English

Predicting Rewards Alongside Tokens: Non-disruptive Parameter Insertion for Efficient Inference Intervention in Large Language Model

Computation and Language 2024-08-21 v1

Abstract

Transformer-based large language models (LLMs) exhibit limitations such as generating unsafe responses, unreliable reasoning, etc. Existing inference intervention approaches attempt to mitigate these issues by finetuning additional models to produce calibration signals (such as rewards) that guide the LLM's decoding process. However, this solution introduces substantial time and space overhead due to the separate models required. This work proposes Non-disruptive parameters insertion (Otter), inserting extra parameters into the transformer architecture to predict calibration signals along with the original LLM output. Otter offers state-of-the-art performance on multiple demanding tasks while saving up to 86.5\% extra space and 98.5\% extra time. Furthermore, Otter seamlessly integrates with existing inference engines, requiring only a one-line code change, and the original model response remains accessible after the parameter insertion. Our code is publicly available at \url{https://github.com/chenhan97/Otter}

Keywords

Cite

@article{arxiv.2408.10764,
  title  = {Predicting Rewards Alongside Tokens: Non-disruptive Parameter Insertion for Efficient Inference Intervention in Large Language Model},
  author = {Chenhan Yuan and Fei Huang and Ru Peng and Keming Lu and Bowen Yu and Chang Zhou and Jingren Zhou},
  journal= {arXiv preprint arXiv:2408.10764},
  year   = {2024}
}

Comments

16 pages

R2 v1 2026-06-28T18:18:02.301Z