论文检索

Rui Wang 0043, Ziang Li, Haiping Huang, Jialin Yu 0001, Yuxiang Zhou, Guozi Sun

Existing multi-label topic models face several challenges when interpreting texts annotated with multiple labels: (1) they often associate irrelevant text segments with incorrect labels, which negatively impacts both segment and label interpretation; (2) they fail to effectively capture the semantic relationships between tokens and labels within the segment; (3) they do not integrate contextualized knowledge that could improve interpretability. To overcome these issues, we introduce the Contextualized Prompting Topic Model (CPTM). CPTM utilizes Chain-of-Thought (CoT) prompting to better align text segments with their semantically relevant labels. Furthermore, it integrates label-specific token visualization and topic mining procedure to facilitate the interpretation of tokens and labels. Experimental evaluations conducted on three multi-label text datasets show that CPTM significantly outperforms existing models in both segment and label interpretation. Human assessments also verify CPTM's effectiveness in accurately identifying label-relevant tokens within segments and providing insightful token-level interpretation.

Hua Wang 0012, Jinghao Lu, Fan Zhang 0045

Transformer-based foundation models have achieved remarkable progress in tasks such as time-series forecasting and image segmentation. However, they frequently suffer from error accumulation in multivariate long-sequence prediction and exhibit vulnerability to out-of-distribution samples in image-related tasks. Furthermore, these challenges become particularly pronounced in large-scale Web data analysis tasks, which typically involve complex temporal patterns and multimodal features. This complexity substantially increases optimization difficulty, rendering models prone to stagnation at saddle points within high-dimensional parameter spaces. To address these issues, we propose a lightweight Transformer architecture in conjunction with a novel Escape-Explore Optimizer (EEO). The optimizer enhances both exploration and generalization while effectively avoiding sharp minima and saddle-point traps. Experimental results show that, in representative Web data scenarios, our method achieves performance on par with state-of-the-art models across 11 time-series benchmark datasets and the Synapse medical image segmentation task. Moreover, it demonstrates superior generalization and stability, thereby validating its potential as a versatile cross-task foundation model for Web-scale data mining and analysis.

Shuili Zhang, Hongzhang Mu, Wenyuan Zhang 0002, Duohe Ma, Tingwen Liu

Attribute-Specific Fashion Retrieval (ASFR) aims to improve fine-grained image retrieval by focusing on specific attributes. However, existing patch-based attention and Transformer methods often misalign with irregular attribute regions and are prone to background noise, limiting their ability to capture subtle, pixel-level microstructures. To tackle these challenges, we propose Super Fashion. , the first ASFR framework that adopts superpixel tokens within a Transformer architecture. Super Fashion initially employs an attribute-guided attention mechanism to extract attribute-related features, which in turn guide the cropping of semantically meaningful image regions. Superpixel segmentation is then leveraged on these regions to generate compact, semantically coherent superpixel tokens. By incorporating modality-specific embeddings for both attribute and superpixel tokens, the superpixel token-based Transformer facilitates adaptive interaction and fusion, thereby enhancing attribute localization and discrimination. Extensive experiments on FashionAI, DARN, and DeepFashion demonstrate relative overall MAP improvements of 1.84%, 9.27%, and 9.35% over prior SOTA. Super Fashion offers a new solution for web-based image retrieval.

Yilin Zhao, Jiawei Huang 0001, Xianshi Su, Weihe Li, Xin Li, Yan Liu, Jiacheng Xie, Qichen Su, Jin Ye 0003, Wanchun Jiang 等

Accurately detecting super host that establishes connections to a large number of distinct peers is significant for mitigating web attacks and ensuring high quality of web service. Existing sketch-based approaches estimate the number of distinct connections called flow cardinality according to full IP addresses, while ignoring the fact that a malicious or victim super host often communicates with hosts within the same subnet, resulting in high false positive rates and low accuracy. Though hierarchical-structure based approaches could capture flow cardinality in subnet, they inherently suffer from high memory usage. To address these limitations, we propose SegSketch, a segmented cardinality estimation approach that employs a lightweight halved-segment hashing strategy to infer common prefix lengths of IP addresses, and estimates cardinality within subnet to enhance detection accuracy under constrained memory size. Experiments driven by real-world traces demonstrate that, SegSketch improves F1-Score by up to 8.04× compared to state-of-the-art solutions, particularly under small memory budgets.

