ICLR

SciLitLLM: How to Adapt LLMs for Scientific Literature Understanding

Sihang Li, Jin Huang, Jiaxi Zhuang, Yaorui Shi, Xiaochen Cai, Mingjun Xu, Xiang Wang, Linfeng Zhang, Guolin Ke, Hengxing Cai

Comments ICLR 2025

ProteinGPT: Multimodal LLM for Protein Property Prediction and Structure Understanding

Yijia Xiao, Edward Sun, Yiqiao Jin, Qifan Wang, Wei Wang

Comments Spotlight, Machine Learning for Genomics Explorations @ ICLR 2025

Does Refusal Training in LLMs Generalize to the Past Tense?

Maksym Andriushchenko, Nicolas Flammarion

Comments Accepted at ICLR 2025. Updates in v2 and v3: added GPT-4o, Claude 3.5 Sonnet, o1-mini, and o1-preview results. Code and jailbreak artifacts: https://github.com/tml-epfl/llm-past-tense

VCR: A Task for Pixel-Level Complex Reasoning in Vision Language Models via Restoring Occluded Text

Tianyu Zhang, Suyuchen Wang, Lu Li, Ge Zhang, Perouz Taslakian, Sai Rajeswar, Jie Fu, Bang Liu, Yoshua Bengio

Comments Accepted at ICLR 2025. Original paper name: VCR: Visual Caption Restoration

Is In-Context Learning Sufficient for Instruction Following in LLMs?

Hao Zhao, Maksym Andriushchenko, Francesco Croce, Nicolas Flammarion

Comments Accepted at ICLR 2025. This camera-ready version v3 adds multi-turn alignment via ICL, revisiting main results on instruct models, and simple mechanistic study. Updates in the v2: experiment with decoding schemes, scaling in-context alignment, ICL vs IFT for instruction following. Code at https://github.com/tml-epfl/icl-alignment

SysCaps: Language Interfaces for Simulation Surrogates of Complex Systems

Patrick Emami, Zhaonan Li, Saumya Sinha, Truc Nguyen

Comments Accepted at ICLR 2025. 23 pages. Updated with final camera ready version

Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks

Maksym Andriushchenko, Francesco Croce, Nicolas Flammarion

Comments Accepted at ICLR 2025. Updates in the v3: GPT-4o and Claude 3.5 Sonnet results, improved writing. Updates in the v2: more models (Llama3, Phi-3, Nemotron-4-340B), jailbreak artifacts for all attacks are available, evaluation with different judges (Llama-3-70B and Llama Guard 2), more experiments (convergence plots, ablation on the suffix length for random search), examples of jailbroken generation

E(3)-equivariant models cannot learn chirality: Field-based molecular generation

Alexandru Dumitrescu, Dani Korpela, Markus Heinonen, Yogesh Verma, Valerii Iakovlev, Vikas Garg, Harri Lähdesmäki

Comments ICLR 2025

GRASP: GRAph-Structured Pyramidal Whole Slide Image Representation

Ali Khajegili Mirabadi, Graham Archibald, Amirali Darbandsari, Alberto Contreras-Sanz, Ramin Ebrahim Nakhli, Maryam Asadi, Allen Zhang, C. Blake Gilks, Peter Black, Gang Wang, Hossein Farahani, Ali Bashashati

Comments Accepted in Learning Meaningful Representations of Life (LMRL) Workshop at ICLR 2025

Generalization through variance: how noise shapes inductive biases in diffusion models

John J. Vastola

Comments Accepted to ICLR 2025

Journal ref The Thirteenth International Conference on Learning Representations, 2025. https://openreview.net/forum?id=7lUdo8Vuqa

On Linear Representations and Pretraining Data Frequency in Language Models

Jack Merullo, Noah A. Smith, Sarah Wiegreffe, Yanai Elazar

Comments ICLR 2025

Themisto: Jupyter-Based Runtime Benchmark

Konstantin Grotov, Sergey Titov

Comments Accepted to the third Deep Learning for Code (DL4C) workshop @ ICLR 2025

Data Metabolism: An Efficient Data Design Schema For Vision Language Model

Jingyuan Zhang, Hongzhi Zhang, Zhou Haonan, Chenxi Sun, Xingguang ji, Jiakang Wang, Fanheng Kong, Yahui Liu, Qi Wang, Fuzheng Zhang

