New methods boost LLM performance by compressing token input.
Runsong Zhao, Pengcheng Huang, Xinyu Liu
― 5 min read
Computer Science - Computation and Language
RSSMQM-APE enhances the quality of machine translation evaluations through advanced error analysis.
Qingyu Lu, Liang Ding, Kanjian Zhang
― 7 min read
This study evaluates how well LLMs understand narrative tropes in movie summaries.
Hung-Ting Su, Ya-Ching Hsu, Xudong Lin
― 4 min read
FLEX method offers a new approach for evaluating text-to-SQL systems accurately.
Heegyu Kim, Taeyang Jeon, Seunghwan Choi
― 6 min read
New features enhance user experience in screen understanding and multilingual interactions.
Naman Goyal
― 6 min read
Using technology to gather plant trait information efficiently from the web.
Diego Marcos, Robert van de Vlasakker, Ioannis N. Athanasiadis
― 4 min read
EVA combines audio and visual signals for better speech recognition accuracy.
Yihan Wu, Yifan Peng, Yichen Lu
― 4 min read
A novel model enhances text embeddings through in-context learning strategies.
Chaofan Li, MingHao Qin, Shitao Xiao
― 5 min read
A new method aims to reduce semantic leakage in cross-lingual sentence embeddings.
Dayeon Ki, Cheonbok Park, Hyunjoong Kim
― 5 min read
New models aim to combat harmful language online through advanced detection techniques.
Tonmoy Roy, Md Robiul Islam, Asif Ahammad Miazee
― 6 min read
QualIT enhances text analysis by combining language models and clustering techniques.
Satya Kapoor, Alex Gil, Sreyoshi Bhaduri
― 5 min read
This study investigates AI's role in salary negotiation advice and potential biases.
R. Stuart Geiger, Flynn O'Sullivan, Elsie Wang
― 4 min read
A new framework improves dialogue quality in educational chatbots for effective learning.
Haoyu Huang, Tong Niu, Rui Yang
― 6 min read
This research investigates LLMs' performance in cognitive tasks similar to infant behavior.
Pengrui Han, Peiyang Song, Haofei Yu
― 6 min read
A new tool evaluates large language models' performance across multiple data types.
Yizhi Li, Ge Zhang, Yinghao Ma
― 5 min read
This article presents a new framework to enhance inference-time techniques for language models.
Jon Saad-Falcon, Adrian Gamarra Lafuente, Shlok Natarajan
― 5 min read
A new method enhances aspect-sentiment triplet extraction accuracy.
Iwo Naglik, Mateusz Lango
― 6 min read
A new framework improves prompt creation for large language models.
Mingqi Li, Karan Aggarwal, Yong Xie
― 6 min read
This study assesses various models for retrieving clinical information effectively.
Skatje Myers, Timothy A. Miller, Yanjun Gao
― 7 min read
A new method enhances Flash Attention performance for sparse attention masks.
Agniv Sharma, Jonas Geiping
― 5 min read
A new metric enhancing the assessment of factual consistency in automatic summaries.
Yuxuan Ye, Edwin Simpson, Raul Santos Rodriguez
― 5 min read
Assessing the effectiveness of LLMs for threat analysis.
Sanchana Srikanth, Mohammad Hasanuzzaman, Farah Tasnur Meem
― 10 min read
Examining the advantages of decoder-only models for machine translation tasks.
Gaëtan Caillaut, Raheel Qader, Mariam Nakhlé
― 6 min read
A new AI tool helps assess COVID-19 risk through patient conversations.
Mohammad Amin Roshani, Xiangyu Zhou, Yao Qiang
― 4 min read
This study enhances key information extraction using a new model for unstructured documents.
Furkan Pala, Mehmet Yasin Akpınar, Onur Deniz
― 9 min read
This study highlights methods to enhance large language models in medical settings.
Clément Christophe, Tathagata Raha, Svetlana Maslenkova
― 6 min read
Examining how AI can identify and measure uncertainty in human beliefs.
Anthony Sicilia, Malihe Alikhani
― 7 min read
This approach simplifies choosing effective pretraining datasets for language models.
Tristan Thrush, Christopher Potts, Tatsunori Hashimoto
― 8 min read
A new approach enhances mental health session summaries through a planning engine.
Aseem Srivastava, Smriti Joshi, Tanmoy Chakraborty
― 7 min read
This framework simplifies understanding of privacy policies using AI technology.
Arda Goknil, Femke B. Gelderblom, Simeon Tverdal
― 8 min read
This study examines how AI can help find historical analogies for current events.
Nianqi Li, Siyu Yuan, Jiangjie Chen
― 5 min read
This research highlights key moments in dialogues through a new dataset and analysis framework.
Gia-Bao Dinh Ho, Chang Wei Tan, Zahra Zamanzadeh Darban
― 7 min read
Research improves data generation in machine learning using synthetic methods for clearer explanations.
Patrick Amadeus Irawan, Genta Indra Winata, Samuel Cahyawijaya
― 5 min read
New AI tool simplifies automatic parallelization for C/C++ programming.
Tal Kadosh, Niranjan Hasabnis, Prema Soundararajan
― 7 min read
Research on how speech reveals signs of depression and its implications.
Sona Binu, Jismi Jose, Fathima Shimna K
― 4 min read
BrainKing assesses language models' problem-solving skills under limited information.
Yuyan Chen, Tianhao Yu, Yueze Li
― 6 min read
ToxiCraft improves detection of harmful online content through synthetic data generation.
Zheng Hui, Zhaoxiao Guo, Hang Zhao
― 6 min read
A method for training language models using focused data selection techniques.
Ernie Chang, Pin-Jie Lin, Yang Li
― 6 min read
A new method for assessing T2I model performance across diverse text prompts.
Jingtao Cao, Zheng Zhang, Hongru Wang
― 7 min read
RAGProbe automates the evaluation of RAG systems, improving their performance and reliability.
Shangeetha Sivasothy, Scott Barnett, Stefanus Kurniawan
― 6 min read