论文检索
输入标题、作者或关键词，从 1,444 篇学术成果中精准定位
Web-based platforms are becoming a primary channel for psychological support, yet most LLM-driven chatbots remain opaque, single-stage, and weakly grounded in established therapeutic practice. To address this gap, we present XInsight, a multi-agent framework that models psychological support as a stage-consistent workflow aligned with the classical Exploration-Insight-Action paradigm. Building on structured client representations, XInsight orchestrates specialized agents under a unified Reason-Intervene-Reflect cycle: an Exploration agent organizes background and concerns into a structured Case Conceptualization Form, a Routing agent performs Adaptive Therapeutic Routing (ATR) across SFBT, CBT, and MBCT, a unified Therapeutic agent executes school-consistent submodules, and a Consolidation agent guides review, skill integration, and relapse-prevention planning. A Recording agent continuously transforms open-ended web dialogues into standardized psychological artifacts, enhancing interpretability, continuity, and accountability. To support transparent assessment, we introduce XInsight-Bench with a Scale-Guided LLM Evaluation (SGLE) protocol that combines therapy-specific clinical scales with general counseling criteria. Experiments show improved paradigm alignment, multi-therapy integration, interaction depth, and interpretability over existing multi-agent counseling systems, indicating that XInsight provides a practical blueprint for integrating counseling-inspired support agents into web applications for digital well-being.
The shortage of mental health professionals has driven the web to become a primary avenue for accessible psychological support. While Large Language Models (LLMs) offer promise for scalable web-based counseling, existing approaches often lack emotional understanding, adaptive strategies, and long-term memory. These limitations pose risks to digital well-being, as disjointed interactions can fail to support vulnerable users effectively. To address these gaps, we introduce TheraMind, a strategic and adaptive agent designed for trustworthy online longitudinal counseling. The cornerstone of TheraMind is a novel dual-loop architecture that decouples the complex counseling process into an Intra-Session Loop for tactical dialogue management and a Cross-Session Loop for strategic therapeutic planning. The Intra-Session Loop perceives the patient's emotional state to dynamically select response strategies while leveraging cross-session memory to ensure continuity. Crucially, the Cross-Session Loop empowers the agent with long-term adaptability by evaluating the efficacy of the applied therapy after each session and adjusting the method for subsequent interactions. We validate our approach in a high-fidelity simulation environment grounded in real clinical cases. Extensive evaluations show that TheraMind outperforms other methods, especially on multi-session metrics like Coherence, Flexibility, and Therapeutic Attunement, validating the effectiveness of its dual-loop design in emulating strategic, adaptive, and longitudinal therapeutic behavior. The code is publicly available at https://github.com/Emo-gml/TheraMind.
The vision of an inclusive World Wide Web is impeded by a severe linguistic divide, particularly for communities in low-resource regions of Southeast Asia. While large language models (LLMs) offer a potential solution for translation, their deployment in data-poor contexts faces a dual challenge: the scarcity of high-quality, culturally relevant data and the prohibitive energy costs of training on massive, noisy web corpora. To resolve the tension between digital inclusion and environmental sustainability, we introduce Sustainable Agent-Guided Expert-tuning (SAGE). This framework pioneers an energy-aware paradigm that prioritizes the ''right data'' over ''big data''. Instead of carbon-intensive training on unfiltered datasets, SAGE employs a reinforcement learning (RL) agent, optimized via Group Relative Policy Optimization (GRPO), to autonomously curate a compact training set. The agent utilizes a semantic reward signal derived from a small, expert-constructed set of community dialogues to filter out noise and cultural misalignment. We then efficiently fine-tune open-source LLMs on this curated data using Low-Rank Adaptation (LoRA). We applied SAGE to translation tasks between English and seven low-resource languages (LRLs) in Southeast Asia. Our approach establishes new state-of-the-art performance on BLEU-4 and COMET-22 metrics, effectively capturing local linguistic nuances. Crucially, SAGE surpasses baselines trained on full datasets while reducing data usage by 97.1% and training energy consumption by 95.2%. By delivering high-performance models with a minimal environmental footprint, SAGE offers a scalable and responsible pathway to bridge the digital divide in the Global South.
