arXiv subjects

Linghe Kong

Publications and source records attributed to Linghe Kong.

At least 19 recordsLinked to original sources

arXiv2026

SWRouter: Similarity-Contractive Window Routing for Multi-Turn Large Language Model Conversations

Large language models exhibit complementary strengths, motivating routing methods that dispatch each query to the most suitable model. Although existing routers are effective in single-turn settings, they do not directly transfer to multi-turn dialogue, where routing performance critically depends on how historical context is segmented, retained, and incorporated into the current prompt. This introduces two fundamental challenges: preventing information loss and information confusion during context construction, and evaluating routing quality without conflating model selection with prompt construction quality. In this paper, we propose SWRouter, a Similarity-Contractive Window Router for multi-turn large language model routing. SWRouter combines a similarity-based context segmentation mechanism for prompt construction with a dual-metric evaluation framework that decouples construction accuracy from router performance. Experiments on multi-turn dialogue benchmarks demonstrate that SWRouter consistently surpasses strong baselines, achieving a 16.26% improvement in evaluation accuracy over the best individual large language model and an additional 8.22% gain over the Conv-ID Context baseline. Our results highlight that multi-turn large language model routing requires a joint design of context construction and evaluation, rather than a direct extension of single-turn routing methods.

cs.CL↗

arXiv2026

Exploring the Initial Performance of NB-IoT NTN over GEO: Measurement and Analysis

With the standardization of Non-Terrestrial Networks (NTN) to provide direct satellite connectivity to massive, low-power Internet of Things (IoT) devices, 3GPP IoT-NTN bridges the worlds of cellular and satellite communications. While holding great potential for global connectivity with IoT devices, there exist several concerns about the system performance of IoT-NTN over Geostationary Earth Orbit (GEO), which covers multiple dimensions such as end-to-end delay and energy consumption considering the ultra-long transmission distance from ground IoT terminals to GEO satellite. To answer these concerns, we have conducted the first comprehensive and in depth measurement for NB-IoT NTN over GEO. Based on real NB-IoT NTN testbeds including both Skylo and Tiantong, measurements covering more than six months confirm that the available implementation of NB-IoT NTN remains in the initial stage. Amplified by ultra-long Round-Trip Time (RTT) between ground IoT terminals and GEO satellite, there exists plenty of time and energy consumption during the access process. Interestingly, it also reveals that as an energy-saving mechanism, Power Saving Mode (PSM) fundamentally reshapes NB-IoT NTN traffic into a bursty and access-driven communication pattern and thus has a considerable influence on the end-to-end delay and energy consumption. Finally, we propose corresponding optimization schemes to reduce the delay and energy, which lays a good foundation for the implementation of NB-IoT NTN in the near future.

cs.NI↗

arXiv2026

ET-Prune: Evidence-Aware Dynamic Budgeting for Visual Token Pruning in Text-Rich MLLMs

Visual token pruning reduces the inference cost of multimodal large language models, but a fixed token ratio is poorly matched to text-rich inputs. In OCR-centric tasks, decisive evidence can be a small number, label, or field whose relevance is specified by the question; indiscriminate pruning can erase that evidence while retaining visually salient but irrelevant regions. We present ET-Prune, a training-free framework that casts pruning as evidence allocation. It derives question-conditioned evidence from a decoder-side partial query-key block, safeguards text-like spatial regions, and converts evidence uncertainty and density into a sample-specific token floor. Three progressive middle-layer events then move the sequence toward this budget, retaining more tokens for diffuse or text-dense evidence and pruning concentrated evidence more aggressively. At the observed point estimates from one deterministic pass per configuration, ET-Prune leads or ties among pruned methods in all six backbone-benchmark comparisons at roughly half tokens. On OCRBench-v2, it leads the strongest pruned baselines by 1.80 and 0.68 percentage points on Qwen3-VL-8B and InternVL3.5-8B, respectively, while retaining about half of the visual tokens; on MMBench v1.1, it reaches 0.8467 circular exact-matching accuracy versus 0.8437 for Vanilla at 54.45% average visual-token retention. These results show a favorable observed quality-cost trade-off for evidence-aware dynamic budgeting in text-rich multimodal inference.

