Zhonghang Yuan, Zhefan Wang, Fang Hu, Zihong Chen, Jinzhe Li, Gang Li, Jie Ying, Huanjun Kong, Songyang Zhang, Nanqing Dong. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Weed control is a critical challenge in modern agriculture, as weeds compete with crops for essential nutrient resources, significantly reducing crop yield and quality. Traditional weed control methods, including chemical and mechanical approaches, have real-life limitations such as associated environmental impact and efficiency. An emerging yet effective approach is laser weeding, which uses a laser beam as the stem cutter. Although there have been studies that use deep learning in weed recognition, its application in intelligent laser weeding still requires a comprehensive understanding. Thus, this study serves the first empirical study on weed recognition for laser weeding. To increase the efficiency of laser beam cut and avoid damaging the crops of interest, the laser beam shall be directly aimed at the weed root. Yet, weed stem detection remains an under-explored problem. We integrate the detection of crop and weed with the localization of weed stem into one end-to-end system. To train and validate the proposed system in a real-life scenario, we curate and construct a high-quality weed stem detection dataset with human annotations. The dataset consists of 7,161 high-resolution pictures collected in the field with annotations of 11,151 instances of weed. The dataset will be released upon acceptance. Experimental results show that, in contrast to seminal weed recognition systems, the proposed system can efficiently improve the weeding accuracy by 5.05% and reduce the energy cost by 32.3%.

Haoyang Su, Renqi Chen, Shixiang Tang, Zhenfei Yin, Xinzhe Zheng, Jinzhe Li, Biqing Qi, Qi Wu, Hui Li, Wanli Ouyang, Philip Torr, Bowen Zhou, Nanqing Dong. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

In recent years, Unsupervised Domain Adaptation (UDA) has gained significant attention in the field of Natural Language Processing (NLP) owing to its ability to enhance model generalization across diverse domains. However, its application for knowledge transfer between distinct legal domains remains largely unexplored. To address the challenges posed by lengthy and complex legal texts and the limited availability of large-scale annotated datasets, we propose JurisCTC, a novel model designed to improve the accuracy of Legal Judgment Prediction (LJP) tasks. Unlike existing approaches, JurisCTC facilitates effective knowledge transfer across various legal domains and employs contrastive learning to distinguish samples from different domains. Specifically, for the LJP task, we enable knowledge transfer between civil and criminal law domains. Compared to other models and specific large language models (LLMs), JurisCTC demonstrates notable advancements, achieving peak accuracies of 76.59% and 78.83%, respectively.1

The Internet of Things (IoT) can only thrive if IoT sensor nodes can be effortlessly deployed and maintained without compromising their general-purpose nature. However, existing low-power sensor systems fail to strike a balance between these two issues, leaving the widespread of IoT sensor nodes as an open problem. In this paper, we propose LEGO+ as a minimalist yet general-purpose sensing edge-end architecture. Instead of running embedded software on a redundant general-purpose microprocessor, LEGO+ can directly construct the desired control functionality for various IoT sensing applications through hardware-level logic orchestration. To achieve this, we first conduct an in-depth analysis of the underlying unit behaviors within IoT sensor systems and, based on this, abstract a uniform logic orchestration model. Next, to enable sensor nodes to comprehend and execute the generated logic, we devise a hierarchical atomic control circuit with negligible overheads. Finally, we develop a task state prediction scheme to further improve the overall operation efficiency among multiple nodes. We prototype LEGO+ for proof-of-concept and conduct comprehensive experiments, and the results demonstrate that LEGO+ can reduce the overall power consumption of sensor nodes by 86% and enhance task efficiency by 49%, thereby facilitating a wider array of IoT sensing applications.

Retrieval-augmented generation (RAG) has become a widely adopted paradigm for enabling knowledge-grounded large language models (LLMs). However, standard RAG pipelines often fail to ensure that model reasoning remains consistent with the evidence retrieved, leading to factual inconsistencies or unsupported conclusions. In this work, we reinterpret RAG as \textit{Retrieval-Augmented Reasoning} and identify a central but underexplored problem: \textit{Reasoning Misalignment}—the divergence between an LLM's internal reasoning trajectory and the evidential constraints provided by retrieval. To address this issue, we propose \textsc{AlignRAG}, a novel iterative framework grounded in \textit{Critique-Driven Alignment (CDA)}. We further introduce \textsc{AlignRAG-auto}, an autonomous variant that dynamically terminates refinement, removing the need to pre-specify the number of critique iterations. At the heart of \textsc{AlignRAG} lies a \textit{contrastive critique synthesis} mechanism that generates retrieval-sensitive critiques while mitigating self-bias. This mechanism trains a dedicated retrieval-augmented \textit{Critic Language Model (CLM)} using labeled critiques that distinguish between evidence-aligned and misaligned reasoning. Empirical evaluations show that our approach significantly improves reasoning fidelity. Our 8B-parameter CLM improves performance over the Self-Refine baseline by \textbf{12.1\%} on out-of-domain tasks and outperforms a standard 72B-parameter CLM by \textbf{2.2\%}. Furthermore, \textsc{AlignRAG-auto} achieves this state-of-the-art performance while dynamically determining the optimal number of refinement steps, enhancing efficiency and usability. \textsc{AlignRAG} remains compatible with existing RAG architectures as a \textit{plug-and-play} module and demonstrates strong robustness under both informative and noisy retrieval scenarios. Overall, \textsc{AlignRAG} offers a principled solution for aligning model reasoning with retrieved evidence, substantially improving the factual reliability and robustness of RAG systems. Our source code is provided at \href{https://github.com/upup-wei/RAG-ReasonAlignment}{link}.

