Searcharxiv⌕ Search

arXiv subjects

Hong Liu

Publications and source records attributed to Hong Liu.

At least 73 records · Page 4Linked to original sources
arXiv2026

Learning step-level dynamic soaring in shear flow

Dynamic soaring enables sustained flight by extracting energy from wind shear, yet it is commonly understood as a cycle-level maneuver that assumes stable flow conditions. In realistic unsteady environments, however, such assumptions are often violated, raising the question of whether explicit cycle-level planning is necessary. Here, we show that dynamic soaring can emerge from step-level, state-feedback control using only local sensing, without explicit trajectory planning. Using deep reinforcement learning as a tool, we obtain policies that achieve robust omnidirectional navigation across diverse shear-flow conditions. The learned behavior organizes into a structured control law that coordinates turning and vertical motion, giving rise to a two-phase strategy governed by a trade-off between energy extraction and directional progress. The resulting policy generalizes across varying conditions and reproduces key features observed in biological flight and optimal-control solutions. These findings identify a feedback-based control structure underlying dynamic soaring, demonstrating that efficient energy-harvesting flight can emerge from local interactions with the flow without explicit planning, and providing insights for biological flight and autonomous systems in complex, flow-coupled environments.

physics.flu-dyn↗
arXiv2026

Radiology Report Generation for Low-Quality X-Ray Images

Vision-Language Models (VLMs) have significantly advanced automated Radiology Report Generation (RRG). However, existing methods implicitly assume high-quality inputs, overlooking the noise and artifacts prevalent in real-world clinical environments. Consequently, current models exhibit severe performance degradation when processing suboptimal images. To bridge this gap, we propose a robust report generation framework explicitly designed for image quality variations. We first introduce an Automated Quality Assessment Agent (AQAA) to identify low-quality samples within the MIMIC-CXR dataset and establish the Low-quality Radiology Report Generation (LRRG) benchmark. To tackle degradation-induced shifts, we propose a novel Dual-loop Training Strategy leveraging bi-level optimization and gradient consistency. This approach ensures the model learns quality-agnostic diagnostic features by aligning gradient directions across varying quality regimes. Extensive experiments demonstrate that our approach effectively mitigates model performance degradation caused by image quality deterioration. The code and data will be released upon acceptance.

cs.CV↗
arXiv2026

Building Trust in the Skies: A Knowledge-Grounded LLM-based Framework for Aviation Safety

The integration of Large Language Models (LLMs) into aviation safety decision-making represents a significant technological advancement, yet their standalone application poses critical risks due to inherent limitations such as factual inaccuracies, hallucination, and lack of verifiability. These challenges undermine the reliability required for safety-critical environments where errors can have catastrophic consequences. To address these challenges, this paper proposes a novel, end-to-end framework that synergistically combines LLMs and Knowledge Graphs (KGs) to enhance the trustworthiness of safety analytics. The framework introduces a dual-phase pipeline: it first employs LLMs to automate the construction and dynamic updating of an Aviation Safety Knowledge Graph (ASKG) from multimodal sources. It then leverages this curated KG within a Retrieval-Augmented Generation (RAG) architecture to ground, validate, and explain LLM-generated responses. The implemented system demonstrates improved accuracy and traceability over LLM-only approaches, effectively supporting complex querying and mitigating hallucination. Results confirm the framework's capability to deliver context-aware, verifiable safety insights, addressing the stringent reliability requirements of the aviation industry. Future work will focus on enhancing relationship extraction and integrating hybrid retrieval mechanisms.

cs.SE↗
arXiv2026

Towards Multi-Object Nonprehensile Transportation via Shared Teleoperation: A Framework Based on Virtual Object Model Predictive Control

