近日，国际人工智能学术会议ICLR 2025论文接收结果公布，香港科技大学（广州）信息枢纽师生有43篇论文被录用。

ICLR，全称为「International Conference on Learning Representations」（国际学习表征会议），被认为是深度学习领域的顶级国际会议之一，关注有关深度学习各个方面的前沿研究，在人工智能、统计和数据科学领域以及机器视觉、语音识别、文本理解等重要应用领域中发布了众多极其有影响力的论文。会议具有广泛且深远的国际影响力，居谷歌学术人工智能会议影响力排行榜前列，与 NeurIPS、ICML 并称为机器学习领域三大顶会。今年，ICLR 2025共接收11,565份投稿，录用率为32.08%。ICLR 2025年会议将于4月24日至28日在新加坡举行，展示人工智能与深度学习领域的最新进展和突破性研究。

数据科学与分析学域被录用论文：

Bridging  and Modeling Correlations in Pairwise Data for Direct Preference Optimization    

Yuxin Jiang, Bo Huang, Yufei Wang, Xingshan Zeng, Liangyou Li, Yasheng Wang, Xin Jiang, Lifeng Shang, Ruiming Tang, Wei Wang

Influence-Guided Diffusion for Dataset Distillation

DSA学域作者：Mingyang Chen, Bo Huang, Wei Wang

STBLLM:  Breaking the 1-Bit Barrier with Structured Binary LLMs

Peijie Dong, Lujun Li, Yuedong Zhong, Dayou Du, Ruibo Fan, Yuhan Chen, Zhenheng Tang, Qiang Wang, Wei Xue, Yike Guo, Xiaowen Chu

AFlow: Automating Agentic Workflow Generation

Jiayi Zhang, Jinyu Xiang, Zhaoyang Yu, Fengwei Teng, Xionghui Chen, Jiaqi Chen, Mingchen Zhuge, Xin Cheng, Sirui Hong, Jinlin Wang, Bingnan Zheng, Bang Liu, Yuyu Luo, Chenglin Wu

You Only Sample Once: Taming One-Step Text-to-Image Synthesis by Self-Cooperative Diffusion GANs

Yihong Luo, Xiaolong Chen, Xinghua Qu, Tianyang Hu, Jing Tang

Decoupled Graph Energy-based Model for Node Out-of-Distribution Detection on Heterophilic Graphs

DSA学域作者：Yihong Luo, Yifan Song, Jing Tang

Structural-Entropy-Based Sample Selection For Efficient and Effective Learning

Tianchi Xie, Jiangning Zhu, Guozu Ma, Minzhi Lin, Wei Chen, Weikai Yang, Shixia Liu

N-ForGOT: Towards Not-forgetting and Generalization of Open Temporal Graph Learning

DSA学域作者：Liping Wang, Xujia Li, Lei Chen

Evaluating and Exploring Large Language Models on Graph Computation

DSA学域作者：Jianheng Tang, Qifan Zhang, Yuhan Li, Jia Li

Semi-Parametric Retrieval via Binary Bag-of-Tokens Index

Jiawei Zhou, Li Dong, Furu Wei, Lei Chen

Agent-Oriented Planning in Multi-Agent Systems

Ao Li, Yuexiang Xie, Songze Li, Fugee Tsung, Bolin Ding, Yaliang Li

Bridging General and Personalized Federated Learning through Selective Model Integration

DSA学域作者：Xiaowen Chu

The Lottery LLM Hypothesis, Rethinking What Abilities Should LLM Compression Preserve?

DSA学域作者：Xiang Liu, Peijie Dong, Xiaowen Chu

Improving Data Efficiency via Curating LLM-Driven Rating Systems

Jinlong Pang, Jiaheng Wei, Ankit Parag Shah, Zhaowei Zhu, Yaxuan Wang, Chen Qian, Yang Liu, Yujia Bao, Wei Wei

LLM Unlearning via Loss Adjustment with Only Forget Data

Yaxuan Wang, Jiaheng Wei, Chris Yuhao Liu, Jinlong Pang, Quan Liu, Ankit Parag Shah, Yujia Bao, Yang Liu, Wei Wei

CirT: Global Subseasonal-to-Seasonal Forecasting with Geometry-inspired Transformer

DSA学域作者：Yang Liu, Zinan Zheng, Jiashun Cheng, Fugee Tsung, Jia Li

InversionGNN: A Dual Path Network for Multi-Property Molecular Optimization

DSA学域作者：Yifan Niu, Ziqi Gao, Yang Liu, Jia Li

SimXRD-4M: Big Simulated X-ray Diffraction Data and Crystalline Symmetry Classification Benchmark

Bin Cao, Yang Liu, Zinan Zheng, Ruifeng Tan, Jia Li, Tong-yi Zhang

Diffusion Actor-Critic: Formulating Constrained Policy Iteration as Diffusion Noise Regression for Offline Reinforcement Learning

Linjiajie Fang, Ruoxue Liu, Jing Zhang, Wenjia Wang, Bing-Yi Jing

Knowledge Distillation with Multi-granularity Mixture of Priors for Image Super-Resolution