cs.CV↗

arXiv2026

FreqForcing: Autoregressive Long Video Generation via Spectral Self-Anchoring

Autoregressive video diffusion models enable real-time streaming video generation. However, errors introduced during self-rollout accumulate over long horizons, manifesting as color drift, motion stagnation, and eventual visual collapse. In this paper, we characterize this phenomenon from a frequency-domain perspective: error accumulation appears as a pronounced energy drift in the low-frequency bands. We further investigate the effectiveness of attention sink in the frequency domain, and find that it improves the video quality by alleviating the spectral energy drift to some extent, but cannot fully resolve it. Motivated by the above analysis, we propose FreqForcing, a training-free framework that addresses error accumulation in long-video generation via Spectral Self-Anchoring (SSA). The proposed SSA leverages the low-frequency components of anchor attention to maintain long-horizon visual stability, while preserving dynamic motion through the high-frequency components of local attention. Our FreqForcing extends Self-Forcing pretrained on 5s clips to two-minute generation, achieving 24x extrapolation. Extensive experiments show that FreqForcing outperforms existing training-free methods quantitatively and qualitatively while remaining competitive with representative training-based approaches.

cs.CV↗

arXiv2026

GeoFlow: Geo-Aware Modeling of Inter-Area Relationships in Origin-Destination Flow Prediction and Generation

Origin-destination (OD) flow modeling underpins urban planning and mobility analysis, but prevailing graph-based methods often neglect salient geographic attributes, limiting their ability to model long-range and multi-area dependencies. In this paper, we introduce GeoFlow, a novel framework that (i) augments area representations with geospatial attributes, including relative positions, k-hop and geodesic distances, (ii) employs a specialized geometric-intrinsic fusion encoder design that combines graph attention for intrinsic area signals with coordinate-aware encoders for global structure, and (iii) adopts an axial-global attention decoder to capture OD-specific competitive dependencies. For OD flow generation, GeoFlow is paired with flow matching models to produce more authentic and diverse mobility samples. Empirically, GeoFlow achieves superior performance in predictive accuracy, while substantially improving generative fidelity and diversity. Ablation and analytical studies confirm the contribution of each component. Code is available at https://github.com/ZheruiHuang/GeoFlow.

cs.LG↗

arXiv2026

Black-Box Continual Learning for Vision-Language Models

The rapid deployment of Vision-Language Models (VLMs) in dynamic environments necessitates the ability to learn continuously without forgetting. However, traditional continual learning (CL) settings often rely on white-box paradigms, which is increasingly invalidated by the shift toward cloud-hosted models. In this paper, we introduce Black-CL, a more realistic benchmark for VLMs that enforces three primary real-world challenges: weight and architecture inaccessibility, constrained computation, and task-agnostic inference. The learner can query only output embeddings or logits, with no gradient flow through or structural modification of the backbone. Current CL methodologies, which rely on backbone backpropagation or complex parameter expansion, are fundamentally incompatible with these constraints. Under this setting, we propose BETA, a simple yet effective baseline built on the key insight that solely optimizing textual prototypes can navigate the complexities of CL. BETA integrates three core components: Semantic Projection Accumulation (SPA) for incremental knowledge acquisition, Latent Distribution Replay (LDR) for anchoring the embedding space against catastrophic forgetting, and Test-Time Prototype Adaptation (TTPA) for dynamic, instance-aware boundary refinement. Extensive experiments across ten diverse datasets and various backbones demonstrate that BETA significantly outperforms existing black-box tuners. Remarkably, with only 0.05 M trainable parameters, a 180--3000$\times$ reduction compared to competitive methods, BETA achieves performance on par with or even exceeding white-box CL methods. We believe Black-CL and BETA provide a foundational framework for future advancements in continual learning and accelerates the transition of continual learning from academia to real-world systems.

