolates the input-specific component with a batch-contrastive baseline. At the sequence level, CREDIT is a teacher-side surrogate for a contrastive pMI objective that also penalizes responses remaining likely under unrelated inputs. Across coding, scientific reasoning, and tool-use benchmarks on two model families, CREDIT delivers the strongest aggregate performance at negligible additional compute.

cs.LG↗
arXiv2026

One for All: A Non-Linear Transformer can Enable Cross-Domain Generalization for In-Context Reinforcement Learning

A central challenge in reinforcement learning (RL) is to learn models that generalize beyond the tasks on which they are trained, a goal traditionally pursued through multi-task and meta RL. Recently, transformer architectures have emerged as a promising approach, enabling adaptation to new tasks via in-context learning without explicit parameter updates. From a functional perspective, a transformer can be viewed as a functional operator that maps a context to a task-specific function. It is thus fundamental to understand and design this operator to support stronger generalization in RL. In this work, we address this resulting question of generalization from a kernel-based perspective by establishing a connection between non-linear transformers and kernel-based temporal difference learning. By interpreting the transformer as performing regression in a Reproducing Kernel Hilbert Space (RKHS), we show that value functions from different domains can be represented using a shared set of weights, provided they lie within the same RKHS. Experiments on multiple MetaWorld domains support this interpretation, demonstrating convergence of the temporal-difference objective.

cs.LG↗
arXiv2026

VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training

Off-policy updates are inevitable in reinforcement learning (RL) for large language models (LLMs) due to rollout staleness from asynchronous training and mismatches between training and inference engines. Naive importance sampling gives an unbiased correction but suffers from high variance, which is amplified by unbounded ratios and autoregressive generation. Prior remedies either rely on scenario-specific engineering, or trade bias for variance via token-level clipping or sequence-level normalization, yet these approaches remain largely heuristic. We propose Variational sEquence-level Soft Policy Optimization (VESPO). By explicitly incorporating variance reduction into a variational formulation, we derive a principled closed-form reshaping kernel that operates directly on sequence-level importance weights, avoids token-level approximation and length normalization, and admits an explicit variance bound for the deployed kernel. Experiments on math reasoning and code generation show that VESPO maintains stable training under severe off-policy conditions (staleness up to 64x) and delivers consistent gains across both dense and Mixture-of-Experts (MoE) models, outperforming recent reshaping baselines under matched setup. Code is available at https://github.com/FloyedShen/VESPO.

cs.LG↗
arXiv2026

SPAT: A Semantic Port-Aware Adaptive-Rate Transmission Protocol for Semantic Communication

With the evolution of 6G, semantic communication has emerged as a promising paradigm by prioritizing the delivery of task-relevant meaning over strict bit-level correctness. However, existing transport mechanisms still rely on explicit port headers and bit-level validation, making them vulnerable to header corruption and the resulting packet loss. To address this issue, this paper proposes a Semantic Port-Aware Adaptive-Rate Transmission Protocol (SPAT) for semantic communication. The proposed framework jointly embeds source and destination port information into semantic representations, thereby reducing dependence on explicit port headers while enabling robust port-aware transmission. Furthermore, a differentiated semantic processing mechanism is developed for uplink and downlink scenarios, where port identification is introduced for uplink service recognition and destination-aware conditional gating is designed for downlink selective decoding. In addition, an adaptive-rate controller is incorporated to dynamically adjust the number of transmitted semantic channels according to channel conditions and feature importance, thereby improving both robustness and transmission efficiency. Experimental results on the AFHQ and ImageNet-10 datasets, together with real-world experimental measurements, demonstrate that SPAT consistently outperforms TCP, UDP, and SITP in reconstruction quality across different SNRs while maintaining low-latency transmission.

eess.SP↗
arXiv2026

Beyond Itinerary Planning-A Real-World Benchmark for Multi-Turn and Tool-Using Travel Tasks

Travel planning is a natural real-world task to test large language models' (LLMs) planning and tool-use abilities. Although prior work has studied LLM performance on travel planning, existing settings still differ from real-world needs, mainly due to limited domain coverage, insufficient modeling of users' implicit preferences in multi-turn conversations, and a lack of evaluation of agents' capability boundaries. To mitigate these gaps, we propose $\textbf{TravelBench}$, a benchmark for $\textit{truly real-world}$ travel planning. We collect user queries, user preferences, and tools from real scenarios, and construct three subtasks -- $\textit{Single-Turn}$, $\textit{Multi-Turn}$, and $\textit{Unsolvable}$ -- to evaluate agents' three core capabilities in real settings: (1) solving problems independently, (2) interacting with users to elicit implicit preferences, and (3) recognizing the capability boundaries. To enable stable tool invocation and reproducible evaluation, we cache real tool-call results and build a sandbox environment which integrates ten travel-related tools, enabling agents to combine these tools to solve most practical travel planning problems. We evaluate multiple LLMs on TravelBench and find that even advanced models exhibit imbalanced performance across different capabilities. Our further systematic verification demonstrates the stability of the proposed benchmark. TravelBench provides a practical and reproducible benchmark to advance research on LLM agents for real-world travel planning.

