The spread of fake news online distorts public judgment and erodes trust in social media platforms. Although recent fake news detection (FND) models perform well in standard settings, they remain vulnerable to adversarial comments-authored by real users or by large language models (LLMs)-that subtly shift model decisions. In view of this, we first present a comprehensive evaluation of comment attacks to existing fake news detectors and then introduce a group-adaptive adversarial training strategy to improve the robustness of FND models. To be specific, our approach comprises three steps: (1) dividing adversarial comments into three psychologically grounded categories: perceptual, cognitive, and societal; (2) generating diverse, category-specific attacks via LLMs to enhance adversarial training; and (3) applying a Dirichlet-based adaptive sampling mechanism (InfoDirichlet Adjusting Mechanism) that dynamically adjusts the learning focus across different comment categories during training. Experiments on benchmark datasets show that our method maintains strong detection accuracy while substantially increasing robustness to a wide range of adversarial comment perturbations.

As large language models (LLMs) advance in role-playing (RP) tasks, existing benchmarks quickly become obsolete due to their narrow scope, outdated interaction paradigms, and limited adaptability across diverse application scenarios. To address this gap, we introduce FURINA-Builder, a novel multi-agent collaboration pipeline that automatically constructs fully customizable RP benchmarks at any scale. It enables evaluation of arbitrary characters across diverse scenarios and prompt formats, as the first benchmark builder in RP area for adaptable assessment. FURINA-Builder simulates dialogues between a test character and other characters drawn from a well-constructed character-scene pool, while an LLM judge selects fine-grained evaluation dimensions and adjusts the test character's responses into final test utterances. Using this pipeline, we build FURINA-Bench, a new comprehensive role-playing benchmark featuring both established and synthesized test characters, each assessed with dimension-specific evaluation criteria. Human evaluation and preliminary separability analysis justify our pipeline and benchmark design. We conduct extensive evaluations of cutting-edge LLMs and find that o3 and DeepSeek-R1 achieve the best performance on English and Chinese RP tasks, respectively. Across all models, established characters consistently outperform synthesized ones, with reasoning capabilities further amplifying this disparity. Interestingly, we observe that model scale does not monotonically reduce hallucinations. More critically, for reasoning LLMs, we uncover a novel trade-off: reasoning improves RP performance but simultaneously increases RP hallucinations. This trade-off extends to a broader Pareto frontier between RP performance and reliability for all LLMs. These findings demonstrate the effectiveness of FURINA-Builder and the challenge posed by FURINA-Bench.

Recent advancements in language agents have led to significant improvements in multi-hop reasoning tasks. However, existing approaches often struggle with handling open-domain problems, which require massive information retrieval due to their reliance on a fixed sequence of actions. To address this, we propose Feedback-Guided Dynamic Interactive Planning (FGDIP), a novel framework tailored to enhance reasoning in LLMs by utilizing dynamic and adaptive strategies for information exploration in open-domain multi-hop reasoning tasks. Our approach begins by identifying key entities relevant to the problem, which serve as the initial nodes in the reasoning process. From these initial nodes, we then generate reasoning child nodes with the process being refined through a combination of historical error analysis and real-time feedback, which allows the framework to dynamically adjust and optimize its reasoning strategies. By integrating depth-first search with an innovative node generation technique, our framework adapts based on both prior error paths and concurrently generated nodes at the same hierarchical level. This dynamic strategy effectively expands the search space while ensuring the reasoning process systematically converges toward accurate solutions. Experimental results show that FGDIP achieved up to 54.47% F1 score on the HotpotQA dataset and 70.05% on the StrategyQA dataset, surpassing the best baseline by 5.03% and 7.25% respectively, highlighting its versatility and potential to enhance language agents in multi-hop reasoning tasks.

