I$^2$S-TFCKD: Intra-Inter Set Knowledge Distillation with Time-Frequency Calibration for Speech Enhancement

In recent years, complexity compression of neural network (NN)-based speech enhancement (SE) models has gradually attracted the attention of researchers, especially in scenarios with limited hardware resources or strict latency requirements. The main difficulties and challenges lie in achieving a balance between complexity and performance according to the characteristics of the task. In this paper, we propose an intra-inter set knowledge distillation (KD) framework with time-frequency calibration (I$^2$S-TFCKD) for SE. Different from previous distillation strategies for SE, the proposed framework fully utilizes the time-frequency differential information of speech while promoting global knowledge flow. Firstly, we propose a multi-layer interactive distillation based on dual-stream time-frequency cross-calibration, which calculates the teacher-student similarity calibration weights in the time and frequency domains respectively and performs cross-weighting, thus enabling refined allocation of distillation contributions across different layers according to speech characteristics. Secondly, we construct a collaborative distillation paradigm for intra-set and inter-set correlations. Within a correlated set, multi-layer teacher-student features are pairwise matched for calibrated distillation. Subsequently, we generate representative features from each correlated set through residual fusion to form the fused feature set that enables inter-set knowledge interaction. The proposed distillation strategy is applied to the dual-path dilated convolutional recurrent network (DPDCRN) that ranked first in the SE track of the L3DAS23 challenge. Objective evaluations demonstrate that the proposed KD strategy consistently and effectively improves the performance of the low-complexity student model and outperforms other distillation schemes.

Video-CoT: A Comprehensive Dataset for Spatiotemporal Understanding of Videos Based on Chain-of-Thought

Video content comprehension is essential for various applications, ranging from video analysis to interactive systems. Despite advancements in large-scale vision-language models (VLMs), these models often struggle to capture the nuanced, spatiotemporal details essential for thorough video analysis. To address this gap, we introduce Video-CoT, a groundbreaking dataset designed to enhance spatiotemporal understanding using Chain-of-Thought (CoT) methodologies. Video-CoT contains 192,000 fine-grained spa-tiotemporal question-answer pairs and 23,000 high-quality CoT-annotated samples, providing a solid foundation for evaluating spatiotemporal understanding in video comprehension. Additionally, we provide a comprehensive benchmark for assessing these tasks, with each task featuring 750 images and tailored evaluation metrics. Our extensive experiments reveal that current VLMs face significant challenges in achieving satisfactory performance, high-lighting the difficulties of effective spatiotemporal understanding. Overall, the Video-CoT dataset and benchmark open new avenues for research in multimedia understanding and support future innovations in intelligent systems requiring advanced video analysis capabilities. By making these resources publicly available, we aim to encourage further exploration in this critical area. Project website:https://video-cot.github.io/ .

Your Classifier Can Do More: Towards Bridging the Gaps in Classification, Robustness, and Generation

Joint Energy-based Models (JEMs), a class of hybrid generative-discriminative models, are well known for their ability to achieve both high classification accuracy and generative capability within a single model. However, their robustness still lags significantly behind the classifiers based adversarial training (AT). Conversely, while AT is currently the most effective approach to improving the classifier's robustness, it typically sacrifices accuracy on clean data and lacks generative capability. The triple trade-off between classification accuracy, generative capability and robustness, raises a natural question: Can a single model simultaneously achieve high classification accuracy, adversarial robustness, and generative performance? -- a goal that has been rarely explored. To address this question, we systematically analyze the energy distribution differences of clean, adversarial, and generated samples across various JEM variants and adversarially trained models. We observe that AT tends to reduce the energy gap between clean and adversarial samples, while JEMs reduce the gap between clean and synthetic ones. This observation suggests a key insight: if the energy distributions of all three data types can be aligned, we might unify the strengths of AT and JEMs, resolving their inherent trade-offs. Building on this idea, we propose Energy-based Joint Distribution Adversarial Training (EB-JDAT), to jointly model the clean data distribution, the adversarial distribution, and the classifier by maximizing their joint probability. EB-JDAT is a general and flexible optimization method, compatible with various JEM variants. Extensive experimental results demonstrate that EB-JDAT not only maintains near original accuracy and generative capability of JEMs, but also significantly enhances robustness, even surpassing state-of-the-art ATs.