Michaël Soumm, Alexandre Fournier-Montgieux, Adrian Popescu 0001, Bertrand Delezoide

The performance of Recommender Systems (RS) varies significantly across users, yet the underlying reasons for this variance remain poorly understood. This paper introduces a unified framework to analyze and explain this performance gap by quantifying user profile characteristics. We propose two novel, information-theoretic measures: Mean Surprise (𝑆(𝑢)), which captures a user's deviation from popular items and is closely related to popularity bias, and Mean Conditional Surprise (𝐂𝑆(𝑢)), which measures the internal coherence of a user's interactions in a domain-agnostic manner. Through extensive experiments on 7 algorithms and 9 datasets, we demonstrate that these measures are strong predictors of recommendation performance. Our analysis reveals that performance gains from complex models are concentrated on ''coherent'' users, while all algorithms perform poorly on ''incoherent'' users. We show how these measures provide practical utility for the Web community by: (1) enabling robust, stratified evaluation to identify model weaknesses; (2) facilitating a novel analysis of the behavioral alignment of recommendations; and (3) guiding targeted system design, which we validate by training a specialized model on a segment of ''coherent'' users that achieves superior performance for that group with significantly less data. This work provides a new lens for understanding user behavior and offers practical tools for building more robust and efficient large-scale recommender systems.

Gu Tang, Jinghe Wang, Jiang Bo, Ze Zhao, Jianping Zhou 0004, Xiaoying Gan, Luoyi Fu, Xinbing Wang, Chenghu Zhou

Rapid development of web services has led to an explosion of multimodal content, making multimodal recommender systems (MRSs) vital tools for mitigating information overload. Current MRSs have achieved remarkable progress by incorporating advanced technologies such as Graph Neural Networks (GNNs) and Large Language Models (LLMs). However, these studies still suffer from the semantic shift problem. Generally, item's multimodal content usually contain multiple objects, including target object (core content of item) and auxiliary objects (decorations of item). Existing MRSs overlooked this distinction, failing to prevent auxiliary objects from dominating the representation, leading to biased item representation. To address this issue, we propose a model-agnostic framework ''TargetMR''. Concretely, TargetMR comprises two core modules, including Object Disentangler and Object Identifier. The Object Disentangler decouples item text and image into multiple objects via text syntactic parsing and image segmentation. The Object Identifier performs knowledge distillation based on LLMs to efficiently identify the target text object. It then identifies the target image object through cross-modal semantic evaluation. Moreover, this module refines the representation of image target object by optimizing the semantic correlation. Owing to the model-agnostic design of TargetMR, it can be integrated into various backbone MRSs. Extensive experiments on three benchmark datasets show that TargetMR consistently improves the performance of five backbone MRSs, with an average improvement of 12.26%. Our codes are available at https://github.com/gutang-97/TargetMR/.

Mingyu Zhao, Haoran Bai, Yu Tian, Bing Zhu, Hengliang Luo

Customer Lifetime Value (LTV) prediction, a central problem in modern marketing, is characterized by a unique zero-inflated and long-tail data distribution. This distribution presents two fundamental challenges: (1) the vast majority of low-to-medium value users numerically overwhelm the small but critically important segment of high-value ''whale'' users, and (2) significant value heterogeneity exists even within the low-to-medium value user base. Common approaches either rely on rigid statistical assumptions or attempt to decouple ranking and regression using ordered buckets; however, they often enforce ordinality through loss-based constraints rather than inherent architectural design, failing to balance global accuracy with high-value precision. To address this gap, we propose Conditional Cascaded Ordinal-Residual Networks (CC-OR-Net), a novel unified framework that achieves a more robust decoupling through structural decomposition, where ranking is architecturally guaranteed. CC-OR-Net integrates three specialized components: a structural ordinal decomposition module for robust ranking, an intra-bucket residual module for fine-grained regression, and a targeted high-value augmentation module for precision on top-tier users. Evaluated on real-world datasets with over 300M users, CC-OR-Net achieves a superior trade-off across all key business metrics, outperforming state-of-the-art methods in creating a holistic and commercially valuable LTV prediction solution.

