Tianyuan Shi
YOU?
Author Swipe
View article: Lookahead Routing for Large Language Models
Lookahead Routing for Large Language Models Open
Large language model (LLM) routers improve the efficiency of multi-model systems by directing each query to the most appropriate model while leveraging the diverse strengths of heterogeneous LLMs. Most existing approaches frame routing as …
View article: Mutual-Taught for Co-adapting Policy and Reward Models
Mutual-Taught for Co-adapting Policy and Reward Models Open
During the preference optimization of large language models (LLMs), distribution shifts may arise between newly generated model samples and the data used to train the reward model (RM). This shift reduces the efficacy of the RM, which in t…
View article: FuseRL: Dense Preference Optimization for Heterogeneous Model Fusion
FuseRL: Dense Preference Optimization for Heterogeneous Model Fusion Open
Heterogeneous model fusion enhances the performance of LLMs by integrating the knowledge and capabilities of multiple structurally diverse models. However, existing approaches often rely solely on selecting the best output for each prompt …
View article: Layer-Specific Scaling of Positional Encodings for Superior Long-Context Modeling
Layer-Specific Scaling of Positional Encodings for Superior Long-Context Modeling Open
Although large language models (LLMs) have achieved significant progress in handling long-context inputs, they still suffer from the ``lost-in-the-middle'' problem, where crucial information in the middle of the context is often underrepre…
View article: Mutual-Taught for Co-adapting Policy and Reward Models
Mutual-Taught for Co-adapting Policy and Reward Models Open
View article: Weighted-Reward Preference Optimization for Implicit Model Fusion
Weighted-Reward Preference Optimization for Implicit Model Fusion Open
While fusing heterogeneous open-source LLMs with varying architectures and sizes can potentially integrate the strengths of different models, existing fusion methods face significant challenges, such as vocabulary alignment and merging dis…
View article: Searching for Best Practices in Retrieval-Augmented Generation
Searching for Best Practices in Retrieval-Augmented Generation Open
Retrieval-augmented generation (RAG) techniques have proven to be effective in integrating up-to-date information, mitigating hallucinations, and enhancing response quality, particularly in specialized domains. While many RAG approaches ha…
View article: Towards Biologically Plausible Computing: A Comprehensive Comparison
Towards Biologically Plausible Computing: A Comprehensive Comparison Open
Backpropagation is a cornerstone algorithm in training neural networks for supervised learning, which uses a gradient descent method to update network weights by minimizing the discrepancy between actual and desired outputs. Despite its pi…
View article: Development of an osteosarcoma model with <i>MYCN</i> amplification and <i>TP53</i> mutation in hiPS cell‐derived neural crest cells
Development of an osteosarcoma model with <i>MYCN</i> amplification and <i>TP53</i> mutation in hiPS cell‐derived neural crest cells Open
Mesenchymal stem cell‐ or osteoblast‐derived osteosarcoma is the most common malignant bone tumor. Its highly metastatic malignant phenotypes, which are often associated with a poor prognosis, have been correlated with the modulation of TP…
View article: The Effects of Documentaries about Wild Life on the Awareness of Environmental Protection of Chinese Primary School Students
The Effects of Documentaries about Wild Life on the Awareness of Environmental Protection of Chinese Primary School Students Open
As far as is known, there are already countless studies around the world about how to help improve the willingness of citizens to protect the environment.However, the research focusing on whether the documentaries of wild life can enhance …
View article: Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation
Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation Open
Comprehensive characterization of the PtAP genes suggests their functional diversity during Populus growth and development. Our findings provide an overall understanding of the AP gene family in trees and establish a better foundation to f…
View article: Additional file 12 of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation
Additional file 12 of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation Open
Table S9. All primers used in this study. (XLSX 10 kb)
View article: Additional file 3: of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation
Additional file 3: of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation Open
Table S2. The 22 PtAP-like genes eliminated in this study. (XLSX 14 kb)
View article: Additional file 4: of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation
Additional file 4: of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation Open
Table S3. Prediction analysis of N- and O-glycosylation sites in the 67 PtAPs. (XLSX 43 kb)
View article: Additional file 5: of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation
Additional file 5: of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation Open
Table S4. Prediction analysis of subcellular localization of the 67 PtAPs. (XLSX 36 kb)
View article: Additional file 11: of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation
Additional file 11: of Genome-wide characterization of aspartic protease (AP) gene family in Populus trichocarpa and identification of the potential PtAPs involved in wood formation Open
Table S8. The eFP values of 57 PtAPs in six tissues downloaded from the Poplar eFP browser. (XLSX 13 kb)