Mental manipulation on social media poses a covert yet serious threat to individuals' psychological well-being and the integrity of online interactions. Detecting such behavior is challenging due to the difficult-to-annotate training data, its highly covert and multi-turn nature, and the lack of real-world datasets. To address these challenges, we propose MentalMAD, a framework that enhances large language models for mental manipulation detection. Our approach consists of three key components: EvoSA, an annotation-free data augmentation method that combines evolutionary operations with speech-act-aware prompting; teacher-model-generated complementary-task supervision; and Complementary-Convergent Distillation, a phase-wise strategy for transferring manipulation-specific knowledge to student models. We then constructed the ReaMent dataset, comprising 5,000 real-world-sourced dialogues. Extensive experiments show that MentalMAD improves accuracy by 14.0%, macro-F1 by 27.3%, and weighted F1 by 15.1% over the strongest baseline. The code and the dataset are publicly available at https://github.com/Yuansheng-Gao/MentalMAD.
Mixture-of-Experts (MoE) models are central to scaling Large Language Models (LLMs), but stateless and compute-intensive routing repeatedly re-explores expert assignments for similar inputs, causing computational redundancy and unstable behavior in web-scale applications like search and dialogue. We reframe expert routing as a retrieval-augmented process and propose the Retrieval-Memory Synergy Mixture-of-Experts (RMS-MoE), which integrates a Co-Activation Memory (CAM) to store and retrieve effective expert teams and a learnable, input-dependent gate to fuse retrieved priors with live routing predictions, enabling consistent expert coordination for semantically related inputs. Extensive experiments on web-scale QA and dialogue tasks show that RMS-MoE achieves a 26% latency reduction, a 2.7-point accuracy gain, and a 3.3% improvement in routing stability, demonstrating architectural memory as a principled path toward more efficient, stable, and scalable LLMs.
Customer service automation has seen growing demand within digital transformation. Existing approaches either rely on modular system designs with extensive agent orchestration or employ over-simplified instruction schemas, providing limited guidance and poor generalizability. This paper introduces an orchestration-free framework using Task-Oriented Flowcharts (TOFs) to enable end-to-end automation without manual intervention. We first define the components and evaluation metrics for TOFs, then formalize a cost-efficient flowchart construction algorithm to abstract procedural knowledge from service dialogues. We emphasize local deployment of small language models and propose decentralized distillation with flowcharts to mitigate data scarcity and privacy issues in model training. Extensive experiments validate the effectiveness in various service tasks, with superior quantitative and application performance compared to strong baselines and market products. By releasing a web-based system demonstration with case studies, we aim to promote streamlined creation of future service automation.
Traditional search engines return uniform results for identical queries, overlooking users' personalized intents. While personalized search has been extensively studied, research on query rewriting for personalized intents has been constrained by traditional approaches like statistical co-occurrence and synonym expansion. Current work primarily addresses multi-turn dialogue scenarios in Conversational AI rather than exploring applications in large-scale search engines. This typically stems from the absence of real search scenario data and the difficulty of inferring users' intents through personalized reasoning. While Large Language Models (LLMs) combined with Chain-of-Thought (CoT) capabilities provide possibilities for personalized reasoning, CoT introduces additional reasoning overhead that is difficult to accept in online scenarios requiring low latency. To address this, this paper proposes PicQue (Personalized Efficient Query Rewrite), a personalized query rewriting model training pipeline aimed at achieving high accuracy with low latency. PicQue contains a two-stage training that first employs a novel Hybrid Supervised Fine-Tuning strategy to retain the model's reasoning capabilities while allowing the decoding process to skip CoT, thereby obtaining CoT's accuracy gains without increasing latency. Building on this foundation, PicQue conducts second-stage reinforcement learning using Group Relative Policy Optimization (GRPO) to further improve rewriting accuracy and reduce the risk of over-rewriting. We also propose a Guided Search strategy to optimize GRPO training, alleviating the reduction in training sample utilization when all sampling rollouts are wrong. Extensive offline and online experiments demonstrate PicQue's effectiveness. In offline metrics, PicQue achieves over 7% improvement in rewriting accuracy compared to baseline methods and compresses up to 95% of decoding tokens. In online A/B tests, user satisfaction increases by 1.78% and query change rate decreases by 0.71%, achieving significant online gains.