Fei Li 0044, Xiaoming Liu, Jiayi Luo, Guibing Guo, Jianzhe Zhao, Xingwei Wang 0001

Lifelong sequential recommendation aims to model users' long-term interests by leveraging their entire interaction history, but the high computational overhead caused by ultra-long sequences poses a core challenge. To address this, existing methods generally adopt the subsequence learning strategies to shorten the input sequence, which can be divided into two categories: (1) Sequence compression methods compress long sequences into multiple subsequence representations through strategies such as uniform segmentation and clustering; (2) Top-k retrieval methods filter the subsequence related to the target item from long sequences via target attention mechanisms or retrieval mechanisms and learn its representation. However, these two types of methods still face challenges in subsequence representation learning: (1) Sequence compression methods struggle to simultaneously balance the high similarity of items within subsequences and smooth temporal continuity (i.e., small temporal intervals between adjacent items), resulting in incorrect learning of subsequence representations; (2) Top-k retrieval methods lose a large amount of effective context information when the length of the retrieved subsequence is much smaller than the original sequence, resulting in incomplete sequence representations. To overcome these challenges, we propose a novel lifelong sequential recommendation method with adaptive subsequence compression and contextual fusion. Specifically, an adaptive subsequence compression module is first designed: it utilizes gradient policy sampling to achieve adaptive segmentation of subsequences, thereby retaining their temporal continuity, and introduces a reward function to enhance the similarity of items within subsequences. Second, a subsequence context fusion module is constructed: it leverages causal cross-attention to fuse recent interactions with their subsequence context and to capture correlations among recent items, thereby learning more complete and accurate sequence representations. We conduct extensive experiments on three public long-sequence recommendation datasets. Experimental results demonstrate that our proposed method consistently outperforms a variety of strong baselines in both predictive accuracy and computational efficiency.

Yu Xia 0038, Qing Tan, He Chen, Jingyu Chen, Qian Dong

Large Language Models (LLMs) with powerful reasoning capabilities, offer new opportunities for recommendation systems (RS) , especially in understanding user preference. However, the direct application of LLMs to this task faces two major challenges: (1) limited context windows struggle to process the extensive user behavior sequences in real-world scenarios; (2) inherent hallucination effect can lead LLMs to infer spurious preferences that contradict true user intent, thereby degrading recommendation quality. To address this, we propose Rec 2, a Reinforcement learning-constrained segmented user modeling framework for recommendation. The framework segments lengthy behavior sequences into multiple segments and generates user preferences in a cascaded fashion. To suppress hallucinations and align with user intent, we introduce reinforcement learning, which uses the subsequent behavior segment as a supervisory signal to constrain and reward the preference inference process on the current segment, thereby generating more predictive and high-fidelity user preferences. Finally, the aligned LLMs infer preferences for each behavior segment; these are then processed by a specially-designed dynamic preference learning module to model preference evolution and are ultimately aggregated into a unified, dynamic long-term user preference embedding. This representation can be integrated into any recommendation model to boost its performance. Extensive experiments demonstrate that Rec2 significantly enhances recommendation performance by effectively capturing dynamic and authentic user preferences.