cs.CV↗

arXiv2026

Measuring Maximum Activations in Open Large Language Models

The dynamic range of activations is a first-order constraint for low-bit quantization, activation scaling, and stable LLM inference. Prior work characterized outlier features and massive activations on pre-2024 LLaMA-style models, and the downstream activation-quantization stack inherits that picture without revisiting it for the post-LLaMA open-model boom. We ask the deployment-oriented question: how large can activations get in modern open LLMs, and how does this magnitude vary across families, generations, and training stages? Under a unified pipeline (5,000-sample multi-domain corpus, family-specific tokenization, identical hooks across embeddings, hidden states, attention, MLP/MoE, SwiGLU gates, and final norm), we measure global and layerwise maxima on 27 checkpoints from 8 open families spanning dense, MoE, vision-language, intermediate-training, and instruction-tuned variants. We find that (i) global maxima span over nearly four orders of magnitude at comparable parameter counts, with Qwen3.5 and MoE checkpoints in the 10^2 to 10^3 range and Gemma3-27B-it reaching ~7 x 10^5; (ii) cross-family and cross-generation comparisons break simple monotonic scaling; and (iii) MoE checkpoints exhibit 14.0-23.4x lower peaks than matched-scale dense counterparts, while the residual stream carries the global maximum in 22/24 checkpoints. A lightweight INT-8 sanity check shows that measured maxima co-vary with low-bit reconstruction error via activation-scale selection. We conclude that maximum activation magnitude is a model property tied to family, architecture, and training stage - not a simple byproduct of size - and should be measured and reported alongside any open-weight release before low-bit deployment. The code is publicly available at https://github.com/clx1415926/Max_act_llm.

cs.CL↗

arXiv2026

EndPrompt: Efficient Long-Context Extension via Terminal Anchoring

Extending the context window of large language models typically requires training on sequences at the target length, incurring quadratic memory and computational costs that make long-context adaptation expensive and difficult to reproduce. We propose EndPrompt, a method that achieves effective context extension using only short training sequences. The core insight is that exposing a model to long-range relative positional distances does not require constructing full-length inputs: we preserve the original short context as an intact first segment and append a brief terminal prompt as a second segment, assigning it positional indices near the target context length. This two-segment construction introduces both local and long-range relative distances within a short physical sequence while maintaining the semantic continuity of the training text--a property absent in chunk-based simulation approaches that split contiguous context. We provide a theoretical analysis grounded in Rotary Position Embedding and the Bernstein inequality, showing that position interpolation induces a rigorous smoothness constraint over the attention function, with shared Transformer parameters further suppressing unstable extrapolation to unobserved intermediate distances. Applied to LLaMA-family models extending the context window from 8K to 64K, EndPrompt achieves an average RULER score of 76.03 and the highest average on LongBench, surpassing LCEG (72.24), LongLoRA (72.95), and full-length fine-tuning (69.23) while requiring substantially less computation. These results demonstrate that long-context generalization can be induced from sparse positional supervision, challenging the prevailing assumption that dense long-sequence training is necessary for reliable context-window extension. The code is available at https://github.com/clx1415926/EndPrompt.

cs.CL↗

arXiv2026

DVFace: Spatio-Temporal Dual-Prior Diffusion for Video Face Restoration

Video face restoration aims to enhance degraded face videos into high-quality results with realistic facial details, stable identity, and temporal coherence. Recent diffusion-based methods have brought strong generative priors to restoration and enabled more realistic detail synthesis. However, existing approaches for face videos still rely heavily on generic diffusion priors and multi-step sampling, which limit both facial adaptation and inference efficiency. These limitations motivate the use of one-step diffusion for video face restoration, yet achieving faithful facial recovery alongside temporally stable outputs remains challenging. In this paper, we propose, DVFace, a one-step diffusion framework for real-world video face restoration. Specifically, we introduce a spatio-temporal dual-codebook design to extract complementary spatial and temporal facial priors from degraded videos. We further propose an asymmetric spatio-temporal fusion module to inject these priors into the diffusion backbone according to their distinct roles. Evaluation on various benchmarks shows that DVFace delivers superior restoration quality, temporal consistency, and identity preservation compared to recent methods. Code: https://github.com/zhengchen1999/DVFace.