Simiao Li, Yun Zhang, Wei Li, Hanting Chen, Wenjia Wang, Bingyi Jing, Shaohui Lin, Jie Hu

AugKD: Ingenious Augmentations Empower Knowledge Distillation for Image Super-Resolution

DSA学域作者：Yun Zhang, Wenjia Wang

FormalAlign: Automated Alignment Evaluation for Autoformalization

DSA学域作者：Zhijiang Guo

Determine-Then-Ensemble: Necessity of Top-k Union for Large Language Model Ensembling

Yuxuan Yao, Han Wu, Mingyang Liu, Sichun Luo, Xiongwei Han, Jie Liu, Zhijiang Guo, Linqi Song

OptiBench Meets ReSocratic: Measure and Improve LLMs for Optimization Modeling

DSA学域作者：Zhicheng Yang, Zhijiang Guo, Jing Tang

Knowledge Benchmark Graph: Assisting Large Language Models in Designing Models by Retrieving Benchmark Knowledge

DSA学域作者：Fan Wang, Juyong Jiang, Jing Tang

Zigzag Diffusion Sampling: Diffusion Models Can Self-Improve via Self-Reflection

Lichen Bai, Shitong Shao, Zikai Zhou, Zipeng Qi, Zhiqiang Xu, Haoyi Xiong, Zeke Xie

IV-mixed Sampler: Leveraging Image Diffusion Model for Enhanced Video Synthesis

Shitong Shao, Zikai Zhou, Lichen Bai, Haoyi Xiong, Zeke Xie

Federated Residual Low-Rank Adaption of Large Language Models

DSA学域作者：Lei Zhu

On the Importance of Language-driven Representation Learning for Heterogeneous Federated Learning

DSA学域作者：Lei Zhu

Meissonic: Revitalizing Masked Generative Transformers for Efficient High-Resolution Text-to-Image Synthesis

DSA学域作者：Lei Zhu

Expand and Compress: Exploring Tuning Principles for Continual Spatio-Temporal Graph Forecasting

DSA学域作者：Yuxuan Liang

Towards Neural Scaling Laws for Time Series Foundation Models

Qingren Yao, Chao-Han Huck Yang, Renhe Jiang, Yuxuan Liang, Ming Jin, Shirui Pan

Open-CK: A Large Multi-Physics Fields Coupling benchmarks in Combustion Kinetics

DSA学域作者：Yuxuan Liang

Air Quality Prediction with Physics-Informed Dual Neural ODEs in Open Systems

Jindong Tian, Yuxuan Liang, Ronghui Xu, Peng Chen, Chenjuan Guo, Aoying Zhou, Lujia Pan, Zhongwen Rao, Bin Yang

KaSA: Knowledge-Aware Singular-Value Adaptation of Large Language Models

Fan Wang, Juyong Jiang, Chansung Park, Sunghun Kim, Jing Tang

人工智能学域被录用论文：

Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Prediction

Jing He*, Haodong Li*, Wei Yin, Yixun Liang*, Leheng Li*, Kaiqiang Zhou, Hongbo Zhang, Bingbing Liu, Ying-Cong Chen*

DisEnvisioner: Disentangled and Enriched Visual Prompt for Customized Image Generation

Jing He*, Haodong Li*, Yongzhe Hu, Guibao Shen*, Yingjie Cai, Weichao Qiu, Ying-Cong Chen*

Mitigating Modality Prior-Induced Hallucinations in Multimodal Large Language Models via Deciphering Attention Causality

Guanyu Zhou*, Yibo Yan*, Xin Zou*, Kun Wang, Aiwei Liu, Xuming Hu*(通讯作者)

Wasserstein-Regularized Conformal Prediction under General Distribution Shift

AI学域作者：Rui Xu, Sihong Xie

Explanations of GNN on Evolving Graphs via Axiomatic Layer edges

AI学域作者：Yazheng Liu, Sihong Xie

IV-mixed Sampler: Leveraging Image Diffusion Model for Enhanced Video Synthesis.

Shitong Shao*, Zikai Zhou*, Lichen Bai, Haoyi Xiong, and Zeke Xie*

Zigzag Diffusion Sampling: Diffusion Models Can Self-Improve via Self-Reflection.

Lichen Bai, Shitong Shao, Zikai Zhou, Zipeng Qi, Zhiqiang Xu, Haoyi Xiong, and Zeke Xie*

Gnothi Seauton: Empowering Faithful Self-Interpretability in Black-Box Models

Shaobo Wang, Hongxuan Tang, Mingyang Wang, Hongrui Zhang, Xuyang Liu, Weiya Li, Xuming Hu* (通讯作者), Linfeng Zhang

Can Watermarked LLMs be Identified by Users via Crafted Prompts?

Aiwei Liu, Sheng Guan, Yiming Liu, Leyi Pan, Yifei Zhang, Liancheng Fang, Lijie Wen, Philip S. Yu, Xuming Hu* (通讯作者)

Graph-Guided Scene Reconstruction from Images with 3D Gaussian Splatting

Chong Cheng, Gaochao Song, Yiyang Yao, Gangjian Zhang, Qinzheng Zhou, Hao Wang