Searcharxiv⌕ Search

arXiv subjects

Bo Jiang

Publications and source records attributed to Bo Jiang.

At least 91 records · Page 5Linked to original sources
arXiv2026

Adaptive Single-Loop Methods for Stochastic Minimax Optimization on Riemannian Manifolds

Stochastic minimax optimization on Riemannian manifolds has recently attracted significant attention due to its broad range of applications, such as robust training of neural networks and robust maximum likelihood estimation. Existing optimization methods for these problems typically require selecting stepsizes based on prior knowledge of specific problem parameters, such as Lipschitz-type constants and (geodesic) strong concavity constants. Unfortunately, these parameters are often unknown in practice. To overcome this issue, we develop single-loop adaptive methods that automatically adjust stepsizes using cumulative Riemannian (stochastic) gradient norms. We first propose a deterministic single-loop Riemannian adaptive gradient descent ascent method and show that it attains an $ε$-stationary point within $O(ε^{-2})$ iterations. This deterministic method is of independent interest and lays the foundation for our subsequent stochastic method. In particular, we propose the Riemannian stochastic adaptive gradient descent ascent method, which finds an $ε$-stationary point in $O(ε^{-6})$ iterations. Under additional second-order smoothness, this iteration complexity is further improved to $O(ε^{-4})$, which even outperforms the corresponding complexity result in Euclidean space. Some numerical experiments on real-world applications are conducted, including the regularized robust maximum likelihood estimation problem, and the robust training of neural networks with orthonormal weights. The results are encouraging and demonstrate the effectiveness of adaptivity in practice.

math.OC↗
arXiv2026

FreqKV: Key-Value Compression in Frequency Domain for Context Window Extension

Existing key-value (KV) cache compression methods for large language models (LLMs) often rely on token eviction, which risks losing critical local information in both long prefilling and decoding scenarios. When extrapolating beyond the pretrained context length, their performance degrades sharply on long-context benchmarks. Motivated by the observation in the frequency domain that the context information is concentrated in the low-frequency components, we propose FreqKV, a parameter-free and architecture-agnostic approach. It iteratively compresses the increasing KV cache in the frequency domain, allowing models to process lengthy contexts efficiently. With minimal training at 8K length, FreqKV extends the context window of LLaMA-2-7B up to 256K tokens while maintaining stable perplexity. Extensive experiments across prefilling and decoding demonstrate that FreqKV enables robust context window extension and consistently outperforms existing KV cache compression methods on LLaMA-2 and LLaMA-3, highlighting its effectiveness for both understanding and generation in long contexts.

cs.CL↗
arXiv2026

Reversible Diffusion Decoding for Diffusion Language Models

Diffusion language models enable parallel token generation through block-wise decoding, but their irreversible commitments can lead to stagnation, where the reverse diffusion process fails to make further progress under a suboptimal context.We propose Reversible Diffusion Decoding (RDD), a decoding framework that introduces reversibility into block-wise diffusion generation. RDD detects stagnation as a state-dependent failure of the reverse process and enables efficient backtracking to earlier blocks without recomputation via cached model states. To avoid repeated failure trajectories, RDD applies confidence-guided re-masking to selectively reinitialize uncertain tokens while preserving reliable context.This reversible formulation allows decoding to recover from early commitment errors while maintaining the parallel efficiency of diffusion-based generation. Experiments show that RDD improves generation robustness and quality over baselines with minimal computational overhead.

cs.CL↗
arXiv2026

A Comprehensive Exploration of Personalized Learning in Smart Education: From Student Modeling to Personalized Recommendations

With the development of artificial intelligence, personalized learning has attracted much attention as an integral part of intelligent education. In recent years, countries and regions such as China, the United States, and the European Union have increasingly recognized the importance of personalized learning, emphasizing its potential to integrate large-scale education with individualized instruction effectively. This survey provides a comprehensive analysis of personalized learning by reviewing relevant studies published in major conferences and journals between January 2017 and April 2025. We examine its definition, objectives, and underlying educational theories, highlighting its pedagogical significance. Furthermore, we explore personalized learning from two key dimensions: student modeling and personalized recommendations. Student modeling is analyzed from both cognitive and non-cognitive perspectives, while recommendation approaches are categorized based on their specific objectives. Additionally, we investigate the interplay between these components and their role in enhancing personalized learning. Beyond theoretical and algorithmic insights, this survey reviews real-world applications, demonstrating personalized learning's effectiveness in educational practice. Finally, we discuss key challenges and future directions, offering a multidimensional perspective that bridges theory and practice.

