Watermarking for Byzantine Robust LLM Training
cs.CR
Submitted: 2026-10-09
Updated: 2026-10-09
Project page: http://skylion007.github.io/OpenWebTextCorpus
Sources
- Finetuned Language Models Are Zero-Shot Learners
- A Survey on Federated Fine-tuning of Large Language Models
- The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI
- Robust Distortion-free Watermarks for Language Models
- On the Learnability of Watermarks for Language Models
- Recent Advances in Algorithmic High-Dimensional Robust Statistics
- Outlier-robust moment-estimation via sum-of-squares
- Qwen Technical Report
- LLaMA: Open and Efficient Foundation Language Models
- Stable Signature is Unstable: Removing Image Watermark from Diffusion Models
- Rethinking the Value of Network Pruning
- Comparing Rewinding and Fine-tuning in Neural Network Pruning
- The Pile: An 800GB Dataset of Diverse Text for Language Modeling
- WizardCoder: Empowering Code Large Language Models with Evol-Instruct
- Adaptive Federated Optimization
- Adam: A Method for Stochastic Optimization
- Measuring the Effects of Non-Identical Data Distribution for Federated Visual Classification
- Local AdaAlter: Communication-Efficient Stochastic Gradient Descent with Adaptive Learning Rates
- FedLALR: Client-Specific Adaptive Learning Rates Achieve Linear Speedup for Non-IID Data
- Local SGD With a Communication Overhead Depending Only on the Number of Workers
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills
- Token Composition: A Graph Based on EVM Logs