most citedRouting to the Expert: Efficient Reward-guided Ensemble of Large Language Models

4 citations · 10 across the 5 of their papers we have counts for

collaborators

9 papers

cs.CL2024

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

Chenhan Yuan, Fei Huang, Ru Peng +4

Transformer-based large language models (LLMs) exhibit limitations such as generating unsafe responses, unreliable reasoning, etc. Existing inference intervention approaches attemp…

cs.CL2024

Online Merging Optimizers for Boosting Rewards and Mitigating Tax in Alignment

Keming Lu, Bowen Yu, Fei Huang +3

Effectively aligning Large Language Models (LLMs) with human-centric values while preventing the degradation of abilities acquired through Pre-training and Supervised Fine-tuning (…

cs.CL20241 cited

Large Language Models are Superpositions of All Characters: Attaining Arbitrary Role-play via Self-Alignment

Keming Lu, Bowen Yu, Chang Zhou +1

Considerable efforts have been invested in augmenting the role-playing proficiency of open-source large language models (LLMs) by emulating proprietary counterparts. Nevertheless,…

cs.CL20234 cited

Routing to the Expert: Efficient Reward-guided Ensemble of Large Language Models

Keming Lu, Hongyi Yuan, Runji Lin +4

The complementary potential of Large Language Models (LLM) assumes off-the-shelf LLMs have heterogeneous expertise in a wide range of domains and tasks so that an ensemble of LLMs…

cs.CL20231 cited

Self-Evolved Diverse Data Sampling for Efficient Instruction Tuning

Shengguang Wu, Keming Lu, Benfeng Xu +3

Enhancing the instruction-following ability of Large Language Models (LLMs) primarily demands substantial instruction-tuning datasets. However, the sheer volume of these imposes a…

cs.CL2023

Speculative Contrastive Decoding

Hongyi Yuan, Keming Lu, Fei Huang +2

Large language models~(LLMs) exhibit exceptional performance in language tasks, yet their auto-regressive inference is limited due to high computational requirements and is sub-opt…