Zobrazeno 1 - 10
of 4 960
pro vyhledávání: '"DONG, YUE"'
Autor:
Joy, Sajib Kumar Saha, Mahy, Arman Hassan, Sultana, Meherin, Abha, Azizah Mamun, Ahmmed, MD Piyal, Dong, Yue, Shahariar, G M
In this study, we investigate gender bias in Bangla pretrained language models, a largely under explored area in low-resource languages. To assess this bias, we applied gender-name swapping techniques to existing datasets, creating four manually anno
Externí odkaz:
http://arxiv.org/abs/2411.10636
Autor:
Bachu, Saketh, Shayegani, Erfan, Chakraborty, Trishna, Lal, Rohit, Dutta, Arindam, Song, Chengyu, Dong, Yue, Abu-Ghazaleh, Nael, Roy-Chowdhury, Amit K.
Vision-language models (VLMs) have improved significantly in multi-modal tasks, but their more complex architecture makes their safety alignment more challenging than the alignment of large language models (LLMs). In this paper, we reveal an unfair d
Externí odkaz:
http://arxiv.org/abs/2411.04291
Key-Value (KV) caching is a common technique to enhance the computational efficiency of Large Language Models (LLMs), but its memory overhead grows rapidly with input length. Prior work has shown that not all tokens are equally important for text gen
Externí odkaz:
http://arxiv.org/abs/2410.19258
Language models (LMs) possess a strong capability to comprehend natural language, making them effective in translating human instructions into detailed plans for simple robot tasks. Nevertheless, it remains a significant challenge to handle long-hori
Externí odkaz:
http://arxiv.org/abs/2409.20560
Recent studies show that text-to-image (T2I) models are vulnerable to adversarial attacks, especially with noun perturbations in text prompts. In this study, we investigate the impact of adversarial attacks on different POS tags within text prompts o
Externí odkaz:
http://arxiv.org/abs/2409.15381
Personalized conversational information retrieval (CIR) combines conversational and personalizable elements to satisfy various users' complex information needs through multi-turn interaction based on their backgrounds. The key promise is that the per
Externí odkaz:
http://arxiv.org/abs/2407.16192
Recent research has shown that pruning large-scale language models for inference is an effective approach to improving model efficiency, significantly reducing model weights with minimal impact on performance. Interestingly, pruning can sometimes eve
Externí odkaz:
http://arxiv.org/abs/2406.17261
Query expansion has been employed for a long time to improve the accuracy of query retrievers. Earlier works relied on pseudo-relevance feedback (PRF) techniques, which augment a query with terms extracted from documents retrieved in a first stage. H
Externí odkaz:
http://arxiv.org/abs/2406.07136
Autor:
Cai, Zefan, Zhang, Yichi, Gao, Bofei, Liu, Yuliang, Liu, Tianyu, Lu, Keming, Xiong, Wayne, Dong, Yue, Chang, Baobao, Hu, Junjie, Xiao, Wen
In this study, we investigate whether attention-based information flow inside large language models (LLMs) is aggregated through noticeable patterns for long context processing. Our observations reveal that LLMs aggregate information through Pyramida
Externí odkaz:
http://arxiv.org/abs/2406.02069
Autor:
Chakraborty, Trishna, Shayegani, Erfan, Cai, Zikui, Abu-Ghazaleh, Nael, Asif, M. Salman, Dong, Yue, Roy-Chowdhury, Amit K., Song, Chengyu
Recent studies reveal that integrating new modalities into Large Language Models (LLMs), such as Vision-Language Models (VLMs), creates a new attack surface that bypasses existing safety training techniques like Supervised Fine-tuning (SFT) and Reinf
Externí odkaz:
http://arxiv.org/abs/2406.02575