Multi-object nonprehensile transportation in teleoperation demands simultaneous trajectory tracking and tray orientation control. Existing methods often struggle with model dependency, uncertain parameters, and multi-object adaptability. We propose a shared teleoperation framework where humans and robots share positioning control, while the robot autonomously manages orientation to satisfy dynamic constraints. Key contributions include: 1) A theoretical dynamic constraint analysis utilizing a novel virtual object (VO)-based method to simplify constraints for trajectory planning. 2) An MPC-based trajectory smoothing algorithm that enforces real-time constraints and coordinates user tracking with orientation control. 3) Validations demonstrating stable manipulation of nine objects at accelerations up to 2.4 m/s2. Compared to the baseline, our approach reduces sliding distance by 72.45% and eliminates tip-overs (0% vs. 13.9%), proving robust adaptability in complex scenarios.

cs.RO↗
arXiv2026

Class-Imbalanced-Aware Adaptive Dataset Distillation for Scalable Pretrained Model on Credit Scoring

The advent of artificial intelligence has significantly enhanced credit scoring technologies. Despite the remarkable efficacy of advanced deep learning models, mainstream adoption continues to favor tree-structured models due to their robust predictive performance on tabular data. Although pretrained models have seen considerable development, their application within the financial realm predominantly revolves around question-answering tasks and the use of such models for tabular-structured credit scoring datasets remains largely unexplored. Tabular-oriented large models, such as TabPFN, has made the application of large models in credit scoring feasible, albeit can only processing with limited sample sizes. This paper provides a novel framework to combine tabular-tailored dataset distillation technique with the pretrained model, empowers the scalability for TabPFN. Furthermore, though class imbalance distribution is the common nature in financial datasets, its influence during dataset distillation has not been explored. We thus integrate the imbalance-aware techniques during dataset distillation, resulting in improved performance in financial datasets (e.g., a 2.5% enhancement in AUC). This study presents a novel framework for scaling up the application of large pretrained models on financial tabular datasets and offers a comparative analysis of the influence of class imbalance on the dataset distillation process. We believe this approach can broaden the applications and downstream tasks of large models in the financial domain.

cs.LG↗
arXiv2026

LongCat-Next: Lexicalizing Modalities as Discrete Tokens

The prevailing Next-Token Prediction (NTP) paradigm has driven the success of large language models through discrete autoregressive modeling. However, contemporary multimodal systems remain language-centric, often treating non-linguistic modalities as external attachments, leading to fragmented architectures and suboptimal integration. To transcend this limitation, we introduce Discrete Native Autoregressive (DiNA), a unified framework that represents multimodal information within a shared discrete space, enabling a consistent and principled autoregressive modeling across modalities. A key innovation is the Discrete Native Any-resolution Visual Transformer (dNaViT), which performs tokenization and de-tokenization at arbitrary resolutions, transforming continuous visual signals into hierarchical discrete tokens. Building on this foundation, we develop LongCat-Next, a native multimodal model that processes text, vision, and audio under a single autoregressive objective with minimal modality-specific design. As an industrial-strength foundation model, it excels at seeing, painting, and talking within a single framework, achieving strong performance across a wide range of multimodal benchmarks. In particular, LongCat-Next addresses the long-standing performance ceiling of discrete vision modeling on understanding tasks and provides a unified approach to effectively reconcile the conflict between understanding and generation. As an attempt toward native multimodality, we open-source the LongCat-Next and its tokenizers, hoping to foster further research and development in the community. GitHub: https://github.com/meituan-longcat/LongCat-Next