cs.CV↗

arXiv2026

AdaFuse: Accelerating Dynamic Adapter Inference via Token-Level Pre-Gating and Fused Kernel Optimization

The integration of dynamic, sparse structures like Mixture-of-Experts (MoE) with parameter-efficient adapters (e.g., LoRA) is a powerful technique for enhancing Large Language Models (LLMs). However, this architectural enhancement comes at a steep cost: despite minimal increases in computational load, the inference latency often skyrockets, leading to decoding speeds slowing by over 2.5 times. Through a fine-grained performance analysis, we pinpoint the primary bottleneck not in the computation itself, but in the severe overhead from fragmented, sequential CUDA kernel launches required for conventional dynamic routing. To address this challenge, we introduce AdaFuse, a framework built on a tight co-design between the algorithm and the underlying hardware system to enable efficient dynamic adapter execution. Departing from conventional layer-wise or block-wise routing, AdaFuse employs a token-level pre-gating strategy, which makes a single, global routing decision for all adapter layers before a token is processed. This "decide-once, apply-everywhere" approach effectively staticizes the execution path for each token, creating an opportunity for holistic optimization. We capitalize on this by developing a custom CUDA kernel that performs a fused switching operation, merging the parameters of all selected LoRA adapters into the backbone model in a single, efficient pass. Experimental results on popular open-source LLMs show that AdaFuse achieves accuracy on par with state-of-the-art dynamic adapters while drastically cutting decoding latency by a factor of over 2.4x, thereby bridging the gap between model capability and inference efficiency.

cs.AI↗

arXiv2026

IDER: IDempotent Experience Replay for Reliable Continual Learning

Catastrophic forgetting, the tendency of neural networks to forget previously learned knowledge when learning new tasks, has been a major challenge in continual learning (CL). To tackle this challenge, CL methods have been proposed and shown to reduce forgetting. Furthermore, CL models deployed in mission-critical settings can benefit from uncertainty awareness by calibrating their predictions to reliably assess their confidences. However, existing uncertainty-aware continual learning methods suffer from high computational overhead and incompatibility with mainstream replay methods. To address this, we propose idempotent experience replay (IDER), a novel approach based on the idempotent property where repeated function applications yield the same output. Specifically, we first adapt the training loss to make model idempotent on current data streams. In addition, we introduce an idempotence distillation loss. We feed the output of the current model back into the old checkpoint and then minimize the distance between this reprocessed output and the original output of the current model. This yields a simple and effective new baseline for building reliable continual learners, which can be seamlessly integrated with other CL approaches. Extensive experiments on different CL benchmarks demonstrate that IDER consistently improves prediction reliability while simultaneously boosting accuracy and reducing forgetting. Our results suggest the potential of idempotence as a promising principle for deploying efficient and trustworthy continual learning systems in real-world applications.Our code is available at https://github.com/YutingLi0606/Idempotent-Continual-Learning.

cs.LG↗

arXiv2026

Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception

Multimodal Large Language Models (MLLMs) excel at broad visual understanding but still struggle with fine-grained perception, where decisive evidence is small and easily overwhelmed by global context. Recent "Thinking-with-Images" methods alleviate this by iteratively zooming in and out regions of interest during inference, but incur high latency due to repeated tool calls and visual re-encoding. To address this, we propose Region-to-Image Distillation, which transforms zooming from an inference-time tool into a training-time primitive, thereby internalizing the benefits of agentic zooming into a single forward pass of an MLLM. In particular, we first zoom in to micro-cropped regions to let strong teacher models generate high-quality VQA data, and then distill this region-grounded supervision back to the full image. After training on such data, the smaller student model improves "single-glance" fine-grained perception without tool use. To rigorously evaluate this capability, we further present ZoomBench, a hybrid-annotated benchmark of 845 VQA data spanning six fine-grained perceptual dimensions, together with a dual-view protocol that quantifies the global--regional "zooming gap". Experiments show that our models achieve leading performance across multiple fine-grained perception benchmarks, and also improve general multimodal cognition on benchmarks such as visual reasoning and GUI agents. We further discuss when "Thinking-with-Images" is necessary versus when its gains can be distilled into a single forward pass. Our code is available at https://github.com/inclusionAI/Zooming-without-Zooming.