Unsupervised analysis of text corpora is challenging, especially in data-scarce domains where traditional topic models struggle. While these models offer a solution, they typically describe clusters with lists of keywords that require significant manual effort to interpret and often lack semantic coherence. To address this critical interpretability gap, we introduce Recursive Thematic Partitioning (RTP), a novel framework that leverages Large Language Models (LLMs) to interactively build a binary tree. Each node in the tree is a natural language question that semantically partitions the data, resulting in a fully interpretable taxonomy where the logic of each cluster is explicit. Our experiments demonstrate that RTP's question-driven hierarchy is more interpretable than the keyword-based topics from a strong baseline like BERTopic. Furthermore, we establish the quantitative utility of these clusters by showing they serve as powerful features in downstream classification tasks, particularly when the data's underlying themes correlate with the task labels. RTP introduces a new paradigm for data exploration, shifting the focus from statistical pattern discovery to knowledge-driven thematic analysis. Furthermore, we demonstrate that the thematic paths from the RTP tree can serve as structured, controllable prompts for generative models. This transforms our analytical framework into a powerful tool for synthesis, enabling the consistent imitation of specific characteristics discovered in the source corpus.

Recent advances in reinforcement learning (RL) have significantly enhanced the agentic capabilities of large language models (LLMs). In long-term and multi-turn agent tasks, existing approaches driven solely by outcome rewards often suffer from the problem of sparse supervision. To address the challenge, we propose Tree-based Group Relative Policy Optimization (Tree-GRPO), a grouped agent RL method based on tree search, where each tree node represents the complete agent interaction step. By sharing common prefixes, the tree search sampling increases the number of rollouts achievable within a fixed budget of tokens or tool calls. Moreover, we find that the tree-structured trajectory naturally allows the construction of step-wise process supervised signals even using only the outcome reward. Based on this, Tree-GRPO estimates the grouped relative advantages both on intra-tree and inter-tree levels. Through theoretical analysis, we demonstrate that the objective of intra-tree level group relative policy optimization is equivalent to that of step-level direct preference learning. Experiments across 11 datasets and 3 types of QA tasks demonstrate the superiority of the proposed tree-based RL over the chain-based RL method.Figure 1: Comparison of chain-based and tree-based sampling strategies in LLM multi-turn agent RL. The tree structure brings two major advantages: (i) less rollout budget (both on tokens and tool-calls); (ii) higher performance. Reinforcement Learning (RL) has emerged as a pivotal post-training paradigm for Large Language Models (LLMs), catalyzing the development of several frontier models (DeepSeek-AI Team, 2025; Y ang et al., 2025a; OpenAI, 2024). RL-tuned LLMs trained only with outcome rewards acquire complex reasoning abilities and achieve remarkable gains in single-turn tasks, such as mathematical proof and code generation (Team et al., 2025b; Y u et al., 2025; Chu et al., 2025a; Shao et al., 2024; Xin et al., 2024). This suggests that LLMs can learn not only through static imitation, but also by actively interacting with dynamic environments. Guided by this prospect, recent works have extended this RL paradigm to more complex agent settings involving dynamic, multi-turn interactions (Feng et al., 2025b; Singh et al., 2025; Wang et al., 2025b; Qian et al., 2025; Feng et al., Work done during internship at AMAP, Alibaba Group. Right (Ours): Tree search with nodes corresponding to complete agent step.

We present MA-RAG, a Multi-Agent framework for Retrieval-Augmented Generation (RAG) that addresses the inherent ambiguities and reasoning challenges in complex information-seeking tasks. Unlike conventional RAG methods that rely on end-to-end fine-tuning or isolated component enhancements, MA-RAG orchestrates a collaborative set of specialized AI agents: Planner, Step Definer, Extractor, and QA Agents, each responsible for a distinct stage of the RAG pipeline. By decomposing tasks into subtasks such as query disambiguation, evidence extraction, and answer synthesis, and enabling agents to communicate intermediate reasoning via chain-of-thought prompting, MA-RAG progressively refines retrieval and synthesis while maintaining modular interpretability. Extensive experiments on multi-hop and ambiguous QA benchmarks, including NQ, HotpotQA, 2WikimQA, and TriviaQA, demonstrate that MA-RAG significantly outperforms standalone LLMs and existing RAG methods across all model scales. Notably, even a small LLaMA3-8B model equipped with MA-RAG surpasses larger standalone LLMs, while larger variants (LLaMA3-70B and GPT-4o-mini) set new state-of-the-art results on challenging multi-hop datasets. Ablation studies reveal that both the planner and extractor agents are critical for multi-hop reasoning, and that high-capacity models are especially important for the QA agent to synthesize answers effectively. Beyond general-domain QA, MA-RAG generalizes to specialized domains such as medical QA, achieving competitive performance against domain-specific models without any domain-specific fine-tuning. Our results highlight the effectiveness of collaborative, modular reasoning in retrieval-augmented systems: MA-RAG not only improves answer accuracy and robustness but also provides interpretable intermediate reasoning steps, establishing a new paradigm for efficient and reliable multi-agent RAG.