cs.CV↗
arXiv2026

Breaking the degeneracy among regular black holes with gravitational lensing

We examine parameter degeneracies in Culetu, Bardeen and Hayward regular black holes across lensing, shadow and quasinormal mode regimes. Our analysis reveals that while Einstein ring data yield extremely loose constraints, with the regularization parameter $q$ exceeding $\mathcal{O}(10^3)$, they fail to improve the parameter estimation when combined with strong lensing observables. In contrast, the Event Horizon Telescope observations provide remarkably tight limits: $0 \leq q < 0.0466 <0.0847$ for Culetu, $0 \leq q < 0.5115 <0.6682$ for Bardeen and $0 \leq q < 1.0258 <1.1881$ for Hayward, which shows that the strong field regime alone dominates the available parameter space. Despite these bounds, leading order geometric observables remain highly degenerate, which masks the microscopic details of non-singular cores. To break this ``macroscopic universality,'' we identify high order signatures, such as the Lyapunov exponent and subleading time delays, as sensitive probes of near horizon curvature. Crucially, we discover that the brightness hierarchy of accretion induced intensity profiles undergoes a fundamental inversion when transitioning from lensing dominated static flows to dynamics dominated infalling flows. These results demonstrate that high resolution temporal and intensity profiles are essential for distinguishing between regular black hole geometries.

gr-qc↗
arXiv2026

AngelSlim: A more accessible, comprehensive, and efficient toolkit for large model compression

This technical report introduces AngelSlim, a comprehensive and versatile toolkit for large model compression developed by the Tencent Hunyuan team. By consolidating cutting-edge algorithms, including quantization, speculative decoding, token pruning, and distillation. AngelSlim provides a unified pipeline that streamlines the transition from model compression to industrial-scale deployment. To facilitate efficient acceleration, we integrate state-of-the-art FP8 and INT8 Post-Training Quantization (PTQ) algorithms alongside pioneering research in ultra-low-bit regimes, featuring HY-1.8B-int2 as the first industrially viable 2-bit large model. Beyond quantization, we propose a training-aligned speculative decoding framework compatible with multimodal architectures and modern inference engines, achieving 1.8x to 2.0x throughput gains without compromising output correctness. Furthermore, we develop a training-free sparse attention framework that reduces Time-to-First-Token (TTFT) in long-context scenarios by decoupling sparse kernels from model architectures through a hybrid of static patterns and dynamic token selection. For multimodal models, AngelSlim incorporates specialized pruning strategies, namely IDPruner for optimizing vision tokens via Maximal Marginal Relevance and Samp for adaptive audio token merging and pruning. By integrating these compression strategies from low-level implementations, AngelSlim enables algorithm-focused research and tool-assisted deployment.

cs.LG↗
arXiv2026

Gamifying Compassion: Mitigating Dialect Prejudice Through An AI-Driven Serious Game

Dialect bias is pervasive yet often unconscious, normalized, or obscured by masking. Existing HCI interventions primarily audit disparities and propose reactive fixes. We present CompassioMate, a dialect-aware serious game that nurtures perspective-taking through AI-mediated play. Players listen to audio samples to identify regional dialects, engage in simulated social interactions involving dialect discrimination, and explore branching narratives that reveal how changes in wording or stance can influence the outcomes. In a three-week field study with 20 university students, participants reported feeling comfortable when observing region-tailored dialogues; several described experiencing perspective change. We contribute: 1) a formative study identifying goals for safe action consequence modelling, 2) the design and evaluation of a serious game integrating dialect audio, region-mapping play, bias; and 3) design implications highlighting listener-side training, transparent evaluation, and narratives maintaining psychological well-being.

cs.HC↗
arXiv2026

Structure Observation Driven Image-Text Contrastive Learning for Computed Tomography Report Generation

Computed Tomography Report Generation (CTRG) aims to automate the clinical radiology reporting process, thereby reducing the workload of report writing and facilitating patient care. While deep learning approaches have achieved remarkable advances in X-ray report generation, their effectiveness may be limited in CTRG due to larger data volumes of CT images and more intricate details required to describe them. This work introduces a novel two-stage (structure- and report-learning) framework tailored for CTRG featuring effective structure-wise image-text contrasting. In the first stage, a set of learnable structure-specific visual queries observe corresponding structures in a CT image. The resulting observation tokens are contrasted with structure-specific textual features extracted from the accompanying radiology report with a structure-wise image-text contrastive loss. In addition, text-text similarity-based soft pseudo targets are proposed to mitigate the impact of false negatives, i.e., semantically identical image structures and texts from non-paired images and reports. Thus, the model learns structure-level semantic correspondences between CT images and reports. Further, a dynamic, diversity-enhanced negative queue is proposed to guide the network in learning to discriminate various abnormalities. In the second stage, the visual structure queries are frozen and used to select the critical image patch embeddings depicting each anatomical structure, minimizing distractions from irrelevant areas while reducing memory consumption. Also, a text decoder is added and trained for report generation.Our extensive experiments on two public datasets demonstrate that our framework establishes new state-of-the-art performance for CTRG in clinical efficiency, and its components are effective.

