Rethinking LLM Ensembling from the Perspective of Mixture Models
This work addresses the computational inefficiency of LLM ensembling for practitioners, offering a practical speedup without sacrificing performance.
The paper proposes Mixture-model-like Ensemble (ME), which reinterprets LLM ensembling as a mixture model to stochastically select one model per token, achieving 1.78x-2.68x speedup over conventional ensemble while being mathematically equivalent.
Model ensembling is a well-established technique for improving the performance of machine learning models. Conventionally, this involves averaging the output distributions of multiple models and selecting the most probable label. This idea has been naturally extended to large language models (LLMs), yielding improved performance but incurring substantial computational cost. This inefficiency stems from directly applying conventional ensemble implementation to LLMs, which require a separate forward pass for each model to explicitly compute the ensemble distribution. In this paper, we propose the Mixture-model-like Ensemble (ME). By reinterpreting the ensemble as a mixture model, ME stochastically selects a single model at each step to generate the next token, thereby avoiding the need to explicitly compute the full ensemble distribution. ME is mathematically equivalent to sampling from the ensemble distribution, but requires invoking only one model, making it 1.78x-2.68x faster than conventional ensemble. Furthermore, this perspective connects LLM ensembling and token-level routing methods, suggesting that LLM ensembling is a special case of routing methods. Our findings open new avenues for efficient LLM ensembling and motivate further exploration of token-level routing strategies for LLMs. Our code is available at https://github.com/jialefu/Mixture-model-like-Ensemble/.