cs.CY↗
arXiv2026

On series identities involving $\binom{4k}k$ and harmonic numbers

The harmonic numbers are those $H_n=\sum_{0<k\le n}\frac1k\ (n=0,1,2,\ldots)$. In this paper we confirm over ten conjectural series identities with summands involving the binomial coefficient $\binom{4k}k$ and harmonic numbers. For example, we prove the identities $$\sum_{k=1}^\infty \frac{\binom{4k}{k}}{16^k}\left((22k^2-92k+11)H_{4k}-\frac{449k-275}{2}-\frac{85}{12k}\right)=-151-\frac{80}{3}\log{2}$$ and $$ \sum_{k=0}^\infty\frac{\binom{4k}{k}((11k^2+8k+1)(10H_{4k}-17H_{2k})+2k+18)}{(3k+1)(3k+2)16^k}=8\log2,$$ which were previously conjectured by Z.-W. Sun.

math.NT↗
arXiv2026

Beyond Nonconvexity: A Universal Trust-Region Method with New Analyses

The trust-region (TR) method is renowned historically for its robustness in nonconvex problems and extraordinary numerical performance, but the study of its performance in convex optimization is somehow limited. This paper complements the existing literature by presenting a universal trust-region method that simultaneously incorporates the quadratic regularization and ball constraint. In particular, we introduce a novel descent property tailored for trust-region-type algorithms, enabling us to unify and streamline the analysis for both convex and nonconvex optimization. Our method exhibits an iteration complexity of $\tilde O(ε^{-3/2})$ to find an $ε$-approximate second-order stationary point for nonconvex optimization. Meanwhile, the analysis reveals that the universal method attains an $O(ε^{-1/2})$ complexity bound for convex optimization. Finally, we develop an adaptive universal method to address practical implementations. The numerical results show the effectiveness of our method in both nonconvex and convex problems.

math.OC↗
arXiv2026

Event-VStream: Event-Driven Real-Time Understanding for Long Video Streams

Real-time understanding of long video streams remains challenging for multimodal large language models (VLMs) due to redundant frame processing and rapid forgetting of past context. Existing streaming systems rely on fixed-interval decoding or cache pruning, which either produce repetitive outputs or discard crucial temporal information. We introduce Event-VStream, an event-aware framework that represents continuous video as a sequence of discrete, semantically coherent events. Our system detects meaningful state transitions by integrating motion, semantic, and predictive cues, and triggers language generation only at those boundaries. Each event embedding is consolidated into a persistent memory bank, enabling long-horizon reasoning while maintaining low latency. Across OVOBench-Realtime, and long-form Ego4D evaluations, Event-VStream achieves competitive performance. It improves over a VideoLLM-Online-8B baseline by +10.4 points on OVOBench-Realtime, achieves performance close to Flash-VStream-7B despite using only a general-purpose LLaMA-3-8B text backbone, and maintains around 70% GPT-5 win rate on 2-hour Ego4D streams.

cs.CV↗
arXiv2026

EduResearchBench: A Hierarchical Atomic Task Decomposition Benchmark for Full-Lifecycle Educational Research

While Large Language Models (LLMs) are reshaping the paradigm of AI for Social Science (AI4SS), rigorously evaluating their capabilities in scholarly writing remains a major challenge. Existing benchmarks largely emphasize single-shot, monolithic generation and thus lack the fine-grained assessments required to reflect complex academic research workflows. To fill this gap, we introduce EduResearchBench, the first comprehensive evaluation platform dedicated to educational academic writing. EduResearchBench is built upon our Hierarchical Atomic Task Decomposition (HATD) framework, which decomposes an end-to-end research workflow into six specialized research modules (e.g., Quantitative Analysis, Qualitative Research, and Policy Research) spanning 24 fine-grained atomic tasks. This taxonomy enables an automated evaluation pipeline that mitigates a key limitation of holistic scoring, where aggregate scores often obscure specific capability bottlenecks, and instead provides fine-grained, diagnostic feedback on concrete deficiencies. Moreover, recognizing the high cognitive load inherent in scholarly writing, we propose a curriculum learning strategy that progressively builds competence from foundational skills to complex methodological reasoning and argumentation. Leveraging 55K raw academic samples, we curate 11K high-quality instruction pairs to train EduWrite, a specialized educational scholarly writing model. Experiments show that EduWrite (30B) substantially outperforms larger general-purpose models (72B) on multiple core metrics, demonstrating that in vertical domains, data quality density and hierarchically staged training curricula are more decisive than parameter scale.