cs.CV↗
arXiv2026

Federated Modality-specific Encoders and Partially Personalized Fusion Decoder for Multimodal Brain Tumor Segmentation

Most existing federated learning (FL) methods for medical image analysis only considered intramodal heterogeneity, limiting their applicability to multimodal imaging applications. In practice, some FL participants may possess only a subset of the complete imaging modalities, posing intermodal heterogeneity as a challenge to effectively training a global model on all participants' data. Meanwhile, each participant expects a personalized model tailored to its local data characteristics in FL. This work proposes a new FL framework with federated modality-specific encoders and partially personalized multimodal fusion decoders (FedMEPD) to address the two concurrent issues. Specifically, FedMEPD employs an exclusive encoder for each modality to account for the intermodal heterogeneity. While these encoders are fully federated, the decoders are partially personalized to meet individual needs -- using the discrepancy between global and local parameter updates to dynamically determine which decoder filters are personalized. Implementation-wise, a server with full-modal data employs a fusion decoder to fuse representations from all modality-specific encoders, thus bridging the modalities to optimize the encoders via backpropagation. Moreover, multiple anchors are extracted from the fused multimodal representations and distributed to the clients in addition to the model parameters. Conversely, the clients with incomplete modalities calibrate their missing-modal representations toward the global full-modal anchors via scaled dot-product cross-attention, making up for the information loss due to absent modalities. FedMEPD is validated on the BraTS 2018 and 2020 multimodal brain tumor segmentation benchmarks. Results show that it outperforms various up-to-date methods for multimodal and personalized FL, and its novel designs are effective.

cs.CV↗
arXiv2026

Efficient Degradation-agnostic Image Restoration via Channel-Wise Functional Decomposition and Manifold Regularization

Degradation-agnostic image restoration aims to handle diverse corruptions with one unified model, but faces fundamental challenges in balancing efficiency and performance across different degradation types. Existing approaches either sacrifice efficiency for versatility or fail to capture the distinct representational requirements of various degradations. We present MIRAGE, an efficient framework that addresses these challenges through two key innovations. First, we propose a channel-wise functional decomposition that systematically repurposes channel redundancy in attention mechanisms by assigning CNN, attention, and MLP branches to handle local textures, global context, and channel statistics, respectively. This principled decomposition enables degradation-agnostic learning while achieving superior efficiency-performance trade-offs. Second, we introduce manifold regularization that performs cross-layer contrastive alignment in Symmetric Positive Definite (SPD) space, which empirically improves feature consistency and generalization across degradation types. Extensive experiments demonstrate that MIRAGE achieves state-of-the-art performance with remarkable efficiency, outperforming existing methods in various all-in-one IR settings while offering a scalable and generalizable solution for challenging unseen IR scenarios.

