2026
Lunar-R1: Towards Efficient Reasoning of Large Language Models for Lunar Exploration via Reinforcement Learning
Xin-Yu Xiao, Zhixian He, Shiqi Wang, Ye Tian, Qianchen Xia
Empirical Methods in Natural Language Processing (EMNLP) Industry Track 2026 Accepted
Lunar-R1 is an 8B reasoning model designed for lunar exploration tasks. It introduces Latent Difficulty Perception to estimate task difficulty and adapt the length of its reasoning process. This mechanism concentrates computation on difficult problems while avoiding unnecessary tokens on simpler ones. Experiments show a 38.9% reduction in token usage together with improved accuracy.
Lunar-R1: Towards Efficient Reasoning of Large Language Models for Lunar Exploration via Reinforcement Learning
Xin-Yu Xiao, Zhixian He, Shiqi Wang, Ye Tian, Qianchen Xia
Empirical Methods in Natural Language Processing (EMNLP) Industry Track 2026 Accepted
Lunar-R1 is an 8B reasoning model designed for lunar exploration tasks. It introduces Latent Difficulty Perception to estimate task difficulty and adapt the length of its reasoning process. This mechanism concentrates computation on difficult problems while avoiding unnecessary tokens on simpler ones. Experiments show a 38.9% reduction in token usage together with improved accuracy.
HyperLEGNN: Hypergraph Label-Element Neural Networks for Low-Resource Chinese Legal Event Detection
Xin-Yu Xiao
PRICAI 2026 2026 Accepted
HyperLEGNN is a hypergraph label-element neural network for low-resource Chinese legal event detection. It models relationships among event labels and their defining elements through a heterogeneous label-side hypergraph. This structure helps the model transfer information across rare and confusable event types. Experiments demonstrate improved detection performance in low-resource settings.
HyperLEGNN: Hypergraph Label-Element Neural Networks for Low-Resource Chinese Legal Event Detection
Xin-Yu Xiao
PRICAI 2026 2026 Accepted
HyperLEGNN is a hypergraph label-element neural network for low-resource Chinese legal event detection. It models relationships among event labels and their defining elements through a heterogeneous label-side hypergraph. This structure helps the model transfer information across rare and confusable event types. Experiments demonstrate improved detection performance in low-resource settings.
Lunar-Bench: Towards Evaluating Task-Oriented Reasoning of LLMs in Lunar Exploration Scenarios
Xin-Yu Xiao, Ye Tian, Erwei Yin, Zhixian He, Shiqi Wang, Yalei Liu, Qianchen Xia
Annual Meeting of the Association for Computational Linguistics (ACL) 2026 Published
Lunar-Bench is a 3,000-task benchmark for evaluating task-oriented reasoning and decision-making in lunar exploration scenarios. Each task tests whether a model can reason about a lunar situation and produce an appropriate decision. Environmental Scenario Indicators measure safety, efficiency, integrity, and alignment. The benchmark provides a structured basis for comparing the reliability and practical usefulness of lunar reasoning systems.
Lunar-Bench: Towards Evaluating Task-Oriented Reasoning of LLMs in Lunar Exploration Scenarios
Xin-Yu Xiao, Ye Tian, Erwei Yin, Zhixian He, Shiqi Wang, Yalei Liu, Qianchen Xia
Annual Meeting of the Association for Computational Linguistics (ACL) 2026 Published
Lunar-Bench is a 3,000-task benchmark for evaluating task-oriented reasoning and decision-making in lunar exploration scenarios. Each task tests whether a model can reason about a lunar situation and produce an appropriate decision. Environmental Scenario Indicators measure safety, efficiency, integrity, and alignment. The benchmark provides a structured basis for comparing the reliability and practical usefulness of lunar reasoning systems.
2025
Lunar Twins: We Choose to Go to the Moon with Large Language Models
Xin-Yu Xiao, Yalei Liu, Xiangyu Liu, Zengrui Li, Erwei Yin, Qianchen Xia
Annual Meeting of the Association for Computational Linguistics (ACL) 2025 Published
Lunar Twins introduces domain-specific large language models and data for lunar exploration. The system includes the Chang'e and Yutu models, together with a collaborative multi-agent workflow for generating and solving lunar tasks. It also establishes a specialized lunar dataset integrating information from Chang'e missions. Experiments show that the resulting models outperform comparable general-purpose models on lunar-domain tasks.
Lunar Twins: We Choose to Go to the Moon with Large Language Models
Xin-Yu Xiao, Yalei Liu, Xiangyu Liu, Zengrui Li, Erwei Yin, Qianchen Xia
Annual Meeting of the Association for Computational Linguistics (ACL) 2025 Published
Lunar Twins introduces domain-specific large language models and data for lunar exploration. The system includes the Chang'e and Yutu models, together with a collaborative multi-agent workflow for generating and solving lunar tasks. It also establishes a specialized lunar dataset integrating information from Chang'e missions. Experiments show that the resulting models outperform comparable general-purpose models on lunar-domain tasks.
Integrating Wavelet Transforms into Image Reconstruction Networks for Effective Style Transfer
Yunfei Chu, Xin-Yu Xiao, Longchen Han, Yaoshun Yue, Maohai Lin
Journal of Imaging Science and Technology 2025 Published
This paper presents an effective method for image style transfer by integrating wavelet transforms into whitening and coloring processes within image reconstruction networks. The proposed Wavelet Transfer Network (WTN) directly aligns the feature covariance of content and style images, yielding high-quality stylized outputs with enhanced efficiency and generalization. Experimental results demonstrate the superiority of WTN over existing methods in both arbitrary and photorealistic style transfer, setting a new benchmark in the field.
Integrating Wavelet Transforms into Image Reconstruction Networks for Effective Style Transfer
Yunfei Chu, Xin-Yu Xiao, Longchen Han, Yaoshun Yue, Maohai Lin
Journal of Imaging Science and Technology 2025 Published
This paper presents an effective method for image style transfer by integrating wavelet transforms into whitening and coloring processes within image reconstruction networks. The proposed Wavelet Transfer Network (WTN) directly aligns the feature covariance of content and style images, yielding high-quality stylized outputs with enhanced efficiency and generalization. Experimental results demonstrate the superiority of WTN over existing methods in both arbitrary and photorealistic style transfer, setting a new benchmark in the field.