6 papers · 1 filter
Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMs
Yubo Gao, Haotian Wu, Hong Chen +8
Chain-of-Thought (CoT) has significantly enhanced LLM reasoning, yet often incurs substantial computational overhead due to "overthinking": generating excessively long rationales w…
EffiReason-Bench: A Unified Benchmark for Evaluating and Advancing Efficient Reasoning in Large Language Models
Junquan Huang, Haotian Wu, Yubo Gao +7
Large language models (LLMs) with Chain-of-Thought (CoT) prompting achieve strong reasoning but often produce unnecessarily long explanations, increasing cost and sometimes reducin…
EasyDistill: A Comprehensive Toolkit for Effective Knowledge Distillation of Large Language Models
Chengyu Wang, Junbing Yan, Wenrui Cai +2
In this paper, we present EasyDistill, a comprehensive toolkit designed for effective black-box and white-box knowledge distillation (KD) of large language models (LLMs). Our frame…
DistilQwen2.5: Industrial Practices of Training Distilled Open Lightweight Language Models
Chengyu Wang, Junbing Yan, Yuanhao Yue +1
Enhancing computational efficiency and reducing deployment costs for large language models (LLMs) have become critical challenges in various resource-constrained scenarios. In this…
Building a Family of Data Augmentation Models for Low-cost LLM Fine-tuning on the Cloud
Yuanhao Yue, Chengyu Wang, Jun Huang +1
Specializing LLMs in various domain-specific tasks has emerged as a critical step towards achieving high performance. However, the construction and annotation of datasets in specif…
Distilling Instruction-following Abilities of Large Language Models with Task-aware Curriculum Planning
Yuanhao Yue, Chengyu Wang, Jun Huang +1
Instruction tuning aims to align large language models (LLMs) with open-domain instructions and human-preferred responses. While several studies have explored autonomous approaches…