cs.CL↗
arXiv2026

PackKV: Reducing KV Cache Memory Footprint through LLM-Aware Lossy Compression

Transformer-based large language models (LLMs) have demonstrated remarkable potential across a wide range of practical applications. However, long-context inference remains a significant challenge due to the substantial memory requirements of the key-value (KV) cache, which can scale to several gigabytes as sequence length and batch size increase. In this paper, we present \textbf{PackKV}, a generic and efficient KV cache management framework optimized for long-context generation. %, which synergistically supports both latency-critical and throughput-critical inference scenarios. PackKV introduces novel lossy compression techniques specifically tailored to the characteristics of KV cache data, featuring a careful co-design of compression algorithms and system architecture. Our approach is compatible with the dynamically growing nature of the KV cache while preserving high computational efficiency. Experimental results show that, under the same and minimum accuracy drop as state-of-the-art quantization methods, PackKV achieves, on average, \textbf{153.2}\% higher memory reduction rate for the K cache and \textbf{179.6}\% for the V cache. Furthermore, PackKV delivers extremely high execution throughput, effectively eliminating decompression overhead and accelerating the matrix-vector multiplication operation. Specifically, PackKV achieves an average throughput improvement of \textbf{75.7}\% for K and \textbf{171.7}\% for V across A100 and RTX Pro 6000 GPUs, compared to cuBLAS matrix-vector multiplication kernels, while demanding less GPU memory bandwidth. Code available on https://github.com/BoJiang03/PackKV

cs.DC↗
arXiv2026

When Prompting Meets Spiking: Graph Sparse Prompting via Spiking Graph Prompt Learning

Graph Prompt Feature (GPF) learning has been widely used in adapting pre-trained GNN model on the downstream task. GPFs first introduce some prompt atoms and then learns the optimal prompt vector for each graph node using the linear combination of prompt atoms. However, existing GPFs generally conduct prompting over node's all feature dimensions which is obviously redundant and also be sensitive to node feature noise. To overcome this issue, for the first time, this paper proposes learning sparse graph prompts by leveraging the spiking neuron mechanism, termed Spiking Graph Prompt Feature (SpikingGPF). Our approach is motivated by the observation that spiking neuron can perform inexpensive information processing and produce sparse outputs which naturally fits the task of our graph sparse prompting. Specifically, SpikingGPF has two main aspects. First, it learns a sparse prompt vector for each node by exploiting a spiking neuron architecture, enabling prompting on selective node features. This yields a more compact and lightweight prompting design while also improving robustness against node noise. Second, SpikingGPF introduces a novel prompt representation learning model based on sparse representation theory, i.e., it represents each node prompt as a sparse combination of prompt atoms. This encourages a more compact representation and also facilitates efficient computation. Extensive experiments on several benchmarks demonstrate the effectiveness and robustness of SpikingGPF.

cs.LG↗
arXiv2026

Sentient: Detecting APTs Via Capturing Indirect Dependencies and Behavioral Logic

Advanced Persistent Threats (APTs) are difficult to detect due to their complexity and stealthiness. To mitigate such attacks, many approaches model entities and their relationship using provenance graphs to detect the stealthy and persistent characteristics of APTs. However, existing detection methods suffer from the flaws of missing indirect dependencies, noisy complex scenarios, and missing behavioral logical associations, which make it difficult to detect complex scenarios and effectively identify stealthy threats. In this paper, we propose Sentient, an APT detection method that combines pre-training and intent analysis. It employs a graph transformer to learn structural and semantic information from provenance graphs to avoid missing indirect dependencies. We mitigate scenario noise by combining global and local information. Additionally, we design an Intent Analysis Module (IAM) to associate logical relationships between behaviors. Sentient is trained solely on easily obtainable benign data to detect malicious behaviors that deviate from benign behavioral patterns. We evaluated Sentient on three widely-used datasets covering real-world attacks and simulated attacks. Notably, compared to six state-of-the-art methods, Sentient achieved an average reduction of 44% in false positive rate(FPR) for detection.

cs.CR↗
arXiv2026

Multi-Subspace Multi-Modal Modeling for Diffusion Models: Estimation, Convergence and Mixture of Experts

