Ruyi Ji

Zilin Zhu, Longteng Guo, Yanghong Mei et al.

Long-horizon household tasks demand robust high-level planning and sustained reasoning capabilities, which are largely overlooked by existing embodied AI benchmarks that emphasize short-horizon navigation or manipulation and rely on fixed task categories. We introduce LongAct, a benchmark designed to evaluate planning-level autonomy in long-horizon household tasks specified through free-form instructions. By abstracting away embodiment-specific low-level control, LongAct isolates high-level cognitive capabilities such as instruction understanding, dependency management, memory maintenance, and adaptive planning. We further propose HoloMind, a VLM-driven agent with a DAG-based long-horizon hierarchical planner, a Multimodal Spatial Memory for persistent world modeling, an Episodic Memory for experience reuse, and a global Critic for reflective supervision. Experiments with GPT-5 and Qwen3-VL models show that HoloMind substantially improves long-horizon performance while reducing reliance on model scale. Even top models achieve only 59% goal completion and 16% full-task success, underscoring the difficulty of LongAct and the need for stronger long-horizon planning in embodied agents.

Zikang Liu, Longteng Guo, Handong Li et al.

Real-time understanding of continuous video streams is essential for interactive assistants and multimodal agents operating in dynamic environments. However, most existing video reasoning approaches follow a batch paradigm that defers reasoning until the full video context is observed, resulting in high latency and growing computational cost that are incompatible with streaming scenarios. In this paper, we introduce ThinkStream, a framework for streaming video reasoning based on a Watch--Think--Speak paradigm that enables models to incrementally update their understanding as new video observations arrive. At each step, the model performs a short reasoning update and decides whether sufficient evidence has accumulated to produce a response. To support long-horizon streaming, we propose Reasoning-Compressed Streaming Memory (RCSM), which treats intermediate reasoning traces as compact semantic memory that replaces outdated visual tokens while preserving essential context. We further train the model using a Streaming Reinforcement Learning with Verifiable Rewards scheme that aligns incremental reasoning and response timing with the requirements of streaming interaction. Experiments on multiple streaming video benchmarks show that ThinkStream significantly outperforms existing online video models while maintaining low latency and memory usage. Code, models and data will be released at https://github.com/johncaged/ThinkStream

Jinzhi Zheng, Ruyi Ji, Libo Zhang et al.

Scene text recognition, as a cross-modal task involving vision and text, is an important research topic in computer vision. Most existing methods use language models to extract semantic information for optimizing visual recognition. However, the guidance of visual cues is ignored in the process of semantic mining, which limits the performance of the algorithm in recognizing irregular scene text. To tackle this issue, we propose a novel cross-modal fusion network (CMFN) for irregular scene text recognition, which incorporates visual cues into the semantic mining process. Specifically, CMFN consists of a position self-enhanced encoder, a visual recognition branch and an iterative semantic recognition branch. The position self-enhanced encoder provides character sequence position encoding for both the visual recognition branch and the iterative semantic recognition branch. The visual recognition branch carries out visual recognition based on the visual features extracted by CNN and the position encoding information provided by the position self-enhanced encoder. The iterative semantic recognition branch, which consists of a language recognition module and a cross-modal fusion gate, simulates the way that human recognizes scene text and integrates cross-modal visual cues for text recognition. The experiments demonstrate that the proposed CMFN algorithm achieves comparable performance to state-of-the-art algorithms, indicating its effectiveness.

Ruyi Ji, Dawei Du, Libo Zhang et al.

The majority of existing human parsing methods formulate the task as semantic segmentation, which regard each semantic category equally and fail to exploit the intrinsic physiological structure of human body, resulting in inaccurate results. In this paper, we design a novel semantic neural tree for human parsing, which uses a tree architecture to encode physiological structure of human body, and designs a coarse to fine process in a cascade manner to generate accurate results. Specifically, the semantic neural tree is designed to segment human regions into multiple semantic subregions (e.g., face, arms, and legs) in a hierarchical way using a new designed attention routing module. Meanwhile, we introduce the semantic aggregation module to combine multiple hierarchical features to exploit more context information for better performance. Our semantic neural tree can be trained in an end-to-end fashion by standard stochastic gradient descent (SGD) with back-propagation. Several experiments conducted on four challenging datasets for both single and multiple human parsing, i.e., LIP, PASCAL-Person-Part, CIHP and MHP-v2, demonstrate the effectiveness of the proposed method. Code can be found at https://isrc.iscas.ac.cn/gitlab/research/sematree.

Ruyi Ji, Longyin Wen, Libo Zhang et al.

Fine-grained visual categorization (FGVC) is an important but challenging task due to high intra-class variances and low inter-class variances caused by deformation, occlusion, illumination, etc. An attention convolutional binary neural tree architecture is presented to address those problems for weakly supervised FGVC. Specifically, we incorporate convolutional operations along edges of the tree structure, and use the routing functions in each node to determine the root-to-leaf computational paths within the tree. The final decision is computed as the summation of the predictions from leaf nodes. The deep convolutional operations learn to capture the representations of objects, and the tree structure characterizes the coarse-to-fine hierarchical feature learning process. In addition, we use the attention transformer module to enforce the network to capture discriminative features. The negative log-likelihood loss is used to train the entire network in an end-to-end fashion by SGD with back-propagation. Several experiments on the CUB-200-2011, Stanford Cars and Aircraft datasets demonstrate that the proposed method performs favorably against the state-of-the-arts.

Linyuan Gong, Ruyi Ji

TextCNN, the convolutional neural network for text, is a useful deep learning algorithm for sentence classification tasks such as sentiment analysis and question classification. However, neural networks have long been known as black boxes because interpreting them is a challenging task. Researchers have developed several tools to understand a CNN for image classification by deep visualization, but research about deep TextCNNs is still insufficient. In this paper, we are trying to understand what a TextCNN learns on two classical NLP datasets. Our work focuses on functions of different convolutional kernels and correlations between convolutional kernels.