ICLR

Overcoming False Illusions in Real-World Face Restoration with Multi-Modal Guided Diffusion Model

Keda Tao, Jinjin Gu, Yulun Zhang, Xiucheng Wang, Nan Cheng

机构 * Xidian University（西安电子科技大学） ； The University of Sydney（悉尼大学） ； Shanghai Jiao Tong University（上海交通大学） ； State Key Laboratory of Integrated Services Networks（集成服务网络国家重点实验室） 纠错

Comments 23 Pages, 28 Figures, ICLR 2025

Depth Pro: Sharp Monocular Metric Depth in Less Than a Second

Aleksei Bochkovskii, Amaël Delaunoy, Hugo Germain, Marcel Santos, Yichao Zhou, Stephan R. Richter, Vladlen Koltun

机构 * Apple（苹果公司） 纠错

Comments Published at ICLR 2025. Code and weights available at https://github.com/apple/ml-depth-pro

Training on the Test Task Confounds Evaluation and Emergence

Ricardo Dominguez-Olmedo, Florian E. Dorner, Moritz Hardt

机构 * Max Planck Institute for Intelligent Systems, Tübingen（马克斯·普朗克智能系统研究所，图宾根） ； Tübingen AI Center（图宾根人工智能中心） ； ETH Zurich（苏黎世联邦理工学院） 纠错

Comments ICLR 2025 (Oral)

Language Representations Can be What Recommenders Need: Findings and Potentials

Leheng Sheng, An Zhang, Yi Zhang, Yuxin Chen, Xiang Wang, Tat-Seng Chua

机构 * National University of Singapore（新加坡国立大学） ； University of Science and Technology of China（中国科学技术大学） 纠错

Comments ICLR 2025 (Oral). Codes are available at https://github.com/LehengTHU/AlphaRec

Inverse Constitutional AI: Compressing Preferences into Principles

Arduin Findeis, Timo Kaufmann, Eyke Hüllermeier, Samuel Albanie, Robert Mullins

机构 * University of Cambridge（剑桥大学） ； LMU Munich（慕尼黑大学） ； DFKI（德国达姆施塔特研究所） 纠错

Comments Accepted at ICLR 2025, v2 is camera-ready version; Main changes from v1: extended experiments, additional baselines

SLMRec: Distilling Large Language Models into Small for Sequential Recommendation

Wujiang Xu, Qitian Wu, Zujie Liang, Jiaojiao Han, Xuying Ning, Yunxiao Shi, Wenfang Lin, Yongfeng Zhang

机构 * Rutgers University（新泽西罗格斯大学） ； Broad Institute of MIT and Harvard（哈佛大学与麻省理工学院Broad研究所） ； Ant Group（蚂蚁集团） ； Dian Diagnostics Group Co.（迪安诊断集团） ； University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校） ； University of Technology Sydney（悉尼技术大学） 纠错

Comments International Conference on Learning Representations (ICLR 2025)

ProtPainter: Draw or Drag Protein via Topology-guided Diffusion

Zhengxi Lu, Shizhuo Cheng, Yuru Jiang, Yan Zhang, Min Zhang

机构 * Zhejiang University（浙江大学） 纠错

Comments Published as a conference paper at ICLR 2025

A synthetic dataset of French electric load curves with temperature conditioning

Tahar Nabil, Ghislain Agoua, Pierre Cauchois, Anne De Moliner, Benoît Grossin

机构 * EDF R&D（EDF研究院） ； Enedis 纠错

Comments Workshop paper at "Tackling Climate Change with Machine Learning", ICLR 2025

Flow to Learn: Flow Matching on Neural Network Parameters

Daniel Saragih, Deyu Cao, Tejas Balaji, Ashwin Santhosh

机构 * University of Toronto（多伦多大学） ； University of Tokyo（东京大学） 纠错

Comments Accepted at the ICLR Workshop on Neural Network Weights as a New Data Modality 2025

VACT: A Video Automatic Causal Testing System and a Benchmark

Haotong Yang, Qingyuan Zheng, Yunjian Gao, Yongkun Yang, Yangbo He, Zhouchen Lin, Muhan Zhang