VTLA: Vision-Tactile-Language-Action Model with Preference Learning for Insertion Manipulation

While vision-language models have advanced significantly, their application in language-conditioned robotic manipulation is still underexplored, especially for contact-rich tasks that extend beyond visually dominant pick-and-place scenarios. To bridge this gap, we introduce Vision-Tactile-Language-Action model, a novel framework that enables robust policy generation in contact-intensive scenarios by effectively integrating visual and tactile inputs through cross-modal language grounding. A low-cost, multi-modal dataset has been constructed in a simulation environment, containing vision-tactile-action-instruction pairs specifically designed for the fingertip insertion task. Furthermore, we introduce Direct Preference Optimization (DPO) to offer regression-like supervision for the VTLA model, effectively bridging the gap between classification-based next token prediction loss and continuous robotic tasks. Experimental results show that the VTLA model outperforms traditional imitation learning methods (e.g., diffusion policies) and existing multi-modal baselines (TLA/VLA), achieving over 90% success rates on unseen peg shapes. Finally, we conduct real-world peg-in-hole experiments to demonstrate the exceptional Sim2Real performance of the proposed VTLA model. For supplementary videos and results, please visit our project website: https://sites.google.com/view/vtla

Multi-Modal Molecular Representation Learning via Structure Awareness

Accurate extraction of molecular representations is a critical step in the drug discovery process. In recent years, significant progress has been made in molecular representation learning methods, among which multi-modal molecular representation methods based on images, and 2D/3D topologies have become increasingly mainstream. However, existing these multi-modal approaches often directly fuse information from different modalities, overlooking the potential of intermodal interactions and failing to adequately capture the complex higher-order relationships and invariant features between molecules. To overcome these challenges, we propose a structure-awareness-based multi-modal self-supervised molecular representation pre-training framework (MMSA) designed to enhance molecular graph representations by leveraging invariant knowledge between molecules. The framework consists of two main modules: the multi-modal molecular representation learning module and the structure-awareness module. The multi-modal molecular representation learning module collaboratively processes information from different modalities of the same molecule to overcome intermodal differences and generate a unified molecular embedding. Subsequently, the structure-awareness module enhances the molecular representation by constructing a hypergraph structure to model higher-order correlations between molecules. This module also introduces a memory mechanism for storing typical molecular representations, aligning them with memory anchors in the memory bank to integrate invariant knowledge, thereby improving the model generalization ability. Extensive experiments have demonstrated the effectiveness of MMSA, which achieves state-of-the-art performance on the MoleculeNet benchmark, with average ROC-AUC improvements ranging from 1.8% to 9.6% over baseline methods.

RoboOS: A Hierarchical Embodied Framework for Cross-Embodiment and Multi-Agent Collaboration

The dawn of embodied intelligence has ushered in an unprecedented imperative for resilient, cognition-enabled multi-agent collaboration across next-generation ecosystems, revolutionizing paradigms in autonomous manufacturing, adaptive service robotics, and cyber-physical production architectures. However, current robotic systems face significant limitations, such as limited cross-embodiment adaptability, inefficient task scheduling, and insufficient dynamic error correction. While End-to-end VLA models demonstrate inadequate long-horizon planning and task generalization, hierarchical VLA models suffer from a lack of cross-embodiment and multi-agent coordination capabilities. To address these challenges, we introduce RoboOS, the first open-source embodied system built on a Brain-Cerebellum hierarchical architecture, enabling a paradigm shift from single-agent to multi-agent intelligence. Specifically, RoboOS consists of three key components: (1) Embodied Brain Model (RoboBrain), a MLLM designed for global perception and high-level decision-making; (2) Cerebellum Skill Library, a modular, plug-and-play toolkit that facilitates seamless execution of multiple skills; and (3) Real-Time Shared Memory, a spatiotemporal synchronization mechanism for coordinating multi-agent states. By integrating hierarchical information flow, RoboOS bridges Embodied Brain and Cerebellum Skill Library, facilitating robust planning, scheduling, and error correction for long-horizon tasks, while ensuring efficient multi-agent collaboration through Real-Time Shared Memory. Furthermore, we enhance edge-cloud communication and cloud-based distributed inference to facilitate high-frequency interactions and enable scalable deployment. Extensive real-world experiments across various scenarios, demonstrate RoboOS's versatility in supporting heterogeneous embodiments. Project website: https://github.com/FlagOpen/RoboOS

MEGA: Second-Order Gradient Alignment for Catastrophic Forgetting Mitigation in GFSCIL

Graph Few-Shot Class-Incremental Learning (GFSCIL) enables models to continually learn from limited samples of novel tasks after initial training on a large base dataset. Existing GFSCIL approaches typically utilize Prototypical Networks (PNs) for metric-based class representations and fine-tune the model during the incremental learning stage. However, these PN-based methods oversimplify learning via novel query set fine-tuning and fail to integrate Graph Continual Learning (GCL) techniques due to architectural constraints. To address these challenges, we propose a more rigorous and practical setting for GFSCIL that excludes query sets during the incremental training phase. Building on this foundation, we introduce Model-Agnostic Meta Graph Continual Learning (MEGA), aimed at effectively alleviating catastrophic forgetting for GFSCIL. Specifically, by calculating the incremental second-order gradient during the meta-training stage, we endow the model to learn high-quality priors that enhance incremental learning by aligning its behaviors across both the meta-training and incremental learning stages. Extensive experiments on four mainstream graph datasets demonstrate that MEGA achieves state-of-the-art results and enhances the effectiveness of various GCL methods in GFSCIL. We believe that our proposed MEGA serves as a model-agnostic GFSCIL paradigm, paving the way for future research.

