ICLR 2025 Top Papers
Analysis of the most influential research by citation metrics
4,434
Total Citations
1,111
Highest Citation Count
221.7
Average Citations
1
KAN: Kolmogorov–Arnold Networks
1111
2
SAM 2: Segment Anything in Images and Videos
795
3
WizardMath: Empowering Mathematical Reasoning for Large Language Models via Reinforced Evol-Instruct
411
4
CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer
366
5
LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
209
6
GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models
201
7
LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Models
177
8
Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks
154
9
MediConfusion: Can you trust your AI radiologist? Probing the reliability of multimodal medical foundation models
146
10
NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models
144
11
Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model
141
12
Scaling and evaluating sparse autoencoders
133
13
Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers
118
14
Generative Representational Instruction Tuning
110
15
BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions
107
16
Training Language Models to Self-Correct via Reinforcement Learning
105
16
CameraCtrl: Enabling Camera Control for Text-to-Video Generation
105
18
Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing
104
19
JudgeLM: Fine-tuned Large Language Models are Scalable Judges
103
20
Generative Verifiers: Reward Modeling as Next-Token Prediction
102