Comments A preliminary version of this paper has been accepted by workshop SCSL@ICLR 2025

Unraveling Arithmetic in Large Language Models: The Role of Algebraic Structures

Fu-Chieh Chang, You-Chen Lin, Pei-Yuan Wu

Journal ref ICLR 2025 Workshop on Reasoning and Planning for Large Language Models

Magnetic Preference Optimization: Achieving Last-iterate Convergence for Language Model Alignment

Mingzhi Wang, Chengdong Ma, Qizhi Chen, Linjian Meng, Yang Han, Jiancong Xiao, Zhaowei Zhang, Jing Huo, Weijie J. Su, Yaodong Yang

机构 * Institute for Artificial Intelligence, Peking University（北京大学人工智能研究院） ； Beijing Academy of Artificial Intelligence（北京人工智能研究院） ； National Key Laboratory for Novel Software Technology, Nanjing University（南京大学新型软件技术国家重点实验室） ； China Telecom（中国电信） ； University of Pennsylvania（宾夕法尼亚大学） 纠错

Comments ICLR 2025

Reasoning-Enhanced Healthcare Predictions with Knowledge Graph Community Retrieval

Pengcheng Jiang, Cao Xiao, Minhao Jiang, Parminder Bhatia, Taha Kass-Hout, Jimeng Sun, Jiawei Han

Comments ICLR 2025 Camera-Ready

Learning Structured Representations by Embedding Class Hierarchy with Fast Optimal Transport

Siqi Zeng, Sixian Du, Makoto Yamada, Han Zhao

机构 * University of Illinois, Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校） ； Stanford University（斯坦福大学） ； Okinawa Institute of Science and Technology（冲绳科学技术研究所） 纠错

Comments 31 pages, 12 figures. Published as a conference paper at ICLR '25

Joint Graph Rewiring and Feature Denoising via Spectral Resonance

Jonas Linkerhägner, Cheng Shi, Ivan Dokmanić

机构 * Department of Mathematics and Computer Science（数学与计算机科学系） 纠错

Comments Accepted as oral at the 13th International Conference on Learning Representations (ICLR), 2025

OpenHands: An Open Platform for AI Software Developers as Generalist Agents

Xingyao Wang, Boxuan Li, Yufan Song, Frank F. Xu, Xiangru Tang, Mingchen Zhuge, Jiayi Pan, Yueqi Song, Bowen Li, Jaskirat Singh, Hoang H. Tran, Fuqiang Li, Ren Ma, Mingzhang Zheng, Bill Qian, Yanjun Shao, Niklas Muennighoff, Yizhe Zhang, Binyuan Hui, Junyang Lin, Robert Brennan, Hao Peng, Heng Ji, Graham Neubig

机构 * UIUC（伊利诺伊大学香槟分校） ； CMU（卡内基梅隆大学） ； Yale（耶鲁大学） ； UC Berkeley（加州大学伯克利分校） ； Contextual AI（上下文AI） ； KAUST（科威特大学） ； ANU（澳大利亚国立大学） ； HCMUT（胡志明市理工大学） ； Alibaba（阿里巴巴） ； All Hands AI（所有手AI） 纠错

Comments Accepted by ICLR 2025; Code: https://github.com/All-Hands-AI/OpenHands

Controlling Space and Time with Diffusion Models

Daniel Watson, Saurabh Saxena, Lala Li, Andrea Tagliasacchi, David J. Fleet

机构 * Google DeepMind（谷歌DeepMind） 纠错

Comments ICLR 2025, First three authors contributed equally

LiveBench: A Challenging, Contamination-Limited LLM Benchmark

Colin White, Samuel Dooley, Manley Roberts, Arka Pal, Ben Feuer, Siddhartha Jain, Ravid Shwartz-Ziv, Neel Jain, Khalid Saifullah, Sreemanti Dey, Shubh-Agrawal, Sandeep Singh Sandha, Siddartha Naidu, Chinmay Hegde, Yann LeCun, Tom Goldstein, Willie Neiswanger, Micah Goldblum