cs.CV↗
arXiv2026

Infinitely many groups exhibiting intermediate growth in maximal sum-free sets

Given an Abelian groups $G$, denote $μ(G)$ the size of its largest sum-free subset and $f_{\max}(G)$ the number of maximal sum-free sets in $G$. Confirming a prediction by Liu and Sharifzadeh, we prove that all even-order $G\ne \mathbb{Z}_2^k$ have exponentially fewer maximal sum-free sets than $\mathbb{Z}_2^k$, i.e. $f_{\max}(G) \leq 2^{(1/2-c)μ(G)}$, where $c > 10^{-64}$. We construct an infinite family of Abelian groups $G$ with intermediate growth in the number of maximal sum-free sets, i.e., with $ 2^{(\frac{1}{2}+c)μ(G)}\leq f_{\max}(G) \leq 3^{(\frac{1}{3}-c)μ(G)} $, where $c=10^{-4}$. This disproves a conjecture of Liu and Sharifzadeh and also answers a question of Hassler and Treglown in the negative. Furthermore, we determine for every even-order group $G$, the number of maximal distinct sum-free sets (where a distinct sum is $a+b= c$ with distinct $a,b,c$): it is $ 2^{(1/2+o(1))μ(G)}$ with the only exception being $G=\mathbb{Z}_2^k \oplus \mathbb{Z}_3$, when this function is $3^{(1/3+o(1))μ(G)}$, refuting a conjecture of Hassler and Treglown. Our proofs rely on a container theorem due to Green and Ruzsa. Another key ingredient is a sharp upper bound we establish on the number of maximal independent sets in graphs with given matching number, which interpolates between the classical results of Moon and Moser, and Hujter and Tuza. A special case of our bound implies that every $n$-vertex graph with a perfect matching has at most $2^{n/2}$ maximal independent sets, resolving another conjecture of Hassler and Treglown.

math.CO↗
arXiv2026

Colour diversity in spanning structures under Dirac-type conditions

Finding spanning structures with many distinct colours in properly edge-coloured graphs is a central theme in extremal combinatorics. A classical result of Andersen shows that every proper edge-colouring of the complete graph $K_n$ contains a Hamilton cycle with $n - O(n^{1/2})$ distinct colours. In the bipartite setting, the analogous question for perfect matchings is closely related to permutations in Latin squares. In this paper, we investigate how a Dirac-type minimum degree condition forces colour diversity in spanning structures. For every constant $1/2 < c \le 1$, we prove the following. $\bullet$ Every properly edge-coloured graph $G$ on $n$ vertices with $δ(G)\ge cn$ contains a Hamilton cycle with at least $cn - O(n^{1/2})$ distinct colours. $\bullet$ Every subset of an $n\times n$ Latin square with at least $cn$ cells in each row and each column contains a permutation with at least $cn - O(n^{2/3})$ distinct symbols. Both bounds are best possible up to the error term.

math.CO↗
arXiv2026

Stability with minuscule structure for chromatic thresholds

The chromatic threshold $δ_χ(H)$ of a graph $H$ is the infimum of $d>0$ such that the chromatic number of every $n$-vertex $H$-free graph with minimum degree at least $d n$ is bounded by a constant depending only on $H$ and $d$. Allen, B{ö}ttcher, Griffiths, Kohayakawa, and Morris determined the chromatic threshold for every $H$; in particular, they showed that if $χ(H)=r\ge 3$, then $δ_χ(H) \in\{\frac{r-3}{r-2},~\frac{2 r-5}{2 r-3},~\frac{r-2}{r-1}\}$. While the chromatic thresholds have been completely determined, rather surprisingly the structural behaviors of extremal graphs near the threshold remain unexplored. In this paper, we establish the stability theorems for chromatic threshold problems. We prove that every $n$-vertex $H$-free graph $G$ with $δ(G)\ge (δ_χ(H)-o(1))n$ and $χ(G)=ω(1)$ must be structurally close to one of the extremal configurations. Furthermore, we give a stronger stability result when $H$ is a clique, showing that $G$ admits a partition into independent sets and a small subgraph on sublinear number of vertices. We show that this small subgraph has fractional chromatic number $2+o(1)$ and is homomorphic to a Kneser graph defined by subsets of a logarithmic size set; both these two bounds are best possible. This is the first stability result that captures the lower-order structural features of extremal graphs. We also study two variations of chromatic thresholds. Replacing chromatic number by its fractional counterpart, we determine the fractional chromatic thresholds for all graphs. Another variation is the bounded-VC chromatic thresholds, which was introduced by Liu, Shangguan, Skokan, and Xu very recently. Extending work of Łuczak and Thomass{é} on the triangle case, we determine the bounded-VC chromatic thresholds for all cliques.