Jinming Wang, Hai Wang 0019, Hongkai Wen 0001, Geyong Min, Man Luo 0001

High-quality GPS trajectories are essential for location-based web services and smart city applications, including navigation, ride-sharing and delivery. However, due to low sampling rates and limited infrastructure coverage during data collection, real-world trajectories are often sparse and feature unevenly distributed location points. Recovering these trajectories into dense and continuous forms is essential but challenging, given their complex and irregular spatio-temporal patterns. In this paper, we introduce a novel diffusion model for TRA jectory rEC overy named TRACE, which reconstruct dense and continuous trajectories from sparse and incomplete inputs. At the core of TRACE, we propose a State Propagation Diffusion Model (SPDM), which integrates a novel memory mechanism, so that during the denoising process, TRACE can retain and leverage intermediate results from previous steps to effectively reconstruct those hard-to-recover trajectory segments. Extensive experiments on multiple real-world datasets show that TRACE outperforms the state-of-the-art, offering >26% accuracy improvement without significant inference overhead. Our work strengthens the foundation for mobile and web-connected location services, advancing the quality and fairness of data-driven urban applications. Code is available at:~ https://github.com/JinmingWang/TRACE

Rohan Bose, Jinwei Zhao, Tanya Shreedhar, Jianping Pan 0001, Nitinder Mohan

Low Earth Orbit (LEO) satellite ISPs promise universal Internet connectivity, yet their interaction with content delivery remains poorly understood. We present the first comprehensive measurement study decomposing Starlink's web content delivery performance decomposed across Point of Presence (PoP), DNS, and CDN layers. To quantify how satellite architecture disrupts terrestrial CDN assumptions, we conduct a measurement study spanning two years. We identify three distinct performance regimes based on infrastructure density. Regions with local content-rich PoPs achieve near-terrestrial latencies with the satellite segment dominating 80–90% of RTT. Infrastructure-sparse regions suffer cascading penalties: remote PoPs force distant resolver selection, which triggers CDN mislocalization, pushing latencies beyond 200 ms. Dense-infrastructure regions show minimal sensitivity to PoP changes. Leveraging Starlink's infrastructure expansion in early 2025 as a natural experiment, we demonstrate that relocating PoPs closer to user location reduces median page-fetch times by 60%. Our findings reveal that infrastructure proximity, not satellite coverage, influences web performance, requiring fundamental changes to CDN mapping and DNS resolution for satellite ISPs.

Jin Zeng 0001, Yupeng Qi, Hui Li 0057, Chengming Li 0004, Ziyu Lyu, Lixin Cui, Lu Bai 0001

Large language models (LLMs) are increasingly adopted as the backbone of recommender systems. However, user–item interactions in real-world scenarios are non-stationary, making preference drift over time inevitable. Existing model update strategies mainly rely on global fine-tuning or pointwise editing, but they face two fundamental challenges: (i) imbalanced update granularity, where global updates perturb behaviors unrelated to the target while pointwise edits fail to capture broader preference shifts; (ii) unstable incremental updates, where repeated edits interfere with prior adaptations, leading to catastrophic forgetting and inconsistent recommendations. To address these issues, we propose Region-Aware Incremental Editing (RAIE), a plug-in framework that freezes the backbone model and performs region-level updates. RAIE first constructs semantically coherent preference regions via spherical k-means in the representation space. It then assigns incoming sequences to regions via confidence-aware gating and performs three localized edit operations-Update, Expand, and Add-to dynamically revise the affected region. Each region is equipped with a dedicated Low-Rank Adaptation (LoRA) module, which is trained only on the region's updated data. During inference, RAIE routes each user sequence to its corresponding region and activates the region-specific adapter for prediction. Experiments on two benchmark datasets under a time-sliced protocol that segments data into Set-up (S), Finetune (F), and Test (T) show that RAIE significantly outperforms state-of-the-art baselines while effectively mitigating forgetting. These results demonstrate that region-aware editing offers an accurate and scalable mechanism for continual adaptation in dynamic recommendation scenarios.