Comments ICLR 2025 Spotlight

Jailbreaking as a Reward Misspecification Problem

Zhihui Xie, Jiahui Gao, Lei Li, Zhenguo Li, Qi Liu, Lingpeng Kong

机构 * The University of Hong Kong（香港大学） ； Huawei Noah’s Ark Lab（华为诺亚实验室） 纠错

Comments Accepted to ICLR 2025. Code: https://github.com/zhxieml/remiss-jailbreak

Aligning Language Models with Demonstrated Feedback

Omar Shaikh, Michelle S. Lam, Joey Hejna, Yijia Shao, Hyundong Cho, Michael S. Bernstein, Diyi Yang

机构 * Stanford University（斯坦福大学） ； USC（美国南加州大学） 纠错

Comments ICLR 2025; 28 pages, 8 figures

Symmetry-Breaking Augmentations for Ad Hoc Teamwork

Ravi Hammond, Dustin Craggs, Mingyu Guo, Jakob Foerster, Ian Reid

机构 * Foerster Lab for AI Research, University of Oxford（牛津大学人工智能研究实验室） ； Australian Institute for Machine Learning, University of Adelaide（阿德莱德大学人工智能研究所） ； Meta AI Research, UK（英国Meta人工智能研究） ； Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学） 纠错

Comments 21 pages, 12 figures, Bidirectional Human-AI Alignment workshop, ICLR 2025

An Optimal Discriminator Weighted Imitation Perspective for Reinforcement Learning

Haoran Xu, Shuozhe Li, Harshit Sikchi, Scott Niekum, Amy Zhang

机构 * University of Texas at Austin（德克萨斯大学奥斯汀分校） ； UMass Amherst（马萨诸塞大学阿姆赫斯特分校） ； Meta AI 纠错

Comments ICLR 2025

SAR Object Detection with Self-Supervised Pretraining and Curriculum-Aware Sampling

Yasin Almalioglu, Andrzej Kucik, Geoffrey French, Dafni Antotsiou, Alexander Adam, Cedric Archambeau

Comments Accepted to ICLR 2025 ML4RS https://ml-for-rs.github.io/iclr2025/

Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model

Zhiwei Xu, Zhiyu Ni, Yixin Wang, Wei Hu

机构 * University of Michigan（密歇根大学） ； University of California, Berkeley（加州大学伯克利分校） 纠错

Comments ICLR 2025

On-Device Watermarking: A Socio-Technical Imperative For Authenticity In The Age of Generative AI

Houssam Kherraz

Comments 10 pages, 3 figures, ICLR 2025, https://openreview.net/forum?id=ygE0U21vxM

Diffusion Transformers for Tabular Data Time Series Generation

Fabrizio Garuti, Enver Sangineto, Simone Luetto, Lorenzo Forni, Rita Cucchiara

Comments Accepted at ICLR 2025. 26 pages, 19 figures, 13 tables

Adiabatic Fine-Tuning of Neural Quantum States Enables Detection of Phase Transitions in Weight Space

Vinicius Hernandes, Thomas Spriggs, Saqar Khaleefah, Eliska Greplova

Comments Accepted at the ICLR Workshop on Neural Network Weights as a New Data Modality 2025. Code available at: https://gitlab.com/QMAI/papers/nqsweights

SurFhead: Affine Rig Blending for Geometrically Accurate 2D Gaussian Surfel Head Avatars

Jaeseong Lee, Taewoong Kang, Marcel C. Bühler, Min-Jung Kim, Sungwon Hwang, Junha Hyung, Hyojin Jang, Jaegul Choo

Comments ICLR 2025, Project page with videos: https://summertight.github.io/SurFhead/

AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents

Maksym Andriushchenko, Alexandra Souly, Mateusz Dziemian, Derek Duenas, Maxwell Lin, Justin Wang, Dan Hendrycks, Andy Zou, Zico Kolter, Matt Fredrikson, Eric Winsor, Jerome Wynne, Yarin Gal, Xander Davies

Comments Accepted at ICLR 2025

Does Spatial Cognition Emerge in Frontier Models?

Santhosh Kumar Ramakrishnan, Erik Wijmans, Philipp Kraehenbuehl, Vladlen Koltun

Comments Published in ICLR 2025