Beyond Chain-of-Thought: A Survey of Chain-of-X Paradigms for LLMs by Y. Xia et al.
Different Approaches of CoX, for example,
Chain-of-Thought
Chain-of-Augmentation: Add external information into each step
Chain-of-Feedback: Provide human feedback into each step
Chain-of-Models: Implement different steps by different LLMs (Agents)
Paper
CHAIN-OF-EXPERTS: WHEN LLMS MEET COMPLEX OPERATIONS RESEARCH PROBLEMS by Z. Xiao et al.
Type: Chain-of-Models
Involved multiple LLMs or so-called agents to implement different steps of thought
Teaching Small Language Models to Reason for Knowledge-Intensive Multi-Hop Question Answering by X. Li et al.
Distillation by generating sub-questions from LLMs to train SLMs with CoT, called Decompose and Response Distillation
Chain-of-Probe: Examing the Necessity and Accuracy of CoT Step-by-Step by Z. Wang et al.
Early Answering :
LLMs have already predicted an answer before generating the CoT.
: Internalizing Symbolic Knowledge for Distilling Better CoT Capabilities into Small Language Models by H. Liao et al.
Teaching Small Language Models to Reason by L. C. Magister, J. Mallinson, J. Adamek, E. Malmi, and A. Severyn
Large Language Models Are Reasoning Teachers by N. Ho, L. Schmid, and S.-Y. Yun
To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoning by Z. Sprague et al.
On-device and Edge-cloud collaboration of LLM/Small
Survey
On-Device Language Models: A Comprehensive Review by J. Xu et al.
What is the Role of Small Models in the LLM Era: A Survey by L. Chen and G. Varoquaux
Paper
PerLLM: Personalized Inference Scheduling with Edge-Cloud Collaboration for Diverse LLM Services by Z. Yang, Y. Yang, C. Zhao, Q. Guo, W. He, and W. Ji
CoGenesis: A Framework Collaborating Large and Small Language Models for Secure Context-Aware Instruction Following by K. Zhang, J. Wang, E. Hua, B. Qi, N. Ding, and B. Zhou
Apple Intelligence Foundation Language Models by T. Gunter et al.
Decentralized LLM Inference over Edge Networks with Energy Harvesting by A. Khoshsirat, G. Perin, and M. Rossi
WDMoE: Wireless Distributed Large Language Models with Mixture of Experts by N. Xue et al.
EdgeShard: Efficient LLM Inference via Collaborative Edge Computing by M. Zhang, J. Cao, X. Shen, and Z. Cui
Crayon: Customized On-Device LLM via Instant Adapter Blending and Edge-Server Hybrid Inference by J. Bang, J. Lee, K. Shim, S. Yang, and S. Chang
LLMCad: Fast and Scalable On-device Large Language Model Inference by D. Xu et al.
SPA: Towards A Computational Friendly Cloud-Base and On-Devices Collaboration Seq2seq Personalized Generation by Y. Liu et al.
Hybrid SLM and LLM for Edge-Cloud Collaborative Inference by Z. Hao, H. Jiang, S. Jiang, J. Ren, and T. Cao
GKT: A Novel Guidance-Based Knowledge Transfer Framework For Efficient Cloud-edge Collaboration LLM Deployment by Y. Yao, Z. Li, and H. Zhao
Hybrid LLM: Cost-Efficient and Quality-Aware Query Routing by D. Ding et al.
Hybrid SLM and LLM for Edge-Cloud Collaborative Inference by Z. Hao, H. Jiang, S. Jiang, J. Ren, and T. Cao
Enhancing On-Device LLM Inference with Historical Cloud-Based LLM Interactions by Y. Ding, C. Niu, F. Wu, S. Tang, C. Lyu, and G. Chen
