ICML 2025 Workshop ES-FoMo-III Submissions

Towards Understanding Self-Pretraining for Sequence Classification
Omar Coser, Antonio Orvieto
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Guided Speculative Inference for Efficient Test-Time Alignment of LLMs
Jonathan Geuter, Youssef Mroueh, David Alvarez-Melis
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III Spotlight
- Readers: Everyone
Kevin: Multi-Turn RL for Generating CUDA Kernels
Carlo Baronio, Pietro Marsella, Ben Pan, Simon Guo, Silas Alberti
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Privacy Isn’t Free: Benchmarking the Systems Cost of Privacy-Preserving ML
Nnaemeka Casmir Obiefuna, Samuel Oladayo Oyeneye, Similoluwa Odunaiya, Iremide Blessing Oyelaja, Steven Kolawole
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Training-Free Semantic Deferrals for Open-Ended LLM Cascades
Duncan Soiffer, Steven Kolawole, Virginia Smith
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Towards Large Scale Training on Apple Silicon
Tycho F. A. van der Ouderaa, Mohamed Baioumy, Matt Beton, Seth Howes, Gelu Vrabie, Alex Cheema
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Efficient and Accurate KV-cache Management for Long-Sequence LLMs
Yuzhen Mao, Qitong Wang, Martin Ester, Ke Li
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
One-Pass to Reason: Token Duplication and Block-Sparse Mask for Efficient Fine-Tuning on Multi-Turn Reasoning
Ritesh Goru, Shanay Mehta, Prateek Jain
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
GPU Kernel Scientist: An LLM-Driven Framework for Iterative Kernel Optimization
Martin Andrews, Sam Witteveen
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
How Many Tokens Do 3D Point Cloud Transformer Architectures Really Need?
Tuan Anh Tran, Duy Minh Ho Nguyen, Hoai-Chau Tran, Michael Barz, Khoa D Doan, Roger Wattenhofer, Vien Anh Ngo, Mathias Niepert, Daniel Sonntag, Paul Swoboda
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Multi-stream Sequence Learning
Mohamed Elsayed, A. Rupam Mahmood
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Mamba Drafters for Speculative Decoding
Daewon Choi, Seunghyuk Oh, Saket Dingliwal, Jihoon Tack, Kyuyoung Kim, Woomin Song, Seojin Kim, Insu Han, Jinwoo Shin, Aram Galstyan, Shubham Katiyar, Sravan Babu Bodapati
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Tail-Optimized Caching for LLM Inference
Wenxin Zhang, Yueying Li, Tianyi Peng, Ciamac C. Moallemi
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Quartet: Native FP4 Training Can Be Optimal for Large Language Models
Roberto L. Castro, Andrei Panferov, Rush Tabesh, Jiale Chen, Oliver Sieberling, Mahdi Nikdan, Saleh Ashkboos, Dan Alistarh
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III Spotlight
- Readers: Everyone
Think Clearly: Improving Reasoning via Redundant Token Pruning
Daewon Choi, Jimin Lee, Jihoon Tack, Woomin Song, Saket Dingliwal, Sai Muralidhar Jayanthi, Bhavana Ganesh, Jinwoo Shin, Aram Galstyan, Sravan Babu Bodapati
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Accelerated Test-Time Scaling with Model-Free Speculative Sampling
Woomin Song, Saket Dingliwal, Sai Muralidhar Jayanthi, Bhavana Ganesh, Jinwoo Shin, Aram Galstyan, Sravan Babu Bodapati
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
LATTICE: Learning to Efficiently Compress the Memory
Mahdi Karami, Vahab Mirrokni
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Compress, Gather, and Recompute: REFORMing Long-Context Processing in Transformers
Woomin Song, Sai Muralidhar Jayanthi, Srikanth Ronanki, Kanthashree Mysore Sathyendra, Jinwoo Shin, Aram Galstyan, Shubham Katiyar, Sravan Babu Bodapati
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Unified Scaling Laws for Compressed Representations
Andrei Panferov, Alexandra Volkova, Ionut-Vlad Modoranu, Vage Egiazarian, Mher Safaryan, Dan Alistarh
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Tiny Reward Models
Sarah Pan
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Context-lite Multi-turn Reinforcement Learning for LLM Agents
Wentse Chen, Jiayu Chen, Hao Zhu, Jeff Schneider
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
CarbonGearRL: Precision-Elastic, Carbon-Aware Scheduling for Foundation-Model Training
Thomas Y Chen
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Beyond Cosine Decay: On the effectiveness of Infinite Learning Rate Schedule for Continual Pre-training
Vaibhav Singh, Paul Janson, Paria Mehrbod, Adam Ibrahim, Irina Rish, Eugene Belilovsky, Benjamin Thérien
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Model Parallelism With Subnetwork Data Parallelism
Vaibhav Singh, Zafir Khalid, Eugene Belilovsky, Edouard Oyallon
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone
Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts
Haizhong Zheng, Yang Zhou, Brian R. Bartoldson, Bhavya Kailkhura, Fan Lai, Jiawei Zhao, Beidi Chen
- Published: 11 Jun 2025, Last Modified: 10 Jul 2025
- ES-FoMo III
- Readers: Everyone