LGAICLSep 25, 2025

HAMMER: Hamiltonian Curiosity Augmented Large Language Model Reinforcement

arXiv:2509.25240v1h-index: 3
Originality Incremental advance
AI Analysis

This addresses a specific bottleneck in LLM reinforcement learning, offering an incremental improvement over existing curriculum methods.

The paper tackles the problem of local optimization in curriculum reinforcement learning for large language models by proposing HAMMER, which orders training samples via a minimum-semantic Hamiltonian path to retain exploration, resulting in a 3% to 4% average accuracy gain across diverse inference benchmarks.

Recent curriculum reinforcement learning for large language models (LLMs) typically rely on difficulty-based annotations for data filtering and ordering. However, such methods suffer from local optimization, where continual training on simple samples in the early steps can cause the policy to lose its exploration. We propose a novel schema, namely Hamiltonian curiosity augmented large language model reinforcement (HAMMER), that transfers diversity metrics, commonly used in dataset evaluation, into the dynamic reinforcement learning procedure, where training samples are ordered via a minimum-semantic Hamiltonian path making the initial training retrain more exploration. From a theoretical perspective of generalization bounds, diversity-driven ordering facilitates stable convergence. Empirical evaluations indicate that HAMMER stimulates model "curiosity" and consistently achieves a 3% to 4% average accuracy gain across diverse inference benchmark.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes