18 citations · 52 across the 18 of their papers we have counts for
7 papers · 1 filter
CoBa: Convergence Balancer for Multitask Finetuning of Large Language Models
Zi Gong, Hang Yu, Cong Liao +3
Multi-task learning (MTL) benefits the fine-tuning of large language models (LLMs) by providing a single model with improved performance and generalization ability across tasks, pr…
Rodimus*: Breaking the Accuracy-Efficiency Trade-Off with Efficient Attentions
Zhihao He, Hang Yu, Zi Gong +3
Recent advancements in Transformer-based large language models (LLMs) have set new standards in natural language processing. However, the classical softmax attention incurs signifi…
E2LLM: Encoder Elongated Large Language Models for Long-Context Understanding and Reasoning
Zihan Liao, Jun Wang, Hang Yu +3
Processing long contexts is increasingly important for Large Language Models (LLMs) in tasks like multi-turn dialogues, code generation, and document summarization. This paper addr…
GALLa: Graph Aligned Large Language Models for Improved Source Code Understanding
Ziyin Zhang, Hang Yu, Shijie Li +3
Programming languages possess rich semantic information - such as data flow - that is represented by graphs and not available from the surface form of source code. Recent code lang…
SQLfuse: Enhancing Text-to-SQL Performance through Comprehensive LLM Synergy
Tingkai Zhang, Chaoyu Chen, Cong Liao +6
Text-to-SQL conversion is a critical innovation, simplifying the transition from complex SQL to intuitive natural language queries, especially significant given SQL's prevalence in…
DUPLEX: Dual GAT for Complex Embedding of Directed Graphs
Zhaoru Ke, Hang Yu, Jianguo Li +1
Current directed graph embedding methods build upon undirected techniques but often inadequately capture directed edge information, leading to challenges such as: (1) Suboptimal re…
D2LLM: Decomposed and Distilled Large Language Models for Semantic Search
Zihan Liao, Hang Yu, Jianguo Li +2
The key challenge in semantic search is to create models that are both accurate and efficient in pinpointing relevant sentences for queries. While BERT-style bi-encoders excel in e…