Siheng Li, Kejiao Li, Zenan Xu, Guanhua Huang, Kun Li, Haoyuan Wu, Wujiajia, Zihao Zheng, Chenchen Zhang, Kun Shi, Xue Gong, Qi Yi, Ruibin Xiong, Tingqiang Xu, Yuhao Jiang, Jianfeng Yan, Yuyuan Zeng, Guanghui Xu, Jinbao Xue, Zhijiang xu, Zheng Fang, Shuai LI, Qibin Liu, Xiaoxue Li, Zhuoyu Li, Yangyu Tao, Fei Gao, Cheng Jiang, Bochao Wang, Kai Liu, Jianchen Zhu, Wai Lam, Bo Zhou, Di Wang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Yichen He, Guanhua Huang, Peiyuan Feng, Yuan Lin, Yuchen Zhang, Hang Li, Weinan E. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Fine-grained geographic named entity recognition involves the identification of precise locations within geographic text. Existing methods face challenges in effectively addressing this task, namely: (1) the complexity and long-dependency nature of various fine-grained locations, which pose difficulties for recognition, and (2) the presence of numerous out-of-vocabulary fine-grained locations that are not present in the training data . In this paper, we propose a Multi-scale Siamese Reconstruction Network (MSRN) to tackle these challenges. To address the first challenge, MSRN employs a multi-scale convolutional network to aggregate interactions among coarse-grained and fine-grained span features. To tackle the second challenge, MSRN introduces a siamese reconstruction network that corrupts the representation and then reconstructs it, preventing the model from relying on rote memorization of biased locations in the training data . This approach enhances the model’s understanding of entity context. We conduct experiments on four real-world fine-grained geographic named entity recognition datasets, which contain a variety of complex long fine-grained locations and geographic entities not present in the training data. These experiments illustrate the challenges commonly encountered in real-world engineering applications and deliver a solid foundation for evaluating the robustness and applicability of our proposed method. The experimental results demonstrate the superiority of MSRN over previous models, achieving significant performance improvements ranging from 2.1 to 3.0. Further analysis confirms the effectiveness of MSRN in recognizing complex long-dependency entities, with a performance improvement of 3.0, as well as out-of-vocabulary locations, with a performance improvement of 6.6 compared to models with similar parameter sizes.

Sense of agency refers to the subjective perception of initiating and controlling one’s actions and their consequences within both individual and social contexts. It encompasses not only self-agency, where individuals feel in control of their actions from a first-person perspective, but also other-agency, where individuals comprehend the control exerted by others over their actions through mentalizing. This study explores the characteristics of implicit self-agency and other-agency in social interaction contexts using an interval estimation paradigm. The results indicate that voluntary actions lead to a temporal binding effect in individual contexts. Moreover, the reported intervals generated by the self in individual contexts are shorter than those in human-human and human-robot interaction contexts, suggesting a reduction in self-agency within social interaction contexts. Conversely, temporal intervals in human-human interactions are reported to be shorter than those in individual and human-robot interactions, indicating that other-agency remains unaffected. These findings highlight the distinct effects of social interaction contexts on self-agency and other-agency, offering novel empirical insights into human interaction behavior.

The widespread use of large language models (LLMs) has sparked concerns about the potential misuse of AI-generated text, as these models can produce content that closely resembles human-generated text.Current detectors for AI-generated text (AIGT) lack robustness against adversarial perturbations, with even minor changes in characters or words causing a reversal in distinguishing between humancreated and AI-generated text.This paper investigates the robustness of existing AIGT detection methods and introduces a novel detector, the Siamese Calibrated Reconstruction Network (SCRN).The SCRN employs a reconstruction network to add and remove noise from text, extracting a semantic representation that is robust to local perturbations.We also propose a siamese calibration technique to train the model to make equally confident predictions under different noise, which improves the model's robustness against adversarial perturbations.Experiments on four publicly available datasets show that the SCRN outperforms all baseline methods, achieving 6.5%-18.25%absolute accuracy improvement over the best baseline method under adversarial attacks.Moreover, it exhibits superior generalizability in crossdomain, cross-genre, and mixed-source scenarios.The code is available at https://github. com/CarlanLark/Robust-AIGC-Detector.