Liang Chen, Xiaoding Wang 0001, Limei Lin, Dajin Wang, Zhiquan Liu 0001, Jie Wu 0001

The explosive growth of multimodal web data demands communication that transmits meaning rather than raw bits. Existing semantic-communication systems often fail under noise, missing modalities, and distribution shifts because they optimize surface features instead of modality-invariant knowledge. We present Grasp, a knowledge-centric framework for cross-modal communication. Grasp segments streams into semantic blocks and builds a graph over them; a lightweight Graph Neural Networks (GNN) produces schedulable, importance-weighted representations. At its core is knowledge purification : we minimize a conditional mutual information upper bound to perform a three-way disentanglement—strongly related, weakly related, and task-irrelevant components—so that only essential semantics are transmitted while non-essential factors are suppressed. To maintain synchrony, we introduce one-to-two temporal contrastive learning to achieve triple alignment of video, audio, and text despite sampling asynchrony. For efficient transmission, Grasp uses a cross-modal shared vector-quantization codebook—a discrete knowledge codebook —updated by multimodal attention. At the receiver, a soft-recovery mechanism leverages this shared knowledge to robustly reconstruct semantics under low signal-to-noise ratio (SNR) or missing modalities, yielding graceful degradation. Across web tasks—including cross-modal retrieval and missing-modality inference—Grasp improves knowledge consistency, semantic fidelity, and downstream performance over strong baselines while maintaining low latency. These results show that communication structured around purified knowledge is key to building robust, semantic-aware systems for the modern web.

Yijie Zhong 0001, Yunfan Gao, Haofen Wang

Long-term memory is critical for dialogue systems that support continuous, sustainable, and personalized interactions. However, existing methods rely on continuous summarization or OpenIE-based graph construction paired with fixed Top- k retrieval, leading to limited adaptability across query categories and high computational overhead. In this paper, we propose HingeMem, a boundary-guided long-term memory that operationalizes event segmentation theory to build an interpretable indexing interface via boundary-triggered hyperedges over four elements: person, time, location, and topic. When any such element changes, HingeMem draws a boundary and writes the current segment, thereby reducing redundant operations and preserving salient context. To enable robust and efficient retrieval under diverse information needs, HingeMem introduces query-adaptive retrieval mechanisms that jointly decide (a) what to retrieve: determine the query-conditioned routing over the element-indexed memory; (b) how much to retrieve: control the retrieval depth based on the estimated query type. Extensive experiments across LLM scales (from 0.6B to production-tier models; e.g., Qwen3-0.6B to Qwen-Flash) on LOCOMO show that HingeMem achieves approximately 20% relative improvement over strong baselines without query categories specification, while reducing computational cost (68%\downarrow question answering token cost compared to HippoRAG2). Beyond advancing memory modeling, HingeMem's adaptive retrieval makes it a strong fit for web applications requiring efficient and trustworthy memory over extended interactions.

Yujie Jin, Wenxin Zhang, Jingjing Wang, Guodong Zhou 0001

In the literature, prior research on Security-oriented Video Understanding (SVU) has predominantly focused on detecting and locating the threats (e.g., shootings, robberies) in videos, while largely lacking the effective capability to generate and evaluate the threat causes. Motivated by these gaps, this paper introduces a new chat paradigm SVU task, i.e., In-depth Security-oriented Video Understanding (DeepSVU), which aims to not only identify and locate the threats but also attribute and evaluate the causes of threatening segments in detail. Furthermore, this paper reveals two key challenges in the proposed task: 1) how to effectively model the coarse-to-fine physical-world information (e.g., human behavior, object interactions and background context) to boost the DeepSVU task, and 2) how to adaptively trade off these factors. Addressing these challenges is crucial for improving VAD, especially for identifying, locating, and attributing anomalies. To tackle these challenges, this paper proposes a new Unified Physical-world Regularized MoE (UPRM) approach. Specifically, UPRM incorporates two key components: the Unified Physical-world Enhanced MoE (UPE) Block and the Physical-world Trade-off Regularizer (PTR), to address the above two challenges, respectively. Extensive experiments conduct on our DeepSVU instructions datasets (i.e., UCF-C instructions and CUVA instructions) demonstrate that UPRM outperforms several advanced Video-LLMs as well as non-LLM approaches. such information.These justify the importance of the coarse-to-fine physical-world information in the DeepSVU task and demonstrate the effectiveness of our UPRM in capturing such information.

