Large Language Models have shown growing ability to generate fluent and coherent texts that are highly similar to the writing style of humans. Current detectors for Machine-Generated Text (MGT) perform well when they are trained and tested in the same domain but generalize poorly to unseen domains, due to domain shift between data from different sources. In this work, we propose MGT-Prism , an MGT detection method from the perspective of the frequency domain for better domain generalization. Our key insight stems from analyzing text representations in the frequency domain, where we observe consistent spectral patterns across diverse domains, while significant discrepancies in magnitude emerge between MGT and human-written texts (HWTs). The observation initiates the design of a low frequency domain filtering module for filtering out the document-level features that are sensitive to domain shift, and a dynamic spectrum alignment strategy to extract the task-specific and domain-invariant features for improving the detector's performance in domain generalization. Extensive experiments demonstrate that MGT-Prism outperforms state‑of‑the‑art baselines by an average of 0.90% in accuracy and 0.92% in F1 score on 11 test datasets across three domain‑generalization scenarios.

Chengzhengxu Li, Xiaoming Liu, Zhaohan Zhang, Shengchao Liu, Guoxin Ma, Yu Lan, Cong Wang, Chao Shen. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Accurately spatio-temporal anomaly event inference is significant to enhance society’s safety, such as crime prevention and traffic collision reduction, etc. However, it is hard to achieve good performance for its complicated process being influenced by various kinds of factors. Previous works mainly focus on employing feature-based regression or fitting models with supposed spatio-temporal distribution to tackle this challenge, but are normally short of the following considerations: 1) mutual evolutionary influence, i.e., the dynamic evolution of interactions and dependencies among anomaly events, is changing along with the timeline and alters the probabilities or patterns of their occurrences dynamically; 2) messy features, i.e., complex attributes in the data but with noise, are difficult to select and aggregate for the representation learning algorithm with uncertainty and redundancy. To address the research gap, we put forward a co-evolutionary graph convolutional network model to explicate the dynamically spatio-temporal anomaly patterns. Specifically, we firstly take advantage of a fuzzy-rough set-based algorithm to select features by discovering the specialty and permanence attributes of different interaction features. Then, we propose a co-evolutionary learning method to embed the dynamically temporal influence into latent features with selected interaction information. Finally, we design a graph convolutional network with an attention mechanism to formulate the mutually spatial effects among the anomaly events. The proposed model is verified on the New York City crime records from the real-world, and extensive experiments show that our approach achieves 0.10129, 0.09958 and 0.10034 of the MAE (hour) in the action, location, and action-location time inference tasks, and 0.7973 and 0.4678 of the accuracy in the action and location type inference tasks, which outperform state-of-arts by 24.00%, 50.60%, 11.10%, 10.56% and 21.53% at most, respectively. Hyper-parameter and ablation experiments are also carried out further to demonstrate the sensitivity and effectiveness of our model.

Prompt-based pre-trained language models (PLMs) paradigm has succeeded substantially in few-shot natural language processing (NLP) tasks. However, prior discrete prompt optimization methods require expert knowledge to design the base prompt set and identify high-quality prompts, which is costly, inefficient, and subjective. Meanwhile, existing continuous prompt optimization methods improve the performance by learning the ideal prompts through the gradient information of PLMs, whose high computational cost, and low readability and generalizability are often concerning. To address the research gap, we propose a Dialogue-comprised Policy-gradient-based Discrete Prompt Optimization (DP_2O) method. We first design a multi-round dialogue alignment strategy for readability prompt set generation based on GPT-4. Furthermore, we propose an efficient prompt screening metric to identify high-quality prompts with linear complexity. Finally, we construct a reinforcement learning (RL) framework based on policy gradients to match the prompts to inputs optimally. By training a policy network with only 0.62M parameters on the tasks in the few-shot setting, DP_2O outperforms the state-of-the-art (SOTA) method by 1.52% in accuracy on average on four open-source datasets. Moreover, subsequent experiments also demonstrate that DP_2O has good universality, robustness and generalization ability.

Shengchao Liu, Xiaoming Liu, Yichen Wang, Zehua Cheng, Chengzhengxu Li, Zhaohan Zhang, Yu Lan, Chao Shen. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Recent advances in prompt optimization have notably enhanced the performance of pre-trained language models (PLMs) on downstream tasks. However, the potential of optimized prompts on domain generalization has been under-explored. To explore the nature of prompt generalization on unknown domains, we conduct pilot experiments and find that (i) Prompts gaining more attention weight from PLMs’ deep layers are more generalizable and (ii) Prompts with more stable attention distributions in PLMs’ deep layers are more generalizable. Thus, we offer a fresh objective towards domain-generalizable prompts optimization named ''Concentration'', which represents the ''lookback'' attention from the current decoding token to the prompt tokens, to increase the attention strength on prompts and reduce the fluctuation of attention distribution. We adapt this new objective to popular soft prompt and hard prompt optimization methods, respectively. Extensive experiments demonstrate that our idea improves comparison prompt optimization methods by 1.42% for soft prompt generalization and 2.16% for hard prompt generalization in accuracy on the multi-source domain generalization setting, while maintaining satisfying in-domain performance. The promising results validate the effectiveness of our proposed prompt optimization objective and provide key insights into domain-generalizable prompts.