We introduce a novel reinforcement learning framework of LLM agents named AGILE (AGent that Interacts and Learns from Environments) designed to perform complex conversational tasks with users, leveraging LLMs, memory, tools, and interactions with experts. The agent possesses capabilities beyond conversation, including reflection, tool usage, and expert consultation. We formulate the construction of such an LLM agent as a reinforcement learning (RL) problem, in which the LLM serves as the policy model. We fine-tune the LLM using labeled data of actions and the PPO algorithm. We focus on question answering and release a dataset for agents called ProductQA, comprising challenging questions in online shopping. Our extensive experiments on ProductQA, MedMCQA and HotPotQA show that AGILE agents based on 7B and 13B LLMs trained with PPO can outperform GPT-4 agents. Our ablation study highlights the indispensability of memory, tools, consultation, reflection, and reinforcement learning in achieving the agent's strong performance. Datasets and code are available at https://github.com/bytarnish/AGILE.

In this work, we propose an Implicit Regularization Enhancement (IRE) framework to accelerate the discovery of flat solutions in deep learning, thereby improving generalization and convergence. Specifically, IRE decouples the dynamics of flat and sharp directions, which boosts the sharpness reduction along flat directions while maintaining the training stability in sharp directions. We show that IRE can be practically incorporated with *generic base optimizers* without introducing significant computational overload. Experiments show that IRE consistently improves the generalization performance for image classification tasks across a variety of benchmark datasets (CIFAR-10/100, ImageNet) and models (ResNets and ViTs). Surprisingly, IRE also achieves a $2\times$ *speed-up* compared to AdamW in the pre-training of Llama models (of sizes ranging from 60M to 229M) on datasets including Wikitext-103, Minipile, and Openwebtext. Moreover, we provide theoretical guarantees, showing that IRE can substantially accelerate the convergence towards flat minima in Sharpness-aware Minimization (SAM).

In document-level event extraction (DEE) tasks, a document typically contains many event records with multiple event roles.Therefore, accurately extracting all event records is a big challenge since the number of event records is not given.Previous works present the entitybased directed acyclic graph (EDAG) generation methods to autoregressively generate event roles, which requires a given generation order.Meanwhile, parallel methods are proposed to generate all event roles simultaneously, but suffer from the inadequate training which manifests zero accuracies on some event roles.In this paper, we propose an Iteratively Parallel Generation method with the Pre-Filling strategy (IPGPF).Event roles in an event record are generated in parallel to avoid order selection, and the event records are iteratively generated to utilize historical results.Experiments on two public datasets show our IPGPF improves 11.7 F1 than previous parallel models and up to 5.1 F1 than auto-regressive models under the control variable settings.Moreover, our enhanced IPGPF outperforms other entityenhanced models and achieves new state-ofthe-art performance 1 .* Work was done when Guanhua was an intern at ByteDance AI Lab.† Corresponding author. 1 Our code is available at https://github.com/ CarlanLark/IPGPF [S6] …, Jinggong Group increased its holdings of the company's stock by 182,038 shares through the secondary market on Dec 15, 2011,… [S7] …, the shares held by Jinggong Group in the company increased from 90,880,020 shares to 91,062,058 shares, … [S9] on Dec 16, 2011, Jinggong Group reduced its holdings of ... 35,000 shares, with an average price of 19.88.[S14] As of the date of this announcement, Jinggong Group holds 91,027,058 shares of the company, … EquityOverweight EquityHolder Jinggong Group

In this paper, we propose a novel Siamese Adversarial Network (SAN) architecture for automatic speech recognition, which aims at solving the difficulty of fuzzy audio recognition. Specifically, SAN constructs two sub-networks to differentiate the audio feature input and then introduces a loss to unify the output distribution of these sub-networks. Adversarial learning enables the network to capture more essential acoustic features and helps the models achieve better performance when encountering fuzzy audio input. We conduct numerical experiments with the SAN model on several datasets for the automatic speech recognition task. All experimental results show that the siamese adversarial nets significantly reduce the character error rate (CER). Specifically, we achieve a competitive 4.37 CER result on the AISHELL-1 dataset without any language models. To reveal the generality of the siamese adversarial net, we also conduct experiments on the phoneme recognition task, which also shows the superiority of the siamese adversarial network.

Network function virtualization (NFV) is a new way to provide services to users in a network. Different from dedicated hardware that realizes the network functions for an IoT application, the network function of an NFV network is executed on general servers, and in order to achieve complete network functions, service function chaining (SFC) chains virtual network functions to work together to support an IoT application. In this paper, we focus on a main challenge in this domain, i.e., resource efficient provisioning for social IoT application oriented SFC requests. We propose an online SFC deployment algorithm based on the layered strategies of physical networks and an evaluation of physical network nodes, which can efficiently reduce bandwidth resource consumption (OSFCD-LSEM) and support the security and privacy of social IoT applications. The results of our simulation show that our proposed algorithm improves the bandwidth carrying rate, time efficiency, and acceptance rate by 50%, 60%, and 15%, respectively.