Teach and Grow: An Agent-Centered Architecture for General Robot Learning

End-to-end vision-language-action (VLA) and world-action models offer an elegant route to general-purpose robotics, but their reliability is bounded by validated physical coverage. When an unfamiliar object, sensor, embodiment, or contact falls outside that coverage and no validated fallback exists, correcting the failure requires new robot data, a policy update, and regression testing. This recurring burden is the retraining tax. Unlike text, embodied data must often be created by operating machines. We present Teach-and-Grow Learning (TGL), an agent-centered architecture for general robot learning. In its general form, a multimodal agent turns a few successful demonstrations into reusable Skill Blocks: closed-loop behaviors for meaningful subgoals. In a new scene, the agent grounds and composes these blocks, selects learned or geometric tools, observes the physical outcome, and revises the route when execution departs from intent. A Skill Library stores executable behavior, while structured Experience Memory carries forward success, failure, and repair. New tasks are acquired without task-specific policy retraining. Our LIBERO evaluation attains state-of-the-art performance; controlled studies expose skill induction, persistent reuse, and agent-directed adaptation. Finally, we propose the Teach-and-Grow scaling-law hypothesis: if X denotes effective reusable experience, future-task error and teaching demand should approach irreducible floors as power laws in X. The architecture therefore treats deployment as a period of continued learning, in which one task can make the next easier.

Omni-Prune: Query-Aware Unified Token Pruning for Efficient Omnimodal Large Language Models

Omnimodal large language models (OmniLLMs) are rapidly extending multimodal reasoning to cover synchronized audio and video. However, the resulting audio-video token sequences are long, leading to high prefill latency and GPU memory usage at inference time. Existing token pruning methods, designed mainly for vision-only inputs, miss both the cross-modal links between audio and video and the user query that decides which content matters. To bridge this gap, we present Omni-Prune, a training-free, query-aware audio-visual token pruning framework that jointly removes redundancy from both modalities while keeping task-relevant cross-modal evidence. Specifically, Omni-Prune first splits the token sequence into adaptive time windows placed at audio saliency peaks, then scores audio and video tokens on a single scale that combines encoder attention with text-query relevance, and pairs related audio-video tokens so that they are kept together. Within each window, a final K-medoids step then selects a few representative tokens, adding diverse cues that score-based selection alone would miss. Extensive experiments demonstrate that Omni-Prune outperforms established baseline methods, delivering up to 3.25x prefill speedup and 1.3x memory reduction while retaining over 99% of full-model performance.

Light-Omni: Reflex over Reasoning in Agentic Video Understanding with Long-Term Memory

Agentic video understanding equips models with long-term memory to autonomously process and respond to continuous, long-horizon multimodal streams. However, advanced video agents often rely on ``detective-style'' iterative reasoning for action control (e.g., $\mathtt{search}$) and evidence aggregation, incurring prohibitive costs and latency. We argue that such heavy reasoning primarily compensates for the lack of global context and semantic misalignment in retrieval. This paper introduces Light-Omni, a multimodal agent framework for reflexive and lightweight video understanding. It achieves this through dual contextual states that instantly build the required context in a single forward pass. First, we maintain a global state, a finite-sized multimodal script continuously consolidated from episodic memory, serving as the global context for Light-Omni. Through hierarchical merging, it preserves recent details while summarizing past events. Second, conditioned on this global context, we generate a parametric latent state that directly drives autonomous actions and produces retrieval embeddings, with minimal latency. Benefiting from this coupled design, Light-Omni achieves semantically aligned retrieval and reflexive responses while avoiding iterative reasoning. Extensive experiments validate the effectiveness of Light-Omni across multiple video benchmarks. Notably, it outperforms M3-Agent with an average 2.4% accuracy gain, a 12.1$\times$ speedup, and a 2.6$\times$ improvement in GPU memory efficiency. Furthermore, it serves as a memory system to enhance both the performance and efficiency of existing MLLMs. Project page: https://clare-nie.github.io/Light-Omni.