Deep learning-based computational methods have achieved promising results in predicting protein-protein interactions (PPIs). However, existing benchmarks predominantly focus on isolated pairwise evaluations, overlooking a model's capability to reconstruct biologically meaningful PPI networks, which is crucial for biology research. To address this gap, we introduce PRING, the first comprehensive benchmark that evaluates PRotein-protein INteraction prediction from a Graph-level perspective. PRING curates a high-quality, multi-species PPI network dataset comprising 21,484 proteins and 186,818 interactions, with well-designed strategies to address both data redundancy and leakage. Building on this golden-standard dataset, we establish two complementary evaluation paradigms: (1) topology-oriented tasks, which assess intra and cross-species PPI network construction, and (2) function-oriented tasks, including protein complex pathway prediction, GO module analysis, and essential protein justification. These evaluations not only reflect the model's capability to understand the network topology but also facilitate protein function annotation, biological module detection, and even disease mechanism analysis. Extensive experiments on four representative model categories, consisting of sequence similarity-based, naive sequence-based, protein language model-based, and structure-based approaches, demonstrate that current PPI models have potential limitations in recovering both structural and functional properties of PPI networks, highlighting the gap in supporting real-world biological applications. We believe PRING provides a reliable platform to guide the development of more effective PPI prediction models for the community. The dataset and source code of PRING are available at https://github.com/SophieSarceau/PRING.

Protein sequence not only determines its structure but also provides important clues of its subcellular localization. Although a series of artificial intelligence models have been reported to predict protein subcellular localization, most of them provide only textual outputs. Here, we present deepGPS, a deep generative model for protein subcellular localization prediction. After training with protein primary sequences and fluorescence images, deepGPS shows the ability to predict cytoplasmic and nuclear localizations by reporting both textual labels and generative images as outputs. In addition, cell-type-specific deepGPS models can be developed by using distinct image datasets from different cell lines for comparative analyses. Moreover, deepGPS shows potential to be further extended for other specific organelles, such as vesicles and endoplasmic reticulum, even with limited volumes of training data. Finally, the openGPS website (https://bits.fudan.edu.cn/opengps) is constructed to provide a publicly accessible and user-friendly platform for studying protein subcellular localization and function.

In the framework of unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC), the UAV’s trajectory design is crucial for the effectiveness of task offloading. However, existing studies suffer from some limitations such as reliance on static decision metrics, prediction-based strategies with poor generalization, or high computational complexity, making them unsuitable for large-scale emergency scenarios with surging task demands. To address this issue, this study introduces the Location Priority Level (LPL) and proposes LPL-based Bi-objective Task Offloading Strategy (LBTOS), which tries to simultaneously minimize task completion time (makespan) and total energy consumption. By jointly considering task load, regional deadlines, and UAV energy constraints, LBTOS introduces the LPL to dynamically map task urgency to UAV trajectories. Leveraging LPL and an adaptive triggering mechanism, the proposed UAV Trajectory Design algorithm (UAVTD) achieves high responsiveness under computationally constrained emergency scenarios. Additionally, a task offloading algorithm named HWPSO-SA is designed, which integrates adaptive and time-varying inertia weights, and combines the Particle Swarm Optimization with Simulated Annealing, thereby enhancing the algorithm’s capability to obtain the global optimal solution. Lastly, simulation experiments are conducted to validate the significant effectiveness of LBTOS in the dual optimization objectives of minimizing task completion time and the total computing and transmission energy consumption of the system.

Summary The Terminator, a specific DNA sequence, provides the transcriptional termination signal to RNA polymerase, making it a critical aspect of transcriptional regulation. This article proposes AMter, the first end‐to‐end model designed for predicting transcriptional terminators, leveraging attention mechanisms. In AMter, rather than manual feature engineering, two distinct modules based on attention mechanism, known as Frequency‐Attention and Allkmer‐Attention, are employed to automatically learn efficient features. Frequency‐Attention generates informative features by autonomously determining the significance of various frequency features, while Allkmer‐Attention aims to capture the relationships among all k ‐mers within a DNA sequence. Features generated by Frequency‐Attention and Allkmer‐Attention demonstrate high informativeness and discriminative capacity, allowing precise discrimination of whether a DNA sequence is a terminator through a simple prediction network. The results of the 5‐fold cross‐validation test indicate the remarkable achievement of our proposed method, attaining 100% accuracy in both the training and validation datasets. Furthermore, AMter demonstrates outstanding prediction accuracy on two independent datasets, with 100% accuracy for Escherichia coli and 99.30% for Bacillus subtilis , marking a significant 94.4% relative improvement over prior methods. Experimental results conclusively demonstrate that AMter surpasses existing approaches, establishing a new state‐of‐the‐art in transcriptional terminator prediction.