The robust unit commitment (UC) is of paramount importance for achieving reliable operations considering the uncertainty of renewable realizations. The typical affine decision rule method and the robust feasible region method may achieve uneconomic dispatches as the dispatch decisions just rely on the current-stage information. Through approximating the future cost-to-go functions, the dual dynamic programming based methods have been shown adaptive to the multistage robust optimization problems, while suffering from high computational complexity. Thus, we propose the robust approximate dynamic programming (RADP) method to promote the computational speed and the economic performance for large-scale robust UC problems. RADP initializes the candidate points for guaranteeing the feasibility of upper bounding the value functions, solves the alternating calculation based bilinear programming to obtain the worst cases, and combines the primal and dual updates for the two-phase robust UC decision-making problem to achieve fast convergence. The finite termination guarantee of the RADP method is verified by the analyses for the multistage robust optimization problems with achieving suboptimal solutions. Numerical tests on 118-bus and 2383-bus transmission systems have demonstrated that RADP can approach the suboptimal economic performance at significantly improved computational efficiency.Note to Practitioners—This paper was motivated by solving the large-scale robust UC problem embedded with the multistage economic dispatch with improved computational and economical performance. Compared to the existing methods, this work suggests a RADP-based approach, which is inspired by the robust dual dynamic programming (RDDP) scheme. The proposed RADP owns lower computational complexity when compared with RDDP. As the problem in this work is formulated in a general form, the proposed RADP can be applied to other robust optimization problems such as the inventory management problem with uncertain demands. To apply the method to large-scale decision-making problems, one needs to solve linear programming problems to obtain the finite upper/lower bounds first, and then initialize the upper-bound points based on the feasible region limits of the decision variables. Conduct the forward pass to generate the candidate points and the backward pass to refine the cost-to-go functions. If the application problems have discrete decision variables as in the two-phase UC, one can solve the nonanticipativity constrained problem to obtain discrete solutions before doing the RADP scheme. The analyses and numerical experiments suggest that this approach can achieve suboptimal solutions. In future research, we will address the accelerated multistage robust decision-making with achieving optimal solutions.

Machine-Generated Text (MGT) detection, a task that discriminates MGT from Human-Written Text (HWT), plays a crucial role in preventing misuse of text generative models, which excel in mimicking human writing style recently.The latest proposed detectors usually take coarse text sequences as input and finetune pre-trained models with standard crossentropy loss.However, these methods fail to consider the linguistic structure of texts.Moreover, they lack the ability to handle the lowresource problem, which could often happen in practice considering the enormous amount of textual data online.In this paper, we present a coherence-based contrastive learning model named COCO to detect the possible MGT under the low-resource scenario.To exploit the linguistic feature, we encode coherence information in the form of graph into the text representation.To tackle the challenges of low data resources, we employ a contrastive learning framework and propose an improved contrastive loss for preventing performance degradation brought by simple samples.The experiment results on two public datasets and two self-constructed datasets prove our approach outperforms the state-of-the-art methods significantly.Also, we surprisingly find that MGTs originated from up-to-date language models could be easier to detect than these from previous models, in our experiments.And we propose some preliminary explanations for this counter-intuitive phenomena.All the codes and datasets are open-sourced.1

Microgrid usually associated with distributed renewable energy sources plays a key role in reducing the requirement of traditional power sources and the emission of CO2. Connecting multiple microgrids to the electric power grid can facilitate a more powerful and reliable operation. Microgrids need decentralized dispatch with the requirement of security and privacy, and at the same time, they need to cooperate to satisfy the constraints from the power grid. We consider online decentralized and cooperative dispatch problem in this paper. In order to capture the randomness in the wind and solar power generation and the system dynamics, we formulate the problem as a decentralized Markov decision process (Dec-MDP). We propose a new method to calculate the approximate Q-value functions by constructing the initial feasible policy with reasonable space and time complexity. Bayesian games (BGs) is applied to model the decentralized dispatch with delayed communication in order not to slow down the online decision-making. Also, the alternating minimization method is used to solve this BGs based on the approximate Q-value functions. Numerical results demonstrate that the proposed algorithm is very effective in reducing the state space and computing time and achieving the requirement of real-time scheduling. Motivated by the problem of decentralized dispatch of multi-microgrids in grid-connected mode, we consider the online decentralized operation and pay special attention to the grid tie-line power fluctuation. This problem is challenging due to the randomness of the renewable power generation, the global constraints about multiple microgrids, and the correlation among stages. We model the decentralized dispatch problem as a rolling horizon Dec-MDP and provide an approximate solution method, namely, alternately minimizing among different microgrids at the online stage and calculating approximate value functions in the offline preparation process. The approximate value functions make the online decision turn out to be a single-stage decentralized decision problem. Online decision calculation based on delayed communication is to guarantee all the microgrids to start conducting an action immediately while receiving the information related to them. We hope this paper brings insight onto the tie-line power fluctuation stabilization while multiple microgrids connecting to the same point of interconnection.