Recently, diffusion models have achieved a great performance with a small dataset of size $n$ and a fast optimization process. However, the estimation error of diffusion models suffers from the curse of dimensionality $n^{-1/D}$ with the data dimension $D$. Since images are usually a union of low-dimensional manifolds, current works model the data as a union of linear subspaces with Gaussian latent and achieve a $1/\sqrt{n}$ bound. Though this modeling reflects the multi-manifold property, the Gaussian latent can not capture the multi-modal property of the latent manifold. To bridge this gap, we propose the mixture subspace of low-rank mixture of Gaussian (MoLR-MoG) modeling, which models the target data as a union of $K$ linear subspaces, and each subspace admits a mixture of Gaussian latent ($n_k$ modals with dimension $d_k$). With this modeling, the corresponding score function naturally has a mixture of expert (MoE) structure, captures the multi-modal information, and contains nonlinear property. We first conduct real-world experiments to show that the generation results of MoE-latent MoG NN are much better than MoE-latent Gaussian score. Furthermore, MoE-latent MoG NN achieves a comparable performance with MoE-latent Unet with $10 \times$ parameters. These results indicate that the MoLR-MoG modeling is reasonable and suitable for real-world data. After that, based on such MoE-latent MoG score, we provide a $R^4\sqrt{Σ_{k=1}^Kn_k}\sqrt{Σ_{k=1}^Kn_kd_k}/\sqrt{n}$ estimation error, which escapes the curse of dimensionality by using data structure. Finally, we study the optimization process and prove the convergence guarantee under the MoLR-MoG modeling. Combined with these results, under a setting close to real-world data, this work explains why diffusion models only require a small training sample and enjoy a fast optimization process to achieve a great performance.

cs.LG↗
arXiv2026

Decoupling Amplitude and Phase Attention in Frequency Domain for RGB-Event based Visual Object Tracking

Existing RGB-Event visual object tracking approaches primarily rely on conventional feature-level fusion, failing to fully exploit the unique advantages of event cameras. In particular, the high dynamic range and motion-sensitive nature of event cameras are often overlooked, while low-information regions are processed uniformly, leading to unnecessary computational overhead for the backbone network. To address these issues, we propose a novel tracking framework that performs early fusion in the frequency domain, enabling effective aggregation of high-frequency information from the event modality. Specifically, RGB and event modalities are transformed from the spatial domain to the frequency domain via the Fast Fourier Transform, with their amplitude and phase components decoupled. High-frequency event information is selectively fused into RGB modality through amplitude and phase attention, enhancing feature representation while substantially reducing backbone computation. In addition, a motion-guided spatial sparsification module leverages the motion-sensitive nature of event cameras to capture the relationship between target motion cues and spatial probability distribution, filtering out low-information regions and enhancing target-relevant features. Finally, a sparse set of target-relevant features is fed into the backbone network for learning, and the tracking head predicts the final target position. Extensive experiments on three widely used RGB-Event tracking benchmark datasets, including FE108, FELT, and COESOT, demonstrate the high performance and efficiency of our method. The source code of this paper will be released on https://github.com/Event-AHU/OpenEvTracking

cs.CV↗
arXiv2026

Robust Graph Fine-Tuning with Adversarial Graph Prompting

Parameter-Efficient Fine-Tuning (PEFT) method has emerged as a dominant paradigm for adapting pre-trained GNN models to downstream tasks. However, existing PEFT methods usually exhibit significant vulnerability to various noise and attacks on graph topology and node attributes/features. To address this issue, for the first time, we propose integrating adversarial learning into graph prompting and develop a novel Adversarial Graph Prompting (AGP) framework to achieve robust graph fine-tuning. Our AGP has two key aspects. First, we propose the general problem formulation of AGP as a min-max optimization problem and develop an alternating optimization scheme to solve it. For inner maximization, we propose Joint Projected Gradient Descent (JointPGD) algorithm to generate strong adversarial noise. For outer minimization, we employ a simple yet effective module to learn the optimal node prompts to counteract the adversarial noise. Second, we demonstrate that the proposed AGP can theoretically address both graph topology and node noise. This confirms the versatility and robustness of our AGP fine-tuning method across various graph noise. Note that, the proposed AGP is a general method that can be integrated with various pre-trained GNN models to enhance their robustness on the downstream tasks. Extensive experiments on multiple benchmark tasks validate the robustness and effectiveness of AGP method compared to state-of-the-art methods.