Pei Chen, Geng Hong, Xinyi Wu, Mengying Wu, Zixuan Zhu, Mingxuan Liu 0006, Baojun Liu 0002, Mi Zhang 0001, Min Yang 0002

The emergence of Large Language Model-enhanced Search Engines (LLMSEs) has revolutionized information retrieval by integrating web-scale search capabilities with AI-powered summarization. While these systems demonstrate improved efficiency over traditional search engines, their security implications against well-established black-hat Search Engine Optimization (SEO) attacks remain unexplored. In this paper, we present the first systematic study of SEO attacks targeting LLMSEs. Specifically, we examine ten representative LLMSE products (e.g., ChatGPT, Gemini) and construct SEO-Bench, a benchmark comprising 1,000 real-world black-hat SEO websites, to evaluate both open- and closed-source LLMSEs. Our measurements show that LLMSEs mitigate over 99.78% of traditional SEO attacks, with the phase of retrieval serving as the primary filter, intercepting the vast majority of malicious queries. We further propose and evaluate seven LLMSEO attack strategies, demonstrating that off-the-shelf LLMSEs are vulnerable to LLMSEO attacks, i.e., rewritten-query stuffing and segmented texts double the manipulation rate compared to the baseline. This work offers the first in-depth security analysis of the LLMSE ecosystem, providing practical insights for building more resilient AI-driven search systems. We have responsibly reported the identified issues to major vendors.

Tao Chen, Jingyi Zhang, Decheng Liu, Chunlei Peng

The proliferation of face forgery content on the Web poses a severe threat to online trust, social media security, and the credibility of digital information. Existing detection approaches often fail to generalize across diverse forgery types and unseen scenarios commonly encountered in web-scale applications. Recent studies have utilized visual large language models (VLMs) to answer not only ''Is this face a forgery?'' but also ''Why is the face a forgery?'' These studies introduced forgery-related attributes, such as forgery location and type, to construct deepfake VQA datasets and train VLMs, achieving high accuracy while providing human-understandable explanatory text descriptions. However, these methods still have limitations. For example, they do not fully leverage face quality-related attributes, which are often abnormal in forged faces, and they lack effective training strategies for forgery-aware VLMs. In this paper, we extend the VQA dataset to create DD-VQA+, which features a richer set of attributes and a more diverse range of samples. Furthermore, we introduce a novel forgery detection framework, MGFFD-VLM, which integrates an Attribute-Driven Hybrid LoRA Strategy to enhance the capabilities of Visual Large Language Models (VLMs). Additionally, our framework incorporates Multi-Granularity Prompt Learning and a Forgery-Aware Training Strategy. By transforming classification and forgery segmentation results into prompts, our method not only improves forgery classification but also enhances interpretability. To further boost detection performance, we design multiple forgery-related auxiliary losses. Experimental results demonstrate that our approach surpasses existing methods in both text-based forgery judgment and analysis, achieving superior accuracy.

Yu Yan, Siqi Lu, Yang Gao, Zhaoxuan Li, Ziming Zhao 0008, Qingjun Yuan, Yongjuan Wang