Although Large Language Models (LLMs) have demonstrated significant proficiency in code generation, their monolithic and correlation-driven nature renders them susceptible to systematic cognitive biases, deficient counterfactual reasoning, and adversarial manipulation—a characteristic we term cognitive vulnerability. Such vulnerabilities compromise the reliability and security of AI-assisted software development, potentially resulting in code that is not only functionally incorrect but also biased, insecure, and difficult to validate using conventional testing paradigms. While recent multi-agent systems enhance workflow efficiency via task decomposition, they do not fundamentally address these reasoning deficits. This highlights the need for a framework capable of proactively identifying and mitigating the cognitive flaws of an LLM during the reasoning process. To address this challenge, we introduce CodeForge, a multi-agent adversarial reasoning framework that reframes code generation as a cognitive crucible. This process involves a structured debate among three specialized agents—an Optimist, a Pragmatist, and an Adversarial Skeptic—who iteratively cross-examine and refine solution plans. Following convergence, an adversarial verification module systematically generates counterfactual perturbations to stress-test and enhance the final plan. Comprehensive evaluations on the HumanEval and MBPP benchmarks demonstrate that CodeForge significantly outperforms state-of-the-art methods, achieving a pass@1 of 97.3% with GPT-4. Ablation studies confirm the necessity of both the adversarial dialogue and counterfactual verification components. This work represents a shift from passive debugging to proactive cognitive hardening, establishing a pathway toward more trustworthy automated software engineering.
The prevalence of sarcasm in multimodal dialogues on the social platforms presents a crucial yet challenging task for understanding the true intent behind online content. Comprehensive sarcasm analysis requires two key aspects: Multimodal Sarcasm Detection (MSD) and Multimodal Sarcasm Explanation (MuSE). Intuitively, the act of detection is the result of the reasoning process that explains the sarcasm. Current research predominantly focuses on addressing either MSD or MuSE as a single task. Even though some recent work has attempted to integrate these tasks, their inherent causal dependency is often overlooked. To bridge this gap, we propose MuVaC, a variational causal inference framework that mimics human cognitive mechanisms for understanding sarcasm, enabling robust multimodal feature learning to jointly optimize MSD and MuSE. Specifically, we first model MSD and MuSE from the perspective of structural causal models, establishing variational causal pathways to define the objectives for joint optimization. Next, we design an alignment-then-fusion approach to integrate multimodal features, providing robust fusion representations for sarcasm detection and explanation generation. Finally, we enhance the reasoning trustworthiness by ensuring consistency between detection results and explanations. Experimental results demonstrate the superiority of MuVaC in public datasets, offering a new perspective for understanding multimodal sarcasm.