math.CO↗
arXiv2026

Scaling Embeddings Outperforms Scaling Experts in Language Models

While Mixture-of-Experts (MoE) architectures have become the standard for sparsity scaling in large language models, they increasingly face diminishing returns and system-level bottlenecks. In this work, we explore embedding scaling as a potent, orthogonal dimension for scaling sparsity. Through a comprehensive analysis and experiments, we identify specific regimes where embedding scaling achieves a superior Pareto frontier compared to expert scaling. We systematically characterize the critical architectural factors governing this efficacy -- ranging from parameter budgeting to the interplay with model width and depth. Moreover, by integrating tailored system optimizations and speculative decoding, we effectively convert this sparsity into tangible inference speedups. Guided by these insights, we introduce LongCat-Flash-Lite, a 68.5B parameter model with ~3B activated trained from scratch. Despite allocating over 30B parameters to embeddings, LongCat-Flash-Lite not only surpasses parameter-equivalent MoE baselines but also exhibits exceptional competitiveness against existing models of comparable scale, particularly in agentic and coding domains.

cs.CL↗
arXiv2026

ERTACache: Error Rectification and Timesteps Adjustment for Efficient Diffusion

Diffusion models suffer from substantial computational overhead due to their inherently iterative inference process. While feature caching offers a promising acceleration strategy by reusing intermediate outputs across timesteps, naive reuse often incurs noticeable quality degradation. In this work, we formally analyze the cumulative error introduced by caching and decompose it into two principal components: feature shift error, caused by inaccuracies in cached outputs, and step amplification error, which arises from error propagation under fixed timestep schedules. To address these issues, we propose ERTACache, a principled caching framework that jointly rectifies both error types. Our method employs an offline residual profiling stage to identify reusable steps, dynamically adjusts integration intervals via a trajectory-aware correction coefficient, and analytically approximates cache-induced errors through a closed-form residual linearization model. Together, these components enable accurate and efficient sampling under aggressive cache reuse. Extensive experiments across standard image and video generation benchmarks show that ERTACache achieves up to 2x inference speedup while consistently preserving or even improving visual quality. Notably, on the state-of-the-art Wan2.1 video diffusion model, ERTACache delivers 2x acceleration with minimal VBench degradation, effectively maintaining baseline fidelity while significantly improving efficiency. The code is available at https://github.com/bytedance/ERTACache.

cs.CV↗
arXiv2026

IDPruner: Harmonizing Importance and Diversity in Visual Token Pruning for MLLMs

Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities, yet they encounter significant computational bottlenecks due to the massive volume of visual tokens. Consequently, visual token pruning, which substantially reduces the token count, has emerged as a critical technique for accelerating MLLM inference. Existing approaches focus on token importance, diversity, or an intuitive combination of both, without a principled framework for their optimal integration. To address this issue, we first conduct a systematic analysis to characterize the trade-off between token importance and semantic diversity. Guided by this analysis, we propose the \textbf{I}mportance and \textbf{D}iversity Pruner (\textbf{IDPruner}), which leverages the Maximal Marginal Relevance (MMR) algorithm to achieve a Pareto-optimal balance between these two objectives. Crucially, our method operates without requiring attention maps, ensuring full compatibility with FlashAttention and efficient deployment via one-shot pruning. We conduct extensive experiments across various model architectures and multimodal benchmarks, demonstrating that IDPruner achieves state-of-the-art performance and superior generalization across diverse architectures and tasks. Notably, on Qwen2.5-VL-7B-Instruct, IDPruner retains 95.18\% of baseline performance when pruning 75\% of the tokens, and still maintains 86.40\% even under an extreme 90\% pruning ratio. Our code is available at https://github.com/Tencent/AngelSlim.

cs.CV↗