FlexAM: Flexible Appearance-Motion Decomposition for Versatile Video Generation Control

Effective and generalizable control in video generation remains a significant challenge. While many methods rely on ambiguous or task-specific signals, we argue that a fundamental disentanglement of "appearance" and "motion" provides a more robust and scalable pathway. We propose FlexAM, a unified framework built upon a novel 3D control signal. This signal represents video dynamics as a point cloud, introducing three key enhancements: multi-frequency positional encoding to distinguish fine-grained motion, depth-aware positional encoding, and a flexible control signal for balancing precision and generative quality. This representation allows FlexAM to effectively disentangle appearance and motion, enabling a wide range of tasks including I2V/V2V editing, camera control, and spatial object editing. Extensive experiments demonstrate that FlexAM achieves superior performance across all evaluated tasks.

Towards Autonomous Mathematics Research

Recent advances in foundational models have yielded reasoning systems capable of achieving a gold-medal standard at the International Mathematical Olympiad. The transition from competition-level problem-solving to professional research, however, requires navigating vast literature and constructing long-horizon proofs. In this work, we introduce Aletheia, a math research agent that iteratively generates, verifies, and revises solutions end-to-end in natural language. Specifically, Aletheia is powered by an advanced version of Gemini Deep Think for challenging reasoning problems, a novel inference-time scaling law that extends beyond Olympiad-level problems, and intensive tool use to navigate the complexities of mathematical research. We demonstrate the capability of Aletheia from Olympiad problems to PhD-level exercises and most notably, through several distinct milestones in AI-assisted mathematics research: (a) a research paper (Feng26) generated by AI without any human intervention in calculating certain structure constants in arithmetic geometry called eigenweights; (b) a research paper (LeeSeo26) demonstrating human-AI collaboration in proving bounds on systems of interacting particles called independent sets; and (c) an extensive semi-autonomous evaluation (Feng et al., 2026a) of 700 open problems on Bloom's Erdos Conjectures database, including autonomous solutions to four open questions. In order to help the public better understand the developments pertaining to AI and mathematics, we suggest quantifying standard levels of autonomy and novelty of AI-assisted results, as well as propose a novel concept of human-AI interaction cards for transparency. We conclude with reflections on human-AI collaboration in mathematics and share all prompts as well as model outputs at https://github.com/google-deepmind/superhuman/tree/main/aletheia.

Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuning

Pretrained on large-scale and diverse datasets, VLA models demonstrate strong generalization and adaptability as general-purpose robotic policies. However, Supervised Fine-Tuning (SFT), which serves as the primary mechanism for adapting VLAs to downstream domains, requires substantial amounts of task-specific data and is prone to catastrophic forgetting. To address these limitations, we propose LifeLong-RFT, a simple yet effective Reinforcement Fine-Tuning (RFT) strategy for VLA models independent of online environmental feedback and pre-trained reward models. By integrating chunking-level on-policy reinforcement learning with the proposed Multi-Dimensional Process Reward (MDPR) mechanism, LifeLong-RFT quantifies the heterogeneous contributions of intermediate action chunks across three dimensions to facilitate policy optimization. Specifically, (1) the Quantized Action Consistency Reward (QACR) ensures accurate action prediction within the discrete action space; (2) the Continuous Trajectory Alignment Reward (CTAR) aligns decoded continuous action chunks with reference trajectories to ensure precise control; (3) the Format Compliance Reward (FCR) guarantees the structural validity of outputs. Comprehensive experiments across SimplerEnv, LIBERO, and real-world tasks demonstrate that LifeLong-RFT exhibits strong performance in multi-task learning. Furthermore, for continual learning on the LIBERO benchmark, our method achieves a 22% gain in average success rate over SFT, while effectively adapting to new tasks using only 20% of the training data. Overall, our method provides a promising post-training paradigm for VLAs.