cs.CV↗

arXiv2026

Pedagogically-Inspired Data Synthesis for Language Model Knowledge Distillation

Knowledge distillation from Large Language Models (LLMs) to smaller models has emerged as a critical technique for deploying efficient AI systems. However, current methods for distillation via synthetic data lack pedagogical awareness, treating knowledge transfer as a one-off data synthesis and training task rather than a systematic learning process. In this paper, we propose a novel pedagogically-inspired framework for LLM knowledge distillation that draws from fundamental educational principles. Our approach introduces a three-stage pipeline -- Knowledge Identifier, Organizer, and Adapter (IOA) -- that systematically identifies knowledge deficiencies in student models, organizes knowledge delivery through progressive curricula, and adapts representations to match the cognitive capacity of student models. We integrate Bloom's Mastery Learning Principles and Vygotsky's Zone of Proximal Development to create a dynamic distillation process where student models approach teacher model's performance on prerequisite knowledge before advancing, and new knowledge is introduced with controlled, gradual difficulty increments. Extensive experiments using LLaMA-3.1/3.2 and Qwen2.5 as student models demonstrate that IOA achieves significant improvements over baseline distillation methods, with student models retaining 94.7% of teacher performance on DollyEval while using less than 1/10th of the parameters. Our framework particularly excels in complex reasoning tasks, showing 19.2% improvement on MATH and 22.3% on HumanEval compared with state-of-the-art baselines.

cs.AI↗

arXiv2026

PlanViz: Evaluating Planning-Oriented Image Generation and Editing for Computer-Use Tasks

Unified multimodal models (UMMs) have shown impressive capabilities in generating natural images and supporting multimodal reasoning. However, their potential in supporting computer-use planning tasks, which are closely related to our lives, remain underexplored. Image generation and editing in computer-use tasks require capabilities like spatial reasoning and procedural understanding, and it is still unknown whether UMMs have these capabilities to finish these tasks or not. Therefore, we propose PlanViz, a new benchmark designed to evaluate image generation and editing for computer-use tasks. To achieve the goal of our evaluation, we focus on sub-tasks which frequently involve in daily life and require planning. Specifically, three representative sub-tasks are designed: route planning, work diagramming, and web&UI displaying. We address challenges in data quality ensuring by curating human-annotated questions and reference images, and a quality control process. For detailed and exact evaluation, a task-adaptive score, PlanScore, is proposed. The score helps understanding the correctness, visual quality and efficiency of generated images. Through experiments, we highlight key limitations and opportunities for future research on this topic.

cs.CV↗

arXiv2026

BinaryDemoire: Moir\'e-Aware Binarization for Image Demoir\'eing

Image demoir\'eing aims to remove structured moir\'e artifacts in recaptured imagery, where degradations are highly frequency-dependent and vary across scales and directions. While recent deep networks achieve high-quality restoration, their full-precision designs remain costly for deployment. Binarization offers an extreme compression regime by quantizing both activations and weights to 1-bit. Yet, it has been rarely studied for demoir\'eing and performs poorly when naively applied. In this work, we propose BinaryDemoire, a binarized demoir\'eing framework that explicitly accommodates the frequency structure of moir\'e degradations. First, we introduce a moir\'e-aware binary gate (MABG) that extracts lightweight frequency descriptors together with activation statistics. It predicts channel-wise gating coefficients to condition the aggregation of binary convolution responses. Second, we design a shuffle-grouped residual adapter (SGRA) that performs structured sparse shortcut alignment. It further integrates interleaved mixing to promote information exchange across different channel partitions. Extensive experiments on four benchmarks demonstrate that the proposed BinaryDemoire surpasses current binarization methods. Code: https://github.com/zhengchen1999/BinaryDemoire.