cs.LG↗
arXiv2025

Benchmarking LLMs for Fine-Grained Code Review with Enriched Context in Practice

Code review is a cornerstone of software quality assurance, and recent advances in Large Language Models (LLMs) have shown promise in its automation. However, existing benchmarks for LLM-based code review face three major limitations. Lack of semantic context: most benchmarks provide only code diffs without textual information such as issue descriptions, which are crucial for understanding developer intent. Data quality issues: without rigorous validation, many samples are noisy-e.g., reviews on outdated or irrelevant code-reducing evaluation reliability. Coarse granularity: most benchmarks operate at the file or commit level, overlooking the fine-grained, line-level reasoning essential for precise review. We introduce ContextCRBench, a high-quality, context-rich benchmark for fine-grained LLM evaluation in code review. Our construction pipeline comprises: Raw Data Crawling, collecting 153.7K issues and pull requests from top-tier repositories; Comprehensive Context Extraction, linking issue-PR pairs for textual context and extracting the full surrounding function or class for code context; and Multi-stage Data Filtering, combining rule-based and LLM-based validation to remove outdated, malformed, or low-value samples, resulting in 67,910 context-enriched entries. ContextCRBench supports three evaluation scenarios aligned with the review workflow: hunk-level quality assessment, line-level defect localization, and line-level comment generation. Evaluating eight leading LLMs (four closed-source and four open-source) reveals that textual context yields greater performance gains than code context alone, while current LLMs remain far from human-level review ability. Deployed at ByteDance, ContextCRBench drives a self-evolving code review system, improving performance by 61.98% and demonstrating its robustness and industrial utility. https://github.com/kinesiatricssxilm14/ContextCRBench.

cs.SE↗
arXiv2025

Context as a Tool: Context Management for Long-Horizon SWE-Agents

Agents based on large language models have recently shown strong potential on real-world software engineering (SWE) tasks that require long-horizon interaction with repository-scale codebases. However, most existing agents rely on append-only context maintenance or passively triggered compression heuristics, which often lead to context explosion, semantic drift, and degraded reasoning in long-running interactions. We propose CAT, a new context management paradigm that elevates context maintenance to a callable tool integrated into the decision-making process of agents. CAT formalizes a structured context workspace consisting of stable task semantics, condensed long-term memory, and high-fidelity short-term interactions, and enables agents to proactively compress historical trajectories into actionable summaries at appropriate milestones. To support context management for SWE-agents, we propose a trajectory-level supervision framework, CAT-GENERATOR, based on an offline data construction pipeline that injects context-management actions into complete interaction trajectories. Using this framework, we train a context-aware model, SWE-Compressor. Experiments on SWE-Bench-Verified demonstrate that SWE-Compressor reaches a 57.6% solved rate and significantly outperforms ReAct-based agents and static compression baselines, while maintaining stable and scalable long-horizon reasoning under a bounded context budget.

cs.CL↗
arXiv2025

Estimating Privacy Leakage of Augmented Contextual Knowledge in Language Models

Language models (LMs) rely on their parametric knowledge augmented with relevant contextual knowledge for certain tasks, such as question answering. However, the contextual knowledge can contain private information that may be leaked when answering queries, and estimating this privacy leakage is not well understood. A straightforward approach of directly comparing an LM's output to the contexts can overestimate the privacy risk, since the LM's parametric knowledge might already contain the augmented contextual knowledge. To this end, we introduce *context influence*, a metric that builds on differential privacy, a widely-adopted privacy notion, to estimate the privacy leakage of contextual knowledge during decoding. Our approach effectively measures how each subset of the context influences an LM's response while separating the specific parametric knowledge of the LM. Using our context influence metric, we demonstrate that context privacy leakage occurs when contextual knowledge is out of distribution with respect to parametric knowledge. Moreover, we experimentally demonstrate how context influence properly attributes the privacy leakage to augmented contexts, and we evaluate how factors -- such as model size, context size, generation position, etc. -- affect context privacy leakage. The practical implications of our results will inform practitioners of the privacy risk associated with augmented contextual knowledge.

cs.CL↗
arXiv2025

On Approximation Algorithms for Commutative Quaternion Polynomial Optimization

Quaternion optimization has attracted significant interest due to its broad applications, including color face recognition, video compression, and signal processing. Despite the growing literature on quadratic and matrix quaternion optimization, to the best of our knowledge, the study on quaternion polynomial optimization still remains blank. In this paper, we introduce the first investigation into this fundamental problem, and focus on the sphere-constrained homogeneous polynomial optimization over the commutative quaternion domain, which includes the best rank-one tensor approximation as a special case. Our study proposes a polynomial-time randomized approximation algorithm that employs tensor relaxation and random sampling techniques to tackle this problem. Theoretically, we prove an approximation ratio for the algorithm providing a worst-case performance guarantee

math.OC↗