EvoEmbedding: Evolvable Representations for Long-Context Retrieval and Agentic Memory

Existing embedding models are inherently static: they encode text segments in isolation, ignoring their surrounding context and temporal order. This paper introduces EvoEmbedding, a novel embedding model that generates evolvable representations for retrieval. It is tailored for long-context scenarios, where information is dynamic, sequential, and requires continuous state tracking. Our design is simple: EvoEmbedding maintains a continuously updated latent memory as it sequentially processes inputs, and uses it alongside the raw content to jointly generate evolvable embeddings. Consequently, for the same query, our model adapts its representation to retrieve distinct targets based on the evolving context, going beyond static semantic search. To equip the model with this capability, we construct EvoTrain-180K, a diverse dataset for the joint optimization of latent memory and retrieval. Furthermore, we introduce a memory queue to prevent representation collapse during recurrent encoding, alongside segment-batching techniques that tackle significant length variance and accelerate training by 3.8$\times$. Extensive experiments show that our model not only outperforms larger-scale specialists (e.g., Qwen3-Embedding-8B and KaLM-Embedding-Gemma3-12B) across a range of long-context retrieval benchmarks, but also generalizes well to downstream tasks (e.g., personalization) with contexts 10$\times$ longer than its training window. Notably, EvoEmbedding seamlessly integrates into agentic workflows to boost performance. For instance, a naive RAG pipeline equipped with our model surpasses dedicated agentic memory systems. Project Page: https://clare-nie.github.io/EvoEmbedding/.

Towards the Vision-Sound-Language-Action Paradigm: The HEAR Framework for Sound-Centric Manipulation

While recent Vision-Language-Action (VLA) models have begun to incorporate audio, they typically treat sound as static pre-execution prompts or focus exclusively on human speech. This leaves a significant gap in real-time, sound-centric manipulation where fleeting environmental acoustics provide critical state verification during task execution. Consequently, key sounds are easily missed due to low-frequency updates or system latency. This problem is exacerbated by action chunking with open-loop execution, which creates a Blind Execution Interval where acoustic events are lost between discrete audio observation windows. Recognizing the necessity of continuous auditory awareness, we formalize Vision-Sound-Language-Action (VSLA) as a continuous control paradigm conditioned on vision, streaming audio, language, and proprioception under delayed decision loops. As an instantiation, we introduce HEAR, a VSLA framework integrating four components: (i) a streaming Historizer to maintain a compact, causal audio context across execution gaps; (ii) an Envisioner adapted from omni foundation models to reason over multi-sensory inputs; (iii) an Advancer, formulated as an audio world model, to learn temporal dynamics by predicting near-future audio codes; and (iv) a flow-matching Realizer policy to generate smooth action chunks. To address the scarcity of pretraining data and evaluations for VSLA, we construct OpenX-Sound for pretraining, alongside HEAR-Bench, the first sound-centric manipulation benchmark with strict causal timing rules. Our results suggest that robust sound-centric manipulation necessitates causal persistence and explicit temporal learning. This framework provides a practical step toward multi-sensory foundation models for embodied agents, enabling robots to perceive and interact with dynamic environments. Code and videos are available at https://hear.irmv.top.

ERMV: Editing 4D Robotic Multi-view images to enhance embodied agents

Robot imitation learning relies on 4D multi-view sequential images. However, the high cost of data collection and the scarcity of high-quality data severely constrain the generalization and application of embodied intelligence policies like Vision-Language-Action (VLA) models. Data augmentation is a powerful strategy to overcome data scarcity, but methods for editing 4D multi-view sequential images for manipulation tasks are currently lacking. Thus, we propose ERMV (Editing Robotic Multi-View 4D data), a novel data augmentation framework that efficiently edits an entire multi-view sequence based on single-frame editing and robot state conditions. This task presents three core challenges: (1) maintaining geometric and appearance consistency across dynamic views and long time horizons; (2) expanding the working window with low computational costs; and (3) ensuring the semantic integrity of critical objects like the robot arm. ERMV addresses these challenges through a series of innovations. First, to ensure spatio-temporal consistency in motion blur, we introduce a novel Epipolar Motion-Aware Attention (EMA-Attn) mechanism that learns pixel shift caused by movement before applying geometric constraints. Second, to maximize the editing working window, ERMV pioneers a Sparse Spatio-Temporal (STT) module, which decouples the temporal and spatial views and remodels a single-frame multi-view problem through sparse sampling of the views to reduce computational demands. Third, to alleviate error accumulation, we incorporate a feedback intervention Mechanism, which uses a Multimodal Large Language Model (MLLM) to check editing inconsistencies and request targeted expert guidance only when necessary. Extensive experiments demonstrate that ERMV-augmented data significantly boosts the robustness and generalization of VLA models in both simulated and real-world environments.

