Continual Semantic Segmentation (CSS) requires learning new classes without forgetting previously acquired knowledge, addressing the fundamental challenge of catastrophic forgetting in dense prediction tasks. However, existing CSS methods typically employ single-stage encoder-decoder architectures where segmentation masks and class labels are tightly coupled, leading to interference between old and new class learning and suboptimal retention-plasticity balance. We introduce DecoupleCSS, a novel two-stage framework for CSS. By decoupling class-aware detection from class-agnostic segmentation, DecoupleCSS enables more effective continual learning, preserving past knowledge while learning new classes. The first stage leverages pre-trained text and image encoders, adapted using LoRA, to encode class-specific information and generate location-aware prompts. In the second stage, the Segment Anything Model (SAM) is employed to produce precise segmentation masks, ensuring that segmentation knowledge is shared across both new and previous classes. This approach improves the balance between retention and adaptability in CSS, achieving state-of-the-art performance across a variety of challenging tasks.

Traditional video reasoning segmentation methods rely on supervised fine-tuning, which limits generalization to out-of-distribution scenarios and lacks explicit reasoning. To address this, we propose VideoSeg-R1, the first framework to introduce reinforcement learning into video reasoning segmentation. It adopts a decoupled architecture that formulates the task as joint referring image segmentation and video mask propagation. It comprises three stages: (1) A hierarchical text-guided frame sampler to emulate human attention; (2) A reasoning model that produces spatial cues along with explicit reasoning chains; and (3) A segmentation-propagation stage using SAM2 and XMem. A task difficulty-aware mechanism adaptively controls reasoning length for better efficiency and accuracy. Extensive evaluations on multiple benchmarks demonstrate that VideoSeg-R1 achieves state-of-the-art performance in complex video reasoning and segmentation tasks.

Zhiyuan Yao, Zishan Xu, Yifu Guo, Zhiguang Han, Cheng Yang, Shuo Zhang, Weinan Zhang, Xingshan Zeng, Weiwen Liu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Recognizing speaker intent in long audio dialogues among speakers has a wide range of applications, but is a non-trivial AI task due to complex inter-dependencies in speaker utterances and scarce annotated data. To address these challenges, an end-to-end framework, namely DialogGraph-LLM, is proposed in the current work. DialogGraph-LLM combines a novel Multi-Relational Dialogue Attention Network (MR-DAN) architecture with multimodal foundation models (e.g., Qwen2.5-Omni-7B) for direct acoustic-to-intent inference. An adaptive semi-supervised learning strategy is designed using LLM with a confidence-aware pseudo-label generation mechanism based on dual-threshold filtering using both global and class confidences, and an entropy-based sample selection process that prioritizes high-information unlabeled instances. Extensive evaluations on the proprietary MarketCalls corpus and the publicly available MIntRec 2.0 benchmark demonstrate DialogGraph-LLM’s superiority over strong audio and text-driven baselines. The framework demonstrates strong performance and efficiency in intent recognition in real world scenario audio dialogues, proving its practical value for audio-rich domains with limited supervision. Our code is available at https://github.com/david188888/DialogGraph-LLM

Large Language Model (LLM)-based agents have recently shown impressive capabilities in complex reasoning and tool use via multi-step interactions with their environments. While these agents have the potential to tackle complicated tasks, their problem-solving process—agents' interaction trajectory leading to task completion—remains underexploited. These trajectories contain rich feedback that can navigate agents toward the right directions for solving problems correctly. Although prevailing approaches, such as Monte Carlo Tree Search (MCTS), can effectively balance exploration and exploitation, they ignore the interdependence among various trajectories and lack the diversity of search spaces, which leads to redundant reasoning and suboptimal outcomes. To address these challenges, we propose SE-Agent, a Self-Evolution framework that enables Agents to optimize their reasoning processes iteratively. Our approach revisits and enhances former pilot trajectories through three key operations: revision, recombination, and refinement. This evolutionary mechanism enables two critical advantages: (1) it expands the search space beyond local optima by intelligently exploring diverse solution paths guided by previous trajectories, and (2) it leverages cross-trajectory inspiration to efficiently enhance performance while mitigating the impact of suboptimal reasoning paths. Through these mechanisms, SE-Agent achieves continuous self-evolution that incrementally improves reasoning quality. We evaluate SE-Agent on SWE-bench Verified to resolve real-world GitHub issues. Experimental results across five strong LLMs show that integrating SE-Agent delivers up to 55% relative improvement, achieving state-of-the-art performance among all open-source agents on SWE-bench Verified.