Retrieval-Augmented Generation (RAG) leverages large language models (LLMs) combined with external contexts to enhance the accuracy and reliability of generated responses. However, reliably attributing generated content to specific context segments, context attribution, remains challenging due to the computationally intensive nature of current methods, which often require extensive fine-tuning or human annotation. In this work, we introduce a novel Jensen-Shannon Divergence driven method to Attribute Response to Context (ARC-JSD), enabling efficient and accurate identification of essential context sentences without additional fine-tuning, gradient-calculation or surrogate modelling. Evaluations on a wide range of RAG benchmarks, such as TyDi QA, Hotpot QA, and Musique, using instruction-tuned LLMs in different scales demonstrate superior accuracy and significant computational efficiency improvements compared to the previous surrogate-based method. Furthermore, our mechanistic analysis reveals specific attention heads and multilayer perceptron (MLP) layers responsible for context attribution, providing valuable insights into the internal workings of RAG models and how they affect RAG behaviours. Our code is available at https://github.com/ruizheliUOA/ARC_JSD.

Matrix factorization is a widely used approach for top-N recommendation and collaborative filtering. When implemented on implicit feedback data (such as clicks), a common heuristic is to upweight the observed interactions. This strategy has been shown to improve performance for certain algorithms. In this paper, we conduct a systematic study of various weighting schemes and matrix factorization algorithms. Somewhat surprisingly, we find that training with unweighted data can perform comparably to, and sometimes outperform, training with weighted data, especially for large models. This observation challenges the conventional wisdom. Nevertheless, we identify cases where weighting can be beneficial, particularly for models with lower capacity and specific regularization schemes. We also derive efficient algorithms for exactly minimizing several weighted objectives that were previously considered computationally intractable. Our work provides a comprehensive analysis of the interplay between weighting, regularization, and model capacity in matrix factorization for recommender systems.

Equity confirmed it was supporting a Scottish actor who believes her image was used in the creation of Tilly Norwood (above), an AI-generated'actor'. Equity confirmed it was supporting a Scottish actor who believes her image was used in the creation of Tilly Norwood (above), an AI-generated'actor'. Equity threatens mass direct action over use of actors' images in AI content The performing arts union Equity has threatened mass direct action over tech and entertainment companies' use of its members' likenesses, images and voices in AI content without permission. Its general secretary, Paul W Fleming, said it planned to coordinate data requests en masse to companies to force them to disclose whether they used members' data in AI-generated material without consent. Last week the union confirmed its was supporting a Scottish actor who believes her image was used in the creation of the "AI actor" Tilly Norwood, which has been widely condemned by the film industry.

Natural language-based user profiles in recommender systems have been explored for their interpretability and potential to help users scrutinize and refine their interests, thereby improving recommendation quality. Building on this foundation, we introduce a human-AI collaborative profile for a movie recommender system that presents editable personalized interest summaries of a user's movie history. Unlike static profiles, this design invites users to directly inspect, modify, and reflect on the system's inferences. In an eight-week online field deployment with 1775 active movie recommender users, we find persistent gaps between user-perceived and system-inferred interests, show how the profile encourages engagement and reflection, and identify design directions for leveraging imperfect AI-powered user profiles to stimulate more user intervention and build more transparent and trustworthy recommender experiences.