MovSAM: A Single-image Moving Object Segmentation Framework Based on Deep Thinking

Moving object segmentation plays a vital role in understanding dynamic visual environments. While existing methods rely on multi-frame image sequences to identify moving objects, single-image MOS is critical for applications like motion intention prediction and handling camera frame drops. However, segmenting moving objects from a single image remains challenging for existing methods due to the absence of temporal cues. To address this gap, we propose MovSAM, the first framework for single-image moving object segmentation. MovSAM leverages a Multimodal Large Language Model (MLLM) enhanced with Chain-of-Thought (CoT) prompting to search the moving object and generate text prompts based on deep thinking for segmentation. These prompts are cross-fused with visual features from the Segment Anything Model (SAM) and a Vision-Language Model (VLM), enabling logic-driven moving object segmentation. The segmentation results then undergo a deep thinking refinement loop, allowing MovSAM to iteratively improve its understanding of the scene context and inter-object relationships with logical reasoning. This innovative approach enables MovSAM to segment moving objects in single images by considering scene understanding. We implement MovSAM in the real world to validate its practical application and effectiveness for autonomous driving scenarios where the multi-frame methods fail. Furthermore, despite the inherent advantage of multi-frame methods in utilizing temporal information, MovSAM achieves state-of-the-art performance across public MOS benchmarks, reaching 92.5\% on J\&F. Our implementation will be available at https://github.com/IRMVLab/MovSAM.

RLSAC: Reinforcement Learning enhanced Sample Consensus for End-to-End Robust Estimation

Robust estimation is a crucial and still challenging task, which involves estimating model parameters in noisy environments. Although conventional sampling consensus-based algorithms sample several times to achieve robustness, these algorithms cannot use data features and historical information effectively. In this paper, we propose RLSAC, a novel Reinforcement Learning enhanced SAmple Consensus framework for end-to-end robust estimation. RLSAC employs a graph neural network to utilize both data and memory features to guide exploring directions for sampling the next minimum set. The feedback of downstream tasks serves as the reward for unsupervised training. Therefore, RLSAC can avoid differentiating to learn the features and the feedback of downstream tasks for end-to-end robust estimation. In addition, RLSAC integrates a state transition module that encodes both data and memory features. Our experimental results demonstrate that RLSAC can learn from features to gradually explore a better hypothesis. Through analysis, it is apparent that RLSAC can be easily transferred to other sampling consensus-based robust estimation tasks. To the best of our knowledge, RLSAC is also the first method that uses reinforcement learning to sample consensus for end-to-end robust estimation. We release our codes at https://github.com/IRMVLab/RLSAC.

ABC: Attention with Bilinear Correlation for Infrared Small Target Detection

Infrared small target detection (ISTD) has a wide range of applications in early warning, rescue, and guidance. However, CNN based deep learning methods are not effective at segmenting infrared small target (IRST) that it lack of clear contour and texture features, and transformer based methods also struggle to achieve significant results due to the absence of convolution induction bias. To address these issues, we propose a new model called attention with bilinear correlation (ABC), which is based on the transformer architecture and includes a convolution linear fusion transformer (CLFT) module with a novel attention mechanism for feature extraction and fusion, which effectively enhances target features and suppresses noise. Additionally, our model includes a u-shaped convolution-dilated convolution (UCDC) module located deeper layers of the network, which takes advantage of the smaller resolution of deeper features to obtain finer semantic information. Experimental results on public datasets demonstrate that our approach achieves state-of-the-art performance. Code is available at https://github.com/PANPEIWEN/ABC

TaCo: Textual Attribute Recognition via Contrastive Learning

As textual attributes like font are core design elements of document format and page style, automatic attributes recognition favor comprehensive practical applications. Existing approaches already yield satisfactory performance in differentiating disparate attributes, but they still suffer in distinguishing similar attributes with only subtle difference. Moreover, their performance drop severely in real-world scenarios where unexpected and obvious imaging distortions appear. In this paper, we aim to tackle these problems by proposing TaCo, a contrastive framework for textual attribute recognition tailored toward the most common document scenes. Specifically, TaCo leverages contrastive learning to dispel the ambiguity trap arising from vague and open-ended attributes. To realize this goal, we design the learning paradigm from three perspectives: 1) generating attribute views, 2) extracting subtle but crucial details, and 3) exploiting valued view pairs for learning, to fully unlock the pre-training potential. Extensive experiments show that TaCo surpasses the supervised counterparts and advances the state-of-the-art remarkably on multiple attribute recognition tasks. Online services of TaCo will be made available.