Unified vision large language models (VLLMs) have recently achieved impressive advancements in both multimodal understanding and generation, powering applications such as visual question answering and text-guided image synthesis. However, progress in unified VLLMs remains constrained by the lack of datasets that fully exploit the synergistic potential between these two core abilities. Existing datasets typically address understanding and generation in isolation, thereby limiting the performance of unified VLLMs. To bridge this critical gap, we introduce a novel dataset construction framework, UnifiedVisual, and present UnifiedVisual-240K, a high-quality dataset meticulously designed to facilitate mutual enhancement between multimodal understanding and generation. UnifiedVisual-240K seamlessly integrates diverse visual and textual inputs and outputs, enabling comprehensive cross-modal reasoning and precise text-to-image alignment. Our dataset encompasses a wide spectrum of tasks and data sources, ensuring rich diversity and addressing key shortcomings of prior resources. Extensive experiments demonstrate that models trained on UnifiedVisual-240K consistently achieve strong performance across a wide range of tasks. Notably, these models exhibit significant mutual reinforcement between multimodal understanding and generation, further validating the effectiveness of our framework and dataset. We believe UnifiedVisual represents a new growth point for advancing unified VLLMs and unlocking their full potential. Our code and datasets is available at https://github.com/fnlp-vision/UnifiedVisual.

Abstract-- The emerging field of Vision-Language-Action (VLA) for humanoid robots faces several fundamental challenges, including the high cost of data acquisition, the lack of a standardized benchmark, and the significant gap between simulation and the real world. T o overcome these obstacles, we propose RealMirror, a comprehensive, open-source embodied AI VLA platform. RealMirror builds an efficient, low-cost data collection, model training, and inference system that enables end-to-end VLA research without requiring a real robot. T o facilitate model evolution and fair comparison, we also introduce a dedicated VLA benchmark for humanoid robots, featuring multiple scenarios, extensive trajectories, and various VLA models. Jun Xiong is with The Chinese University of Hong Kong, Shenzhen, China. In conclusion, with the unification of these critical components, RealMirror provides a robust framework that significantly accelerates the development of VLA models for humanoid robots. I. INTRODUCTION The rapid evolution of Large Language Models (LLMs) like GPT [1], Qwen [2], and Deepseek [3] has significantly advanced the development of Artificial General Intelligence (AGI). While exhibiting remarkable model performance, they lack the ability to perform tasks in the real world.

We formulate an attention mechanism for continuous and ordered sequences that explicitly functions as an alignment model, which serves as the core of many sequence-to-sequence tasks. Standard scaled dot-product attention relies on positional encodings and masks but does not enforce continuity or monotonicity, which are crucial for frame-synchronous targets. We propose learned nonnegative \emph{clocks} to source and target and model attention as the meeting probability of these clocks; a path-integral derivation yields a closed-form, Gaussian-like scoring rule with an intrinsic bias toward causal, smooth, near-diagonal alignments, without external positional regularizers. The framework supports two complementary regimes: normalized clocks for parallel decoding when a global length is available, and unnormalized clocks for autoregressive decoding -- both nearly-parameter-free, drop-in replacements. In a Transformer text-to-speech testbed, this construction produces more stable alignments and improved robustness to global time-scaling while matching or improving accuracy over scaled dot-product baselines. We hypothesize applicability to other continuous targets, including video and temporal signal modeling.

Modeling semantic and structural information from tabular data remains a core challenge for effective table understanding. Existing Table-as-Text approaches flatten tables for large language models (LLMs), but lose crucial structural cues, while Table-as-Image methods preserve structure yet struggle with fine-grained semantics. Recent Table-as-Multimodality strategies attempt to combine textual and visual views, but they (1) statically process both modalities for every query-table pair within a large multimodal LLMs (MLLMs), inevitably introducing redundancy and even conflicts, and (2) depend on costly fine-tuning of MLLMs. In light of this, we propose TableDART, a training-efficient framework that integrates multimodal views by reusing pretrained single-modality models. TableDART introduces a lightweight 2.59M-parameter MLP gating network that dynamically selects the optimal path (either Text-only, Image-only, or Fusion) for each table-query pair, effectively reducing redundancy and conflicts from both modalities. In addition, we propose a novel agent to mediate cross-modal knowledge integration by analyzing outputs from text- and image-based models, either selecting the best result or synthesizing a new answer through reasoning. This design avoids the prohibitive costs of full MLLM fine-tuning. Extensive experiments on seven benchmarks show that TableDART establishes new state-of-the-art performance among open-source models, surpassing the strongest baseline by an average of 4.02%. The code is available at: https://anonymous.4open.science/r/TableDART-C52B

Spatial audio reasoning enables machines to interpret auditory scenes by understanding events and their spatial attributes. In this work, we focus on spatial audio understanding with an emphasis on reasoning about moving sources. First, we introduce a spatial audio encoder that processes spatial audio to detect multiple overlapping events and estimate their spatial attributes, Direction of Arrival (DoA) and source distance, at the frame level. To generalize to unseen events, we incorporate an audio grounding model that aligns audio features with semantic audio class text embeddings via a cross-attention mechanism. Second, to answer complex queries about dynamic audio scenes involving moving sources, we condition a large language model (LLM) on structured spatial attributes extracted by our model. Finally, we introduce a spatial audio motion understanding and reasoning benchmark dataset and demonstrate our framework's performance against the baseline model.