Multimodal dialogue emotion recognition seeks to integrate information from text, audio, and video to accurately determine the emotional state of each utterance. However, prevailing approaches often depend on fixed fusion strategies, failing to account for quality variations across modalities. Consequently, noise from less informative modalities can degrade overall performance. Moreover, most models treat utterances as independent units for static classification, overlooking the dynamic evolution of emotions throughout a dialogue. To overcome these limitations, we propose MIDE, a novel approach that leverages mutual information enhancement and dynamic modality slection for multimodal emotion recognition. Our approach dynamically selects high-quality modality pairs for fusion, minimizing interference from noisy or redundant sources. Specifically, it employs mutual information maximization to achieve cross-modal semantic alignment and incorporates an autonomous modality selection mechanism to assess inter-modal compatibility. Furthermore, an emotion transition prediction module, implemented with a Gated Recurrent Unit (GRU), captures temporal emotional dependencies, enabling joint optimization of static emotion classification and dynamic emotion trajectory prediction. Extensive experiments on the IEMOCAP and MELD datasets demonstrate that MIDE significantly surpasses existing models in accuracy and robustness, highlighting its strength in adaptive fusion for complex dialogue scenarios.
Large language models have given rise to interactive recommendation agents (IRAs). Through proactive clarification, tool invocation, and dynamic dialogue, IRAs shift recommender systems from passive prediction to interactive, proactive intelligence. For training IRAs, agentic reinforcement learning offers a natural pathway, as it enables models to learn interactive capabilities directly from environmental feedback without requiring costly annotated data. However, this process faces three key challenges: credit assignment in multi-turn interactions, efficient exploration in large action spaces, and coordinated learning of multiple interactive skills. To tackle this challenge, we present a new preference optimization paradigm, GIMO, which treats the interaction and learning process of IRAs in sparse environments as the continuous fulfillment and stimulation of three intrinsic drives: Autonomy, Competence, and Relatedness. Instead of relying solely on external rewards, GIMO estimates the potential between adjacent states in a generative manner to construct intrinsic rewards. Such a generative motivation structure not only enables fine-grained credit assignment, but also naturally inspires a hint-guided action proposal mechanism that facilitates efficient exploration. Furthermore, during the multi-skill coordination training stage, we introduce an explicit KL regularization into policy orchestration to prevent global policy collapse. Theoretically, we prove that GIMO guarantees global policy consistency, and empirically, experiments in interactive recommendation environments built from three datasets demonstrate that GIMO consistently outperforms existing methods. Our code is anonymously available at https://github.com/XueyangFeng/GIMO.
Conversational recommender systems (CRSs) aim to provide real-time recommendations through dynamic interactions between users and the system. Recent studies have revealed the value of personalized information derived from users' historical dialogue records in refining user preferences. However, existing methods often utilize the entire historical dialogue of a user indiscriminately, leading to the issue of cognitive negative transfer, wherein historical dialogue sessions impede rather than facilitate current decision-making. This ultimately degrades the performance of conversational recommendations. To address this issue, inspired by the behavioral decision theory, this paper proposes a novel model, Personalization-Driven Agent Debate for Conversational Recommendation, named EyeCRS. EyeCRS is comprised of two core components: (i) The Multi-Agent Debate Module employs supporting and opposing agents to simulate a human-like debate process, thereby enabling a rigorous evaluation of whether historical dialogue sessions may introduce cognitive negative transfer into the recommender. (ii) The Parametric Injection Module, subsequently, parameterizes the historical dialogues that are validated as positive and injects them into the intrinsic parameter space of the downstream recommender, thereby effectively enhancing the model's ability to internalize and utilize personalized knowledge. Experimental results on two real-world datasets show that EyeCRS effectively mitigates cognitive negative transfer and achieves superior performance in conversational recommendation.