cs.CV↗

arXiv2026

LSGQuant: Layer-Sensitivity Guided Quantization for One-Step Diffusion Real-World Video Super-Resolution

One-Step Diffusion Models have demonstrated promising capability and fast inference in video super-resolution (VSR) for real-world. Nevertheless, the substantial model size and high computational cost of Diffusion Transformers (DiTs) limit downstream applications. While low-bit quantization is a common approach for model compression, the effectiveness of quantized models is challenged by the high dynamic range of input latent and diverse layer behaviors. To deal with these challenges, we introduce LSGQuant, a layer-sensitivity guided quantizing approach for one-step diffusion-based real-world VSR. Our method incorporates a Dynamic Range Adaptive Quantizer (DRAQ) to fit video token activations. Furthermore, we estimate layer sensitivity and implement a Variance-Oriented Layer Training Strategy (VOLTS) by analyzing layer-wise statistics in calibration. We also introduce Quantization-Aware Optimization (QAO) to jointly refine the quantized branch and a retained high-precision branch. Extensive experiments demonstrate that our method has nearly performance to origin model with full-precision and significantly exceeds existing quantization techniques. Code is available at: https://github.com/zhengchen1999/LSGQuant.

cs.CV↗

arXiv2026

VEQ: Modality-Adaptive Quantization for MoE Vision-Language Models

Mixture-of-Experts(MoE) Vision-Language Models (VLMs) offer remarkable performance but incur prohibitive memory and computational costs, making compression essential. Post-Training Quantization (PTQ) is an effective training-free technique to address the massive memory and computation overhead. Existing quantization paradigms fall short as they are oblivious to two critical forms of heterogeneity: the inherent discrepancy between vision and language tokens, and the non-uniform contribution of different experts. To bridge this gap, we propose Visual Expert Quantization (VEQ), a dual-aware quantization framework designed to simultaneously accommodate cross-modal differences and heterogeneity between experts. Specifically, VEQ incorporates 1)Modality-expert-aware Quantization, which utilizes expert activation frequency to prioritize error minimization for pivotal experts, and 2)Modality-affinity-aware Quantization, which constructs an enhanced Hessian matrix by integrating token-expert affinity with modality information to guide the calibration process. Extensive experiments across diverse benchmarks verify that VEQ consistently outperforms state-of-the-art baselines. Specifically, under the W3A16 configuration, our method achieves significant average accuracy gains of 2.04\% on Kimi-VL and 3.09\% on Qwen3-VL compared to the previous SOTA quantization methods, demonstrating superior robustness across various multimodal tasks. Our code will be available at https://github.com/guangshuoqin/VEQ.

cs.CV↗

arXiv2026

FlexSpec: Frozen Drafts Meet Evolving Targets in Edge-Cloud Collaborative LLM Speculative Decoding

Deploying large language models (LLMs) in mobile and edge computing environments is constrained by limited on-device resources, scarce wireless bandwidth, and frequent model evolution. Although edge-cloud collaborative inference with speculative decoding (SD) can reduce end-to-end latency by executing a lightweight draft model at the edge and verifying it with a cloud-side target model, existing frameworks fundamentally rely on tight coupling between the two models. Consequently, repeated model synchronization introduces excessive communication overhead, increasing end-to-end latency, and ultimately limiting the scalability of SD in edge environments. To address these limitations, we propose FlexSpec, a communication-efficient collaborative inference framework tailored for evolving edge-cloud systems. The core design of FlexSpec is a shared-backbone architecture that allows a single and static edge-side draft model to remain compatible with a large family of evolving cloud-side target models. By decoupling edge deployment from cloud-side model updates, FlexSpec eliminates the need for edge-side retraining or repeated model downloads, substantially reducing communication and maintenance costs. Furthermore, to accommodate time-varying wireless conditions and heterogeneous device constraints, we develop a channel-aware adaptive speculation mechanism that dynamically adjusts the speculative draft length based on real-time channel state information and device energy budgets. Extensive experiments demonstrate that FlexSpec achieves superior performance compared to conventional SD approaches in terms of inference efficiency.

cs.DC↗