Zobrazeno 1 - 10
of 64
pro vyhledávání: '"Xia, Heming"'
Speculative decoding (SD) has emerged as a widely used paradigm to accelerate the inference of large language models (LLMs) without compromising generation quality. It works by first employing a compact model to draft multiple tokens efficiently and
Externí odkaz:
http://arxiv.org/abs/2410.06916
Tool learning aims to enhance and expand large language models' (LLMs) capabilities with external tools, which has gained significant attention recently. Current methods have shown that LLMs can effectively handle a certain amount of tools through in
Externí odkaz:
http://arxiv.org/abs/2406.17465
Autor:
Luo, Weiyao, Zheng, Suncong, Xia, Heming, Wang, Weikang, Lei, Yan, Liu, Tianyu, Chen, Shuang, Sui, Zhifang
Large language models (LLMs) have shown promising efficacy across various tasks, becoming powerful tools in numerous aspects of human life. However, Transformer-based LLMs suffer a performance degradation when modeling long-term contexts due to they
Externí odkaz:
http://arxiv.org/abs/2406.10985
Understanding the deep semantics of images is essential in the era dominated by social media. However, current research works primarily on the superficial description of images, revealing a notable deficiency in the systematic investigation of the in
Externí odkaz:
http://arxiv.org/abs/2402.11281
Autor:
Xia, Heming, Yang, Zhe, Dong, Qingxiu, Wang, Peiyi, Li, Yongqi, Ge, Tao, Liu, Tianyu, Li, Wenjie, Sui, Zhifang
To mitigate the high inference latency stemming from autoregressive decoding in Large Language Models (LLMs), Speculative Decoding has emerged as a novel decoding paradigm for LLM inference. In each decoding step, this method first drafts several fut
Externí odkaz:
http://arxiv.org/abs/2401.07851
Autor:
Tong, Shoujie, Xia, Heming, Dai, Damai, Xu, Runxin, Liu, Tianyu, Lin, Binghuai, Cao, Yunbo, Sui, Zhifang
Pretrained language models have achieved remarkable success in natural language understanding. However, fine-tuning pretrained models on limited training data tends to overfit and thus diminish performance. This paper presents Bi-Drop, a fine-tuning
Externí odkaz:
http://arxiv.org/abs/2305.14760
Recently, Large Language Models (LLMs) have been serving as general-purpose interfaces, posing a significant demand for comprehensive visual knowledge. However, it remains unclear how well current LLMs and their visually augmented counterparts (VaLMs
Externí odkaz:
http://arxiv.org/abs/2305.15028
Continual relation extraction (CRE) models aim at handling emerging new relations while avoiding catastrophically forgetting old ones in the streaming data. Though improvements have been shown by previous CRE studies, most of them only adopt a vanill
Externí odkaz:
http://arxiv.org/abs/2305.04636
Autor:
Dong, Qingxiu, Li, Lei, Dai, Damai, Zheng, Ce, Ma, Jingyuan, Li, Rui, Xia, Heming, Xu, Jingjing, Wu, Zhiyong, Liu, Tianyu, Chang, Baobao, Sun, Xu, Sui, Zhifang
With the increasing capabilities of large language models (LLMs), in-context learning (ICL) has emerged as a new paradigm for natural language processing (NLP), where LLMs make predictions based on contexts augmented with a few examples. It has been
Externí odkaz:
http://arxiv.org/abs/2301.00234
We study lossless acceleration for seq2seq generation with a novel decoding algorithm -- Aggressive Decoding. Unlike the previous efforts (e.g., non-autoregressive decoding) speeding up seq2seq generation at the cost of quality loss, our approach aim
Externí odkaz:
http://arxiv.org/abs/2205.10350