Diffusion language models (DLMs) have recently emerged as a compelling alternative to autoregressive generation, offering parallel generation and improved global coherence. During inference, DLMs generate text by iteratively denoising masked sequences in parallel; however, determining which positions to unmask and which tokens to commit forms a large combinatorial search problem. Existing inference methods approximate this search using heuristics, which often yield suboptimal decoding paths; other approaches instead rely on additional training to guide token selection. To introduce a principled search mechanism for DLMs inference, we introduce MEDAL, an inference-time scaling framework that integrates Monte Carlo Tree SEarch initialization for Diffusion LAnguage Model inference. We employ Monte Carlo Tree Search at the initialization stage to explore promising unmasking trajectories, providing a robust starting point for subsequent refinement. This design enables efficient inference-time scaling, allowing generation quality to improve as the search budget increases, without additional training. Across multiple benchmarks, MEDAL achieves up to 22.0% improvement over existing inference strategies, establishing a new paradigm for search-based inference in DLMs.

With the rapid development of large language models (LLMs), role-playing dialogue systems have become increasingly popular for immersive and personalized human–AI interaction. However, in real-world role-playing scenarios, users often provide lazy queries—brief and ambiguous responses such as “oh,” “hmm,” or “?”—which lack explicit intent and emotional cues, making it difficult for role-playing LLMs to generate engaging replies. Existing query rewriting methods mainly target functional tasks such as information retrieval and are not tailored to role-playing dialogue, where rewrite quality depends on role-playing-specific factors such as character consistency and personalized user preferences. To address this, we first propose a listwise Generative Reward Model (GRM) trained from real user preference data collected through online regeneration behavior, which provides fine-grained process and final preference signals for training a Lazy Query Rewriter. Using rejection fine-tuning and reinforcement learning, the LQ-Rewriter learns to transform underspecified user inputs into richer expressions while preserving user intent and dialogue context. Experiments on real-world dialogue logs from a mainstream Chinese AI-powered romantic role-playing platform show that our GRM substantially outperforms general-purpose LLM judges. Moreover, our LQ-Rewriter achieves the highest GRM-based and human preference win rates among all rewriting baselines, and brings tangible improvements in online deployment.

Mis- and disinformation online have become a major societal problem as major sources of online harms of different kinds. One common form of mis- and disinformation is out-of-context (OOC) information, where different pieces of information are falsely associated, e.g., a real image combined with a false textual caption or a misleading textual description. Although some past studies have attempted to defend against OOC mis- and disinformation through external evidence, they tend to disregard the role of different pieces of evidence with different stances. Motivated by the intuition that the stance of evidence represents a bias towards different detection results, we propose a stance extraction network (SEN) that can extract the stances of different pieces of multi-modal evidence in a unified framework. Moreover, we introduce a support-refutation score calculated based on the co-occurrence relations of named entities into the textual SEN. Extensive experiments on a public large-scale dataset demonstrated that our proposed method outperformed the state-of-the-art baselines, with the best model achieving a performance gain of 3.2% in accuracy.

Information extraction from documents such as receipts or invoices is a fundamental and crucial step for office automation. Many approaches focus on extracting entities and relationships from plain texts, however, when it comes to document images, such demand becomes quite challenging since visual and layout information are also of great significance to help tackle this problem. In this work, we propose the attention-based graph neural network to combine textual and visual information from document images. Moreover, the global node is introduced in our graph construction algorithm which is used as a virtual hub to collect the information from all the nodes and edges to help improve the performance. Extensive experiments on real-world datasets show that our method outperforms baseline methods by significant margins.