FastRSR: Efficient and Accurate Road Surface Reconstruction from Bird's Eye View

Road Surface Reconstruction (RSR) is crucial for autonomous driving, enabling the understanding of road surface conditions. Recently, RSR from the Bird's Eye View (BEV) has gained attention for its potential to enhance performance. However, existing methods for transforming perspective views to BEV face challenges such as information loss and representation sparsity. Moreover, stereo matching in BEV is limited by the need to balance accuracy with inference speed. To address these challenges, we propose two efficient and accurate BEV-based RSR models: FastRSR-mono and FastRSR-stereo. Specifically, we first introduce Depth-Aware Projection (DAP), an efficient view transformation strategy designed to mitigate information loss and sparsity by querying depth and image features to aggregate BEV data within specific road surface regions using a pre-computed look-up table. To optimize accuracy and speed in stereo matching, we design the Spatial Attention Enhancement (SAE) and Confidence Attention Generation (CAG) modules. SAE adaptively highlights important regions, while CAG focuses on high-confidence predictions and filters out irrelevant information. FastRSR achieves state-of-the-art performance, exceeding monocular competitors by over 6.0% in elevation absolute error and providing at least a 3.0x speedup by stereo methods on the RSRD dataset. The source code will be released.

Reason-RFT: Reinforcement Fine-Tuning for Visual Reasoning

Visual reasoning abilities play a crucial role in understanding complex multimodal data, advancing both domain-specific applications and artificial general intelligence (AGI). Existing methods improve VLM reasoning via Chain-of-Thought (CoT) supervised fine-tuning, using meticulously annotated training data to enhance visual reasoning capabilities. However, this training paradigm may lead to overfitting and cognitive rigidity, restricting the model's ability to transfer visual reasoning skills across domains and limiting its real-world applicability. To address these limitations, we propose Reason-RFT, a novel reinforcement fine-tuning framework that significantly enhances generalization capabilities in visual reasoning tasks. Reason-RFT introduces a two-phase training framework for visual reasoning: (1) Supervised Fine-Tuning (SFT) with curated Chain-of-Thought (CoT) data activates the reasoning potential of Vision-Language Models (VLMs), followed by (2) Group Relative Policy Optimization (GRPO)-based reinforcement learning that generates multiple reasoning-response pairs, significantly enhancing generalization in visual reasoning tasks. To evaluate Reason-RFT's visual reasoning capabilities, we reconstructed a comprehensive dataset spanning visual counting, structure perception, and spatial transformation. Experimental results demonstrate Reasoning-RFT's three key advantages: (1) Performance Enhancement: achieving state-of-the-art results across multiple tasks, outperforming most mainstream open-source and proprietary models; (2) Generalization Superiority: consistently maintaining robust performance across diverse tasks and domains, outperforming alternative training paradigms; (3) Data Efficiency: excelling in few-shot learning scenarios while surpassing full-dataset SFT baselines. Project website: https://tanhuajie.github.io/ReasonRFT

TLA: Tactile-Language-Action Model for Contact-Rich Manipulation

Significant progress has been made in vision-language models. However, language-conditioned robotic manipulation for contact-rich tasks remains underexplored, particularly in terms of tactile sensing. To address this gap, we introduce the Tactile-Language-Action (TLA) model, which effectively processes sequential tactile feedback via cross-modal language grounding to enable robust policy generation in contact-intensive scenarios. In addition, we construct a comprehensive dataset that contains 24k pairs of tactile action instruction data, customized for fingertip peg-in-hole assembly, providing essential resources for TLA training and evaluation. Our results show that TLA significantly outperforms traditional imitation learning methods (e.g., diffusion policy) in terms of effective action generation and action accuracy, while demonstrating strong generalization capabilities by achieving over 85\% success rate on previously unseen assembly clearances and peg shapes. We publicly release all data and code in the hope of advancing research in language-conditioned tactile manipulation skill learning. Project website: https://sites.google.com/view/tactile-language-action/