paper-with-me

Papers

Incentivizing Permissionless Distributed Learning of LLMs

2025-05-27 · Joel Lidin, Amir Sarfi, Evangelos Pappas, Samuel Dare, Eugene Belilovsky, Jacob Steeves

We describe an incentive system for distributed deep learning of foundational models where peers are rewarded for contributions. The incentive system, \textit{Gauntlet}, has been deployed on the bittensor blockchain and used to train a 1.2B LLM with completely permissionless contributions of pseudo-gradients: no control over the users that can register or their hardware. \textit{Gauntlet} can be applied to any synchronous distributed training scheme that relies on aggregating updates or pseudo-gradients. We rely on a two-stage mechanism for fast filtering of peer uptime, reliability, and synchronization, combined with the core component that estimates the loss before and after individual pseudo-gradient contributions. We utilized an OpenSkill rating system to track competitiveness of pseudo-gradient scores across time. Finally, we introduce a novel mechanism to ensure peers on the network perform unique computations. Our live 1.2B run, which has paid out real-valued tokens to participants based on the value of their contributions, yielded a competitive (on a per-iteration basis) 1.2B model that demonstrates the utility of our incentive system.

📄 PDF Abstract BibTeX arXiv:2505.21684

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

A Distributed Computation Model Based on Federated Learning Integrates Heterogeneous models and Consortium Blockchain for Solving Time-Varying Problems

2023-06-28 · Zhihao Hao, Guancheng Wang, Chunwei Tian, Bob Zhang

The recurrent neural network has been greatly developed for effectively solving time-varying problems corresponding to complex environments. However, limited by the way of centralized processing, the model performance is…

Federated Learning

Panoptic: the perpetual, oracle-free options protocol

2022-04-27 · Guillaume Lambert, Jesper Kristensen

Panoptic is the perpetual, oracle-free, instant-settlement options trading protocol on the Ethereum blockchain. Panoptic enables the permissionless trading of options on top of any asset pool in the Uniswap v3 ecosystem …

Covenant-72B: Pre-Training a 72B LLM with Trustless Peers Over-the-Internet

2026-03-09 · Joel Lidin, Amir Sarfi, Erfan Miahi, Quentin Anthony 외 arxiv

Recently, there has been increased interest in globally distributed training, which has the promise to both reduce training costs and democratize participation in building large-scale foundation models. However, existing…

Incentivizing Reasoning from Weak Supervision

2025-05-26 · Yige Yuan, Teng Xiao, Shuchang Tao, Xue Wang 외

Large language models (LLMs) have demonstrated impressive performance on reasoning-intensive tasks, but enhancing their reasoning abilities typically relies on either reinforcement learning (RL) with verifiable signals o…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)

On the Day They Experience: Awakening Self-Sovereign Experiential AI Agents

2025-05-20 · Botao Amber Hu, Helena Rong

Drawing on Andrew Parker's "Light Switch" theory-which posits that the emergence of vision ignited a Cambrian explosion of life by driving the evolution of hard parts necessary for survival and fueling an evolutionary ar…