VersaViT: Enhancing MLLM Vision Backbones via Task-Guided Optimization

Multimodal Large Language Models (MLLMs) have recently achieved remarkable success in visual-language understanding, demonstrating superior high-level semantic alignment within their vision encoders. An important question thus arises: Can these encoders serve as versatile vision backbones, capable of reliably performing classic vision-centric tasks as well? To address the question, we make the following contributions: (i) we identify that the vision encoders within MLLMs exhibit deficiencies in their dense feature representations, as evidenced by their suboptimal performance on dense prediction tasks (e.g., semantic segmentation, depth estimation); (ii) we propose VersaViT, a well-rounded vision transformer that instantiates a novel multi-task framework for collaborative post-training. This framework facilitates the optimization of the vision backbone via lightweight task heads with multi-granularity supervision; (iii) extensive experiments across various downstream tasks demonstrate the effectiveness of our method, yielding a versatile vision backbone suited for both language-mediated reasoning and pixel-level understanding.

SRSUPM: Sequential Recommender System Based on User Psychological Motivation

Sequential recommender infers users' evolving psychological motivations from historical interactions to recommend the next preferred items. Most existing methods compress recent behaviors into a single vector and optimize it toward a single observed target item, but lack explicit modeling of psychological motivation shift. As a result, they struggle to uncover the distributional patterns across different shift degrees and to capture collaborative knowledge that is sensitive to psychological motivation shift. We propose a general framework, the Sequential Recommender System Based on User Psychological Motivation, to enhance sequential recommenders with psychological motivation shift-aware user modeling. Specifically, the Psychological Motivation Shift Assessment quantitatively measures psychological motivation shift; guided by PMSA, the Shift Information Construction models dynamically evolving multi-level shift states, and the Psychological Motivation Shift-driven Information Decomposition decomposes and regularizes representations across shift levels. Moreover, the Psychological Motivation Shift Information Matching strengthens collaborative patterns related to psychological motivation shift to learn more discriminative user representations. Extensive experiments on three public benchmarks show that SRSUPM consistently outperforms representative baselines on diverse sequential recommender tasks.

SDFed: Bridging Local Global Discrepancy via Subspace Refinement and Divergence Control in Federated Prompt Learning

Vision-language pretrained models offer strong transferable representations, yet adapting them in privacy-sensitive multi-party settings is challenging due to the high communication cost of federated optimization and the limited local data on clients. Federated prompt learning mitigates this issue by keeping the VLPM backbone frozen and collaboratively training lightweight prompt parameters. However, existing approaches typically enforce a unified prompt structure and length across clients, which is inadequate under practical client heterogeneity in both data distributions and system resources, and may further introduce conflicts between globally shared and locally optimal knowledge. To address these challenges, we propose \textbf{SDFed}, a heterogeneous federated prompt learning framework that bridges Local-Global Discrepancy via Subspace Refinement and Divergence Control. SDFed maintains a fixed-length global prompt for efficient aggregation while allowing each client to learn a variable-length local prompt to better match its data characteristics and capacity. To mitigate local-global conflicts and facilitate effective knowledge transfer, SDFed introduces a subspace refinement method for local prompts and an information retention and divergence control strategy that preserves key local information while maintaining appropriate separability between global and local representations. Extensive experiments on several datasets demonstrate that SDFed consistently improves performance and robustness in heterogeneous federated settings.

POINTS-GUI-G: GUI-Grounding Journey

The rapid advancement of vision-language models has catalyzed the emergence of GUI agents, which hold immense potential for automating complex tasks, from online shopping to flight booking, thereby alleviating the burden of repetitive digital workflows. As a foundational capability, GUI grounding is typically established as a prerequisite for end-to-end task execution. It enables models to precisely locate interface elements, such as text and icons, to perform accurate operations like clicking and typing. Unlike prior works that fine-tune models already possessing strong spatial awareness (e.g., Qwen3-VL), we aim to master the full technical pipeline by starting from a base model with minimal grounding ability, such as POINTS-1.5. We introduce POINTS-GUI-G-8B, which achieves state-of-the-art performance with scores of 59.9 on ScreenSpot-Pro, 66.0 on OSWorld-G, 95.7 on ScreenSpot-v2, and 49.9 on UI-Vision. Our model's success is driven by three key factors: (1) Refined Data Engineering, involving the unification of diverse open-source datasets format alongside sophisticated strategies for augmentation, filtering, and difficulty grading; (2) Improved Training Strategies, including continuous fine-tuning of the vision encoder to enhance perceptual accuracy and maintaining resolution consistency between training and inference; and (3) Reinforcement Learning (RL) with Verifiable Rewards. While RL is traditionally used to bolster reasoning, we demonstrate that it significantly improves precision in the perception-intensive GUI grounding task. Furthermore, GUI grounding provides a natural advantage for RL, as rewards are easily verifiable and highly accurate.