The ultimate goal of code agents is to solve complex tasks autonomously. Although large language models (LLMs) have made substantial progress in code generation, real-world tasks typically demand full-fledged code repositories rather than simple scripts. Building such repositories from scratch remains a major challenge. Fortunately, GitHub hosts a vast, evolving collection of open-source repositories, which developers frequently reuse as modular components for complex tasks. Yet, existing frameworks like OpenHands and SWE-Agent still struggle to effectively leverage these valuable resources. Relying solely on README files provides insufficient guidance, and deeper exploration reveals two core obstacles: overwhelming information and tangled dependencies of repositories, both constrained by the limited context windows of current LLMs. To tackle these issues, we propose RepoMaster, an autonomous agent framework designed to explore and reuse GitHub repositories for solving complex tasks. For efficient understanding, RepoMaster constructs function-call graphs, module-dependency graphs, and hierarchical code trees to identify essential components, providing only identified core elements to the LLMs rather than the entire repository. During autonomous execution, it progressively explores related components using our exploration tools and prunes information to optimize context usage. Evaluated on the adjusted MLE-bench, RepoMaster achieves a 110\% relative boost in valid submissions over the strongest baseline OpenHands. On our newly released GitTaskBench, RepoMaster lifts the task-pass rate from 40.7% to 62.9% while reducing token usage by 95%. Our code and demonstration materials are publicly available at https://github.com/QuantaAlpha/RepoMaster.

Automated scoring of open-ended questions continues to face significant challenges in modeling student cognition, ensuring scoring consistency, and providing interpretability. Although large language models (LLMs) have demonstrated substantial potential, single-model architectures exhibit structural limitations in handling complex reasoning and cognitive alignment. To address these issues, we propose CogMAS, a Cognitively-Grounded Multi-Agent Scoring Framework that incorporates three types of agents, i.e., student agents, teacher agents, and evaluation agents, to enable multidimensional and interpretable scoring of open-ended responses. CogMAS leverages Bloom’s taxonomy to construct a mapping between questions and cognitive dimensions, guiding teacher agents to perform dimension-aware scoring. A dual-stage semantic retrieval module is introduced to provide contextually relevant exemplars. Evaluation agents are responsible for detecting explanation path biases and deriving high-confidence reasoning chains and final scores. Teacher agents are further trained using Direct Preference Optimization (DPO) to improve the quality and consistency of scoring explanations. High-confidence score–explanation pairs are stored in a retrievable memory module to support continuous optimization in future tasks. Experiments on three public open-ended question scoring datasets demonstrate that CogMAS achieves state-of-the-art performance in both scoring accuracy and consistency, validating its effectiveness and generalizability.

This paper presents a 1st-order extended-counting (EC) incremental analog-to-digital converter (IADC) based on dynamic amplifier (DA). The wide output swing and high gain requirement of the amplifier makes it infeasible to design an energy-efficient high-resolution EC IADC based on DA. Conventional gain-boosting methods, such as multi-stage and cascoding, suffer from the stability issue and limited output swing respectively. To solve this problem, a foreground gain calibration method is proposed for reducing the gain requirement by compensating the quantization error caused by the limited gain of DA. Additionally, a two-stage DA is proposed to achieve wide output swing and high energy-efficiency, which combines the floating-inverter-based amplifier (FIA) and the switched-inverter-based amplifier (SIA). SIA works as the 2nd stage, providing wide output swing while FIA as the 1st stage maintains the gain > 50 dB. A prototype of 1st-order EC IADC based on the proposed DA with gain calibration is fabricated in 180-nm CMOS technology. Operating under 1.8-V power supply, the EC IADC achieves a SNDR/DR of 94.6 dB/ 98.1 dB and 20μW power consumption at 921 kHz offs, leading to a Schreier figure-of-merit (FOMSNDR) of 174.1 dB.

With the development of cloud computing,as a cloud computing service, cloud storage widely applied to enterprises and people's daily life in the form of the public cloud storage, hybrid cloud storage , internal cloud storage. In the current internet environment, the annual investment in providing public cloud storage services is more than ?500 million. More and more companies have joined the R & D team on cloud storage. Because the product differentiation of Cloud storage products is small, the personal preferences of user groups is completed, the market is a buyer's market, and other reasons. Further, the profit of private cloud storage model is also not clear. Cloud storage pricing model has become the focus for the user and vendor. In this article, the authors will be starting from different pricing strategies to discuss the pricing model of cloud storage.

This paper mainly develops the operation strategy of an enterprise drive based on viral marketing. Firstly, it introduces the concept, the features and successful cases of viral marketing. Next, it introduces what is Kingdee cloud drive and shows the comparative results among Kingdee cloud drive and other drives. With respect to the market positioning, this paper designs the operation strategy. Based on viral marketing, the detailed progress and reward mechanism are put forward in order to attract more users. Finally, the lottery model is developed based on the analytic network process in the reward mechanism.