Large Language Models (LLMs) have shown strong potential as conversational agents. Yet, their effectiveness remains limited by deficiencies in robust long-term memory—particularly in complex, long-term Web-based services such as online emotional support. However, existing long-term dialogue benchmarks primarily focus on static and explicit fact retrieval, failing to evaluate agents in these critical scenarios where user information is dispersed, implicit, and continuously evolving. To address this gap, we introduce ES-MemEval, a comprehensive benchmark that systematically evaluates five core memory capabilities—information extraction, temporal reasoning, conflict detection, abstention, and user modeling—in long-term emotional support scenarios, covering question answering, summarization, and dialogue generation tasks. To support the benchmark, we also propose EvoEmo, the first multi-session dataset for personalized long-term emotional support scenarios, capturing fragmented, implicit user disclosures and evolving user states. Extensive experiments on open-source long-context, commercial, and retrieval-augmented (RAG) LLMs reveal that explicit long-term memory is essential to reduce hallucinations and enable effective personalization. At the same time, RAG enhances factual consistency but struggles with temporal dynamics and evolving user states. These findings highlight both the potential and limitations of current paradigms, encouraging the development of more robust memory–retrieval integration in long-term personalized dialogue systems.
We present S-Path-RAG, a semantic-aware shortest-path Retrieval-Augmented Generation framework designed to improve multi-hop question answering over large knowledge graphs. S-Path-RAG departs from one-shot, text-heavy retrieval by enumerating bounded-length, semantically weighted candidate paths using a hybrid weighted k-shortest, beam, and constrained random-walk strategy, learning a differentiable path scorer together with a contrastive path encoder and lightweight verifier, and injecting a compact soft mixture of selected path latents into a language model via cross-attention. The system runs inside an iterative Neural-Socratic Graph Dialogue loop in which concise diagnostic messages produced by the language model are mapped to targeted graph edits or seed expansions, enabling adaptive retrieval when the model expresses uncertainty. This combination yields a retrieval mechanism that is both token-efficient and topology-aware while preserving interpretable path-level traces for diagnostics and intervention. We validate S-Path-RAG on standard multi-hop KGQA benchmarks and through ablations and diagnostic analyses. The results demonstrate consistent improvements in answer accuracy, evidence coverage, and end-to-end efficiency compared to strong graph- and LLM-based baselines. We further analyze trade-offs between semantic weighting, verifier filtering, and iterative updates, and report practical recommendations for deployment under constrained compute and token budgets.
LLM-based empathetic dialogue systems enhance agents' emotional support capabilities. Previous approaches primarily relied on Chain-of-Thought (CoT) prompting to extract key dialogue cues and further strengthened the agent's sensitivity to these signals through supervised fine-tuning. However, such methods overly depend on the information extraction capability of LLMs, leading to unstable reasoning and limited interpretability. To simultaneously improve an agent's ability to proactively explore solutions through rational reasoning while attending to users' sensitive emotions via empathetic understanding, we propose Neuro-Sym Supporter, a hybrid decision-making emotional support agent that integrates symbolic reasoning with deep learning. This model combines rational inference with emotional empathy, enabling the agent to generate supportive responses that balance logic and emotion. Specifically, we introduce Sym-Mind, a differentiable logic-based reasoning framework for emotional support strategy selection, which unifies interpretability with stable performance. Experimental results on public datasets demonstrate that our approach consistently outperforms multiple competitive baselines in both automatic and human evaluations, validating its effectiveness.
While Large Language Models (LLMs) are increasingly deployed in Web applications such as search, dialogue, and recommendation systems, their reliance on large-scale Web data raises serious privacy concerns, particularly the risk of memorizing sensitive content. Existing Membership Inference Attacks (MIA) rely heavily on the surface form of inputs, rendering them ineffective against semantically preserved but structurally altered samples. This methodological weakness results in widespread false negatives and compromises the integrity of privacy evaluations in large-scale Web corpora. To address this limitation, we propose Adversarial Semantic Membership Inference Attack (ASMIA). ASMIA enhances MIA effectiveness by generating semantically diverse adversarial samples, extracting multi-layer attention features from the target model, and training a contrastive classifier that leverages similarity metrics and logarithmic probabilities to distinguish members from non-members. Experiments on LLMs trained with Wikipedia, a representative large-scale Web corpus, demonstrate that ASMIA significantly outperforms existing methods, highlighting the value of semantic perturbations and attention patterns in detecting training data leakage.
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.