LMMRec: LLM-driven Motivation-aware Multimodal Recommendation

Motivation-based recommendation systems uncover user behavior drivers. Motivation modeling, crucial for decision-making and content preference, explains recommendation generation. Existing methods often treat motivation as latent variables from interaction data, neglecting heterogeneous information like review text. In multimodal motivation fusion, two challenges arise: 1) achieving stable cross-modal alignment amid noise, and 2) identifying features reflecting the same underlying motivation across modalities. To address these, we propose LLM-driven Motivation-aware Multimodal Recommendation (LMMRec), a model-agnostic framework leveraging large language models for deep semantic priors and motivation understanding. LMMRec uses chain-of-thought prompting to extract fine-grained user and item motivations from text. A dual-encoder architecture models textual and interaction-based motivations for cross-modal alignment, while Motivation Coordination Strategy and Interaction-Text Correspondence Method mitigate noise and semantic drift through contrastive learning and momentum updates. Experiments on three datasets show LMMRec achieves up to a 4.98\% performance improvement.

Maneuverable-Jamming-Aided Secure Communication and Sensing in A2G-ISAC Systems

In this paper, we propose a maneuverablejamming-aided secure communication and sensing (SCS) scheme for an air-to-ground integrated sensing and communication (A2G-ISAC) system, where a dual-functional source UAV and a maneuverable jamming UAV operate collaboratively in a hybrid monostatic-bistatic radar configuration. The maneuverable jamming UAV emits artificial noise to assist the source UAV in detecting multiple ground targets while interfering with an eavesdropper. The effects of residual interference caused by imperfect successive interference cancellation on the received signal-to-interference-plus-noise ratio are considered, which degrades the system performance. To maximize the average secrecy rate (ASR) under transmit power budget, UAV maneuvering constraints, and sensing requirements, the dual-UAV trajectory and beamforming are jointly optimized. Given that secure communication and sensing fundamentally conflict in terms of resource allocation, making it difficult to achieve optimal performance for both simultaneously, we adopt a two-phase design to address this challenge. By dividing the mission into the secure communication (SC) phase and the SCS phase, the A2G-ISAC system can focus on optimizing distinct objectives separately. In the SC phase, a block coordinate descent algorithm employing the trust-region successive convex approximation and semidefinite relaxation iteratively optimizes dual-UAV trajectory and beamforming. For the SCS phase, a weighted distance minimization problem determines the suitable dual-UAV sensing positions by a greedy algorithm, followed by the joint optimization of source beamforming and jamming beamforming. Simulation results demonstrate that the proposed scheme achieves the highest ASR among benchmarks while maintaining robust sensing performance, and confirm the impact of the SIC residual interference on both secure communication and sensing.

Fedcompass: Federated Clustered and Periodic Aggregation Framework for Hybrid Classical-Quantum Models

Federated learning enables collaborative model training across decentralized clients under privacy constraints. Quantum computing offers potential for alleviating computational and communication burdens in federated learning, yet hybrid classical-quantum federated learning remains susceptible to performance degradation under non-IID data. To address this,we propose FEDCOMPASS, a layered aggregation framework for hybrid classical-quantum federated learning. FEDCOMPASS employs spectral clustering to group clients by class distribution similarity and performs cluster-wise aggregation for classical feature extractors. For quantum parameters, it uses circular mean aggregation combined with adaptive optimization to ensure stable global updates. Experiments on three benchmark datasets show that FEDCOMPASS improves test accuracy by up to 10.22% and enhances convergence stability under non-IID settings, outperforming six strong federated learning baselines.