While Large Reasoning Models (LRMs) generate extensive chain-of-thought reasoning, we lack a principled framework for understanding how these thoughts are structured. In this paper, we introduce a novel approach by applying Schoenfeld's Episode Theory, a classic cognitive framework for human mathematical problem-solving, to analyze the reasoning traces of LRMs. We annotated thousands of sentences and paragraphs from model-generated solutions to math problems using seven cognitive labels (e.g., Plan, Implement, Verify). The result is the first publicly available benchmark for the fine-grained analysis of machine reasoning, including a large annotated corpus and detailed annotation guidebooks. Our preliminary analysis reveals distinct patterns in LRM reasoning, such as the transition dynamics between cognitive states. This framework provides a theoretically grounded methodology for interpreting LRM cognition and enables future work on more controllable and transparent reasoning systems.

As large language models~(LLMs) become widely adopted, ensuring their alignment with human values is crucial to prevent jailbreaks where adversaries manipulate models to produce harmful content. While most defenses target single-turn attacks, real-world usage often involves multi-turn dialogues, exposing models to attacks that exploit conversational context to bypass safety measures. We introduce MUSE, a comprehensive framework tackling multi-turn jailbreaks from both attack and defense angles. For attacks, we propose MUSE-A, a method that uses frame semantics and heuristic tree search to explore diverse semantic trajectories. For defense, we present MUSE-D, a fine-grained safety alignment approach that intervenes early in dialogues to reduce vulnerabilities. Extensive experiments on various models show that MUSE effectively identifies and mitigates multi-turn vulnerabilities. Code is available at \href{https://github.com/yansiyu02/MUSE}{https://github.com/yansiyu02/MUSE}.

Abstract--Bird's-Eye-View (BEV) representation offers a metric-scaled planar workspace, facilitating the simplification of 6-DoF ego-motion to a more robust 3-DoF model for monocular visual odometry (MVO) in intelligent transportation systems. However, existing BEV methods suffer from sparse supervision signals and information loss during perspective-to-BEV projection. Our approach introduces: (1) dense BEV optical flow supervision constructed from 3-DoF pose ground truth for pixel-level guidance; (2) PV-BEV fusion that computes correlation volumes before projection to preserve 6-DoF motion cues while maintaining scale consistency. The framework employs three supervision levels derived solely from pose data: dense BEV flow, 5-DoF for the PV branch, and final 3-DoF output. Extensive evaluation on KITTI, NCL T, Oxford, and our newly collected ZJH-VO multi-scale dataset demonstrates state-of-the-art performance, achieving 40% improvement in RTE compared to previous BEV methods. The ZJH-VO dataset, covering diverse ground vehicle scenarios from underground parking to outdoor plazas, is publicly available to facilitate future research. IRD'S-EYE-VIEW (BEV) representation has become a cornerstone for perception and localization tasks in modern intelligent transportation systems [1]-[3], offering a powerful solution to the scale drift problem inherent in Monoc-ular Visual Odometry (MVO) [4], [5]. For ground vehicles like autonomous cars and logistics robots, motion is predominantly planar [6]. This allows for simplifying pose estimation from six degrees of freedom (6-DoF) to a more robust 3-DoF model (x,y, yaw), which naturally aligns with the unified, metric-scaled grid of BEV representation [7]. This simplification not only reduces computational complexity but also mitigates the accumulation of errors in non-primary motion axes, a common source of drift in long-range navigation.

Understanding and reasoning about entire software repositories is an essential capability for intelligent software engineering tools. While existing benchmarks such as CoSQA and CodeQA have advanced the field, they predominantly focus on small, self-contained code snippets. These setups fail to capture the complexity of real-world repositories, where effective understanding and reasoning often require navigating multiple files, understanding software architecture, and grounding answers in long-range code dependencies. In this paper, we present SWE-QA, a repository-level code question answering (QA) benchmark designed to facilitate research on automated QA systems in realistic code environments. SWE-QA involves 576 high-quality question-answer pairs spanning diverse categories, including intention understanding, cross-file reasoning, and multi-hop dependency analysis. To construct SWE-QA, we first crawled 77,100 GitHub issues from 11 popular repositories. Based on an analysis of naturally occurring developer questions extracted from these issues, we developed a two-level taxonomy of repository-level questions and constructed a set of seed questions for each category. For each category, we manually curated and validated questions and collected their corresponding answers. As a prototype application, we further develop SWE-QA-Agent, an agentic framework in which LLM agents reason and act to find answers automatically. We evaluate six advanced LLMs on SWE-QA under various context augmentation strategies. Experimental results highlight the promise of LLMs, particularly our SWE-QA-Agent framework, in addressing repository-level QA, while also revealing open challenges and pointing to future research directions.

As privacy and security take center stage in AI, machine unlearning, the ability to erase specific knowledge from models, has garnered increasing attention. However, existing methods overly prioritize efficiency and aggressive forgetting, which introduces notable limitations. In particular, radical interventions like gradient ascent, influence functions, and random label noise can destabilize model weights, leading to collapse and reduced reliability. To address this, we propose CUFG (Curriculum Unlearning via Forgetting Gradients), a novel framework that enhances the stability of approximate unlearning through innovations in both forgetting mechanisms and data scheduling strategies. Specifically, CUFG integrates a new gradient corrector guided by forgetting gradients for fine-tuning-based unlearning and a curriculum unlearning paradigm that progressively forgets from easy to hard. These innovations narrow the gap with the gold-standard Retrain method by enabling more stable and progressive unlearning, thereby improving both effectiveness and reliability. Furthermore, we believe that the concept of curriculum unlearning has substantial research potential and offers forward-looking insights for the development of the MU field. Extensive experiments across various forgetting scenarios validate the rationale and effectiveness of our approach and CUFG. Codes are available at https://anonymous.4open.science/r/CUFG-6375.