cs.AI↗
arXiv2026

ChemVLR: Prioritizing Reasoning in Perception for Chemical Vision-Language Understanding

While Vision-Language Models (VLMs) have demonstrated significant potential in chemical visual understanding, current models are predominantly optimized for direct visual question-answering tasks. This paradigm often results in "black-box" systems that fail to utilize the inherent capability of Large Language Models (LLMs) to infer underlying reaction mechanisms. In this work, we introduce ChemVLR, a chemical VLM designed to prioritize reasoning within the perception process. Unlike conventional chemical VLMs, ChemVLR analyzes visual inputs in a fine-grained manner by explicitly identifying granular chemical descriptors, such as functional groups, prior to generating answers. This approach ensures the production of explicit and interpretable reasoning paths for complex visual chemical problems. To facilitate this methodology, we implement a cross-modality reverse-engineering strategy, combined with a rigorous filtering pipeline, to curate a large-scale reasoning-and-captioning dataset comprising 760k high-quality samples across molecular and reaction tasks. Furthermore, we adopt a three-stage training framework that systemically builds model perception and reasoning capacity. Experiments demonstrate that ChemVLR achieves state-of-the-art (SOTA) performance, surpassing both leading proprietary models and domain-specific open-source baselines. We also provide comprehensive ablation studies to validate our training strategy and data generation designs. Code and model weights will be available at https://github.com/xxlllz/ChemVLR.

cs.CL↗
arXiv2026

Seeing new depths: Three-dimensional flow of a free-swimming alga

A swimming microorganism stirs the surrounding fluid, creating a flow field that governs not only its locomotion and nutrient uptake, but also its interactions with other microorganisms and the environment. Despite its fundamental importance, capturing this flow field and unraveling its biological implications remains a challenge. Here, we report the first direct, time-resolved measurements of the three-dimensional (3D) flow field generated by a single, free-swimming microalga, Chlamydomonas reinhardtii, a model organism for microbial locomotion and flagellar dynamics. Supported by hydrodynamic modeling and simulations, our measurements resolve how established two-dimensional (2D) flow features such as in-plane vortices and the stagnation point emerge from and shape the full algal flow in 3D. Moreover, we reveal unexpected low-Reynolds-number flow phenomena including micron-sized vortex rings and periodically recurring translating vortices and uncover topological changes in the underlying flow structure associated with the puller-to-pusher transition of an alga. Biologically, access to the 3D flow field enables rigorous quantification of the alga's energy expenditure, as well as its swimming and feeding efficiency, improving the precision of these physiological metrics. Taken together, our study demonstrates rich vortex dynamics in inertialess flows and shows their influence on microbial motility. The work also introduces a new experimental method for mapping the fluid environment sculpted by beating flagella.

physics.flu-dyn↗
arXiv2026

Multi-Modal Intelligent Channel Modeling: From Fine-tuned LLMs to Pre-trained Foundation Models

To meet the evolving demands of sixth-generation (6G) wireless channel modeling, such as precise prediction capability, extension capabilities, and system participation capability, multi-modal intelligent channel modeling (MMICM) has been proposed based on Synesthesia of Machines (SoM) which explores the mapping relationship between multi-modal sensing in physical environment and channel characteristics in electromagnetic space. Furthermore, for integrating heterogeneous sensing, reasoning across scales, and generalizing to complex air-space-ground-sea communication environments, two new paradigms of MMICM are explored, including fine-tuned large language models (LLMs) for Channel Modeling (LLM4CM) and Wireless Channel Foundation Model (WiCo). LLM4CM leverages pre-trained LLMs on channel representations for cross-modal alignment and lightweight adaptation, enabling flexible channel modeling for 6G multi-band and multi-scenario communication systems. WiCo, which pre-trained on physically valid channel realizations and their associated environmental and modal observations, embeds electromagnetic equations for physical interpretability and uses parameterized adapters for scalability. This article details the architectures and features of LLM4CM and WiCo, laying a foundation for artificial intelligence (AI)-native 6G wireless communication systems. Then, we conducts a comparative analysis of the two emerging paradigms, focusing on their distinct characteristics, relative advantages, inherent limitations, and performance attributes. Finally, we discuss the future research directions.

