Zobrazeno 1 - 10
of 828
pro vyhledávání: '"LI, Junlong"'
Large language model pre-training has traditionally relied on human experts to craft heuristics for improving the corpora quality, resulting in numerous rules developed to date. However, these rules lack the flexibility to address the unique characte
Externí odkaz:
http://arxiv.org/abs/2409.17115
Autor:
Fan, Run-Ze, Li, Xuefeng, Zou, Haoyang, Li, Junlong, He, Shwai, Chern, Ethan, Hu, Jiewen, Liu, Pengfei
The quality of finetuning data is crucial for aligning large language models (LLMs) with human values. Current methods to improve data quality are either labor-intensive or prone to factual errors caused by LLM hallucinations. This paper explores ele
Externí odkaz:
http://arxiv.org/abs/2402.12219
As a relative quality comparison of model responses, human and Large Language Model (LLM) preferences serve as common alignment goals in model fine-tuning and criteria in evaluation. Yet, these preferences merely reflect broad tendencies, resulting i
Externí odkaz:
http://arxiv.org/abs/2402.11296
Critique, as a natural language description for assessing the quality of model-generated content, has played a vital role in the training, evaluation, and refinement of LLMs. However, a systematic method to evaluate the quality of critique is lacking
Externí odkaz:
http://arxiv.org/abs/2401.04518
In open-domain question-answering (ODQA), most existing questions require single-hop reasoning on commonsense. To further extend this task, we officially introduce open-domain multi-hop reasoning (ODMR) by answering multi-hop questions with explicit
Externí odkaz:
http://arxiv.org/abs/2310.13552
The rapid development of Large Language Models (LLMs) has substantially expanded the range of tasks they can address. In the field of Natural Language Processing (NLP), researchers have shifted their focus from conventional NLP tasks (e.g., sequence
Externí odkaz:
http://arxiv.org/abs/2310.05470
In recent years, significant progress has been made in video instance segmentation (VIS), with many offline and online methods achieving state-of-the-art performance. While offline methods have the advantage of producing temporally consistent predict
Externí odkaz:
http://arxiv.org/abs/2309.11857
Autor:
Sun, Hualei, Qiu, Liang, Han, Yifeng, Yi, Enkui, Li, Junlong, Huo, Mengwu, Huang, Chaoxin, Liu, Hui, Li, Manrong, Wang, Weiliang, Yao, Dao-Xin, Frandsen, Benjamin A., Shen, Bing, Hou, Yusheng, Wang, Meng
NiPSe3 is regarded as a bandwidth-controlled Mott insulator, distinct from the widely studied Mott insulating magnetic graphene MPSe3 (M = Mn and Fe) family. By employing high-pressure synchrotron X-ray diffraction, we observe two structural transiti
Externí odkaz:
http://arxiv.org/abs/2301.04329
Open-Domain Question Answering (ODQA) aims to answer questions without explicitly providing specific background documents. This task becomes notably challenging in a zero-shot setting where no data is available to train tailored retrieval-reader mode
Externí odkaz:
http://arxiv.org/abs/2212.08635
Publikováno v:
Shipin Kexue, Vol 45, Iss 5, Pp 126-136 (2024)
The volatile aroma compounds of wine fermented from goji berries were evaluated by liquid-liquid extraction-solvent-assisted evaporation-gas chromatography-mass spectrometry-olfactometry (LLE-SAFE-GC-MS-O) combined with GC-quadrupole-MS (GC-qMS) and
Externí odkaz:
https://doaj.org/article/346b921c34c7499dba330edab5f360e5