Comments To be presented at ICLR 2025, First Workshop on Open Science for Foundation Models

Learning to Help in Multi-Class Settings

Yu Wu, Yansong Li, Zeyu Dong, Nitya Sathyavageeswaran, Anand D. Sarwate

Comments 30 pages, 7 figures, conference, ICLR 2025

Test-time Alignment of Diffusion Models without Reward Over-optimization

Sunwoo Kim, Minkyu Kim, Dongmin Park

Comments ICLR 2025 (Spotlight). The Thirteenth International Conference on Learning Representations. 2025

Multi-Field Adaptive Retrieval

Millicent Li, Tongfei Chen, Benjamin Van Durme, Patrick Xia

Comments ICLR 2025, Spotlight

Do Vision-Language Models Represent Space and How? Evaluating Spatial Frame of Reference Under Ambiguities

Zheyuan Zhang, Fengyuan Hu, Jayjun Lee, Freda Shi, Parisa Kordjamshidi, Joyce Chai, Ziqiao Ma

Comments Accepted to ICLR 2025 (Oral) | Project page: https://spatial-comfort.github.io/

Arithmetic Transformers Can Length-Generalize in Both Operand Length and Count

Hanseul Cho, Jaeyoung Cha, Srinadh Bhojanapalli, Chulhee Yun

Comments 44 pages, 20 figures, 26 tables, accepted to ICLR 2025

Systems with Switching Causal Relations: A Meta-Causal Perspective

Moritz Willig, Tim Nelson Tobiasch, Florian Peter Busch, Jonas Seng, Devendra Singh Dhami, Kristian Kersting

Comments 21 pages, 3 figures, 4 tables, ICLR 2025 Camera Ready Version

Evaluating Semantic Variation in Text-to-Image Synthesis: A Causal Perspective

Xiangru Zhu, Penglei Sun, Yaoxian Song, Yanghua Xiao, Zhixu Li, Chengyu Wang, Jun Huang, Bei Yang, Xiaoxiao Xu

Comments Accepted by ICLR 2025

Presto! Distilling Steps and Layers for Accelerating Music Generation

Zachary Novack, Ge Zhu, Jonah Casebeer, Julian McAuley, Taylor Berg-Kirkpatrick, Nicholas J. Bryan

Comments Accepted as Spotlight at ICLR 2025

Visually Consistent Hierarchical Image Classification

Seulki Park, Youren Zhang, Stella X. Yu, Sara Beery, Jonathan Huang

Comments Accepted to ICLR 2025

Generating Pragmatic Examples to Train Neural Program Synthesizers

Saujas Vaduguru, Daniel Fried, Yewen Pu

Comments ICLR 2024

FLIP Reasoning Challenge

Andreas Plesner, Turlan Kuzhagaliyev, Roger Wattenhofer

Comments Published at First Workshop on Open Science for Foundation Models at ICLR 2025

Leave-One-Out Stable Conformal Prediction

Kiljae Lee, Yuan Zhang

Comments Accepted at ICLR 2025

CoMotion: Concurrent Multi-person 3D Motion

Alejandro Newell, Peiyun Hu, Lahav Lipson, Stephan R. Richter, Vladlen Koltun

Comments Accepted at ICLR 2025, for code and weights go to https://github.com/apple/ml-comotion

Synthetic Data for Blood Vessel Network Extraction

Joël Mathys, Andreas Plesner, Jorel Elmiger, Roger Wattenhofer

Comments Presented at SynthData Workshop at ICLR 2025

GrabS: Generative Embodied Agent for 3D Object Segmentation without Scene Supervision

Zihui Zhang, Yafei Yang, Hongtao Wen, Bo Yang

Comments ICLR 2025 Spotlight. Code and data are available at: https://github.com/vLAR-group/GrabS

Neural ODE Transformers: Analyzing Internal Dynamics and Adaptive Fine-tuning

Anh Tong, Thanh Nguyen-Tang, Dongeun Lee, Duc Nguyen, Toan Tran, David Hall, Cheongwoong Kang, Jaesik Choi

Comments ICLR 2025

Geometry-aware RL for Manipulation of Varying Shapes and Deformable Objects

Tai Hoang, Huy Le, Philipp Becker, Vien Anh Ngo, Gerhard Neumann

Comments Accepted at ICLR 2025 (Oral)