eess.SP↗
arXiv2026

Improving Hydrodynamic Modeling of Free-Swimming Algae Using a Modified Three-Sphere Approach

The beating flagella of the green alga Chlamydomonas reinhardtii play a prominent role in cellular mechanics, enabling cells to both displace and sense surrounding fluid. Specifically, flagellum-induced fluid transport enables microalgae to swim through fluid media and interact with other microorganisms. Minimal models, such as the three-sphere model with one sphere representing the cell body and two orbiting spheres mimicking the flagella, have been widely adopted to study various aspects of algal motility, including the synchronization of flagellar beating, run-and-tumble swimming, responses to shear flow, cell-body rolling, and helical navigation. However, detailed investigation of the algal flow fields generated by this minimal model remains limited. In this study, we systematically examine the time-averaged and time-resolved fluid flows generated by the three-sphere algae model and compare the numerical predictions with experimental data. Our findings reveal that the standard three-sphere model fails to produce key flow characteristics observed experimentally. To address this discrepancy, we explore a modified three-sphere model with refined flagellar beating dynamics and identify that differential drag acting on the flagellar spheres is the dominant factor influencing the fidelity of the modeled flow fields. These results advance the fundamental understanding of the flagellum-fluid interactions and algal flow and enhance our ability to accurately simulate microswimmer dynamics.

physics.flu-dyn↗
arXiv2026

Bootstrapping Exploration with Group-Level Natural Language Feedback in Reinforcement Learning

Large language models (LLMs) typically receive diverse natural language (NL) feedback through interaction with the environment. However, current reinforcement learning (RL) algorithms rely solely on scalar rewards, leaving the rich information in NL feedback underutilized and leading to inefficient exploration. In this work, we propose GOLF, an RL framework that explicitly exploits group-level language feedback to guide targeted exploration through actionable refinements. GOLF aggregates two complementary feedback sources: (i) external critiques that pinpoint errors or propose targeted fixes, and (ii) intra-group attempts that supply alternative partial ideas and diverse failure patterns. These group-level feedbacks are aggregated to produce high-quality refinements, which are adaptively injected into training as off-policy scaffolds to provide targeted guidance in sparse-reward regions. Meanwhile, GOLF jointly optimizes generation and refinement within a unified RL loop, creating a virtuous cycle that continuously improves both capabilities. Experiments on both verifiable and non-verifiable benchmarks show that GOLF achieves superior performance and exploration efficiency, achieving 2.2$\times$ improvements in sample efficiency compared to RL methods trained solely on scalar rewards. Code is available at https://github.com/LuckyyySTA/GOLF.

cs.CL↗
arXiv2026

From Specialist to Large Models: A Paradigm Evolution Towards Semantic-Aware MIMO

The sixth generation (6G) network is expected to deploy larger multiple-input multiple-output (MIMO) arrays to support massive connectivity, which will increase overhead and latency at the physical layer. Meanwhile, emerging 6G demands such as immersive communications and environmental sensing pose challenges to traditional signal processing. To address these issues, we propose the ``semantic-aware MIMO'' paradigm, which leverages specialist models and large models to perceive, utilize, and fuse the inherent semantics of channels and sources for improved performance. Moreover, for representative MIMO physical-layer tasks, e.g., random access activity detection, channel feedback, and precoding, we design specialist models that exploit channel and source semantics for better performance. Additionally, in view of the more diversified functions of 6G MIMO, we further explore large models as a scalable solution for multi-task semantic-aware MIMO and review recent advances along with their advantages and limitations. Finally, we discuss the challenges, insights, and prospects of the evolution of specialist models and large models empowered semantic-aware MIMO paradigms.

cs.IT↗
arXiv2026

Variational Trajectory Optimization of Anisotropic Diffusion Schedules

We introduce a variational framework for diffusion models with anisotropic noise schedules parameterized by a matrix-valued path $M_t(θ)$ that allocates noise across subspaces. Central to our framework is a trajectory-level objective that jointly trains the score network and learns $M_t(θ)$, which encompasses general parameterization classes of matrix-valued noise schedules. We further derive an estimator for the derivative with respect to $θ$ of the score that enables efficient optimization of the $M_t(θ)$ schedule. For inference, we develop an efficiently-implementable reverse-ODE solver that is an anisotropic generalization of the second-order Heun discretization algorithm. Across CIFAR-10, AFHQv2, FFHQ, and ImageNet-64, our method consistently improves upon the baseline EDM model in all NFE regimes. Code is available at https://github.com/lizeyu090312/anisotropic-diffusion-paper.

cs.LG↗