Large Language Models (LLMs), as common components of modern web application backends and online services, are being widely deployed across various web infrastructures in the .gguf single-file format. This trend exposes their model parameter space to an unprecedented hardware attack surface, such as Bit-Flip attacks (BFA). This paper is the first to systematically discover and validate the existence of single-bit vulnerabilities in LLMs weight files: In the .gguf quantization format of mainstream open-source models (such as DeepSeek, QWEN), flipping a single bit can induce three types of targeted semantic-level faults, respectively-Artificial Flawed Intelligence (outputting factual errors), Artificial Weak Intelligence (catastrophic model failure), and Artificial Bad Intelligence (generating harmful content). By building an information-theoretic weight sensitivity entropy model and a probabilistic heuristic scanning framework called BitSifter, we achieved efficient localization of critical vulnerable bits in models with hundreds of millions of parameters. Furthermore, an end-to-end remote BFA chain was designed, enabling semantic-level attacks in real-world web server deployment scenarios: At an attack frequency of 464.3 times per second, the average time required for the first successful flip of the target bit is 31.7 seconds, without requiring high-cost equipment or complex prompt engineering. This study reveals a critical finding: under relatively modest remote-attack conditions, requiring only conventional network connectivity, flipping a single vulnerable bit within the tensor data segment can cause models deployed in web service environments to autonomously generate extremely malicious responses, such as ''humans should be exterminated'', or produce naturally fluent and difficult-to-detect erroneous replies to ordinary user queries. This demonstrates a pervasive and exploitable security vulnerability in LLMs systems at the fundamental hardware level.

Jihao Zhao, Daixuan Li, Pengfei Li, Shuaishuai Zu, Biao Qin, Hongyan Liu 0002

The effectiveness upper bound of retrieval-augmented generation (RAG) is fundamentally constrained by the semantic integrity and information granularity of text chunks in its knowledge base. Moreover, domain documents are characterized by dense terminology and strong contextual dependencies, which exacerbate the semantic fragmentation of text chunks, thereby making it difficult to efficiently utilize their key information. To address these challenges, this paper proposes QChunker, which restructures the RAG paradigm from retrieval-augmentation to understanding-retrieval-augmentation. Firstly, QChunker models the text chunking as a composite task of text segmentation and knowledge completion to ensure the logical coherence and integrity of text chunks. Drawing inspiration from Hal Gregersen's ''Questions Are the Answer'' theory, we design a multi-agent debate framework comprising four specialized components: a question outline generator, text segmenter, integrity reviewer, and knowledge completer. This framework operates on the principle that questions serve as catalysts for profound insights. Through this pipeline, we successfully construct a high-quality dataset of 45K entries and transfer this capability to small language models. Additionally, to handle long evaluation chains and low efficiency in existing chunking evaluation methods, which overly rely on downstream QA tasks, we introduce a novel direct evaluation metric, ChunkScore. Both theoretical and experimental validations demonstrate that ChunkScore can directly and efficiently discriminate the quality of text chunks. Furthermore, during the text segmentation phase, we utilize document outlines for multi-path sampling to generate multiple candidate chunks and select the optimal solution employing ChunkScore. Extensive experimental results across four heterogeneous domains exhibit that QChunker effectively resolves aforementioned issues by providing RAG with more logically coherent and information-rich text chunks. Notably, this study also establishes a small-domain QA dataset concerning hazardous chemical safety, which fully reveals the significant value of RAG in specialized domains and the generalization capability of the QChunker framework.

Qishen Han, Grant Schoenebeck, Biaoshuai Tao, Lirong Xia

We investigate majority voting where agents possess private information about an unobservable ground truth that determines their preferences. In such settings, agents may hold different preferences and (collectively) engage in counterintuitive strategic behaviors. Previous work either assumes strategic behavior occurs without coordination or with unlimited coordination, yielding overly inclusive or exclusive predictions about voting outcomes. We incorporate coordination ability—the largest coalition size at which agents could strategically coordinate—into the analysis. Under the ex-ante Bayesian k-strong equilibrium framework, where no group of at most k agents can benefit from deviation, we provide closed-form characterizations of when informed majority decisions, the decision favored by the majority if the ground truth is common knowledge, are achievable. Specifically, we determine (1) when all k-strong equilibria reach the informed majority decision and (2) when at least one such equilibrium exists. These conditions depend on three factors: coordination ability, fraction of majority agents, and information structure. The boundary for the second question exhibits surprising complexity--non-continuous, non-linear, and segmental. Our results reveal the complicated landscape and provide refined predictions for strategic behavior across different coordination levels.