Zobrazeno 1 - 10
of 246
pro vyhledávání: '"Bach, Stephen"'
Many recent works have explored using language models for planning problems. One line of research focuses on translating natural language descriptions of planning tasks into structured planning languages, such as the planning domain definition langua
Externí odkaz:
http://arxiv.org/abs/2407.03321
Detoxifying multilingual Large Language Models (LLMs) has become crucial due to their increasing global use. In this work, we explore zero-shot cross-lingual generalization of preference tuning in detoxifying LLMs. Unlike previous studies that show l
Externí odkaz:
http://arxiv.org/abs/2406.16235
Recent works often assume that Vision-Language Model (VLM) representations are based on visual attributes like shape. However, it is unclear to what extent VLMs prioritize this information to represent concepts. We propose Extract and Explore (EX2),
Externí odkaz:
http://arxiv.org/abs/2403.16442
We introduce Bonito, an open-source model for conditional task generation that converts unannotated text into task-specific training datasets for instruction tuning. We aim to enable zero-shot task adaptation of large language models on users' specia
Externí odkaz:
http://arxiv.org/abs/2402.18334
Data scarcity in low-resource languages can be addressed with word-to-word translations from labeled task data in high-resource languages using bilingual lexicons. However, bilingual lexicons often have limited lexical overlap with task data, which r
Externí odkaz:
http://arxiv.org/abs/2402.14086
Prompted weak supervision (PromptedWS) applies pre-trained large language models (LLMs) as the basis for labeling functions (LFs) in a weak supervision framework to obtain large labeled datasets. We further extend the use of LLMs in the loop to addre
Externí odkaz:
http://arxiv.org/abs/2402.01867
Autor:
Esfandiarpoor, Reza, Bach, Stephen H.
A promising approach for improving the performance of vision-language models like CLIP for image classification is to extend the class descriptions (i.e., prompts) with related attributes, e.g., using brown sparrow instead of sparrow. However, curren
Externí odkaz:
http://arxiv.org/abs/2311.07593
AI safety training and red-teaming of large language models (LLMs) are measures to mitigate the generation of unsafe content. Our work exposes the inherent cross-lingual vulnerability of these safety mechanisms, resulting from the linguistic inequali
Externí odkaz:
http://arxiv.org/abs/2310.02446
Fine-tuning vision-language models (VLMs) like CLIP to downstream tasks is often necessary to optimize their performance. However, a major obstacle is the limited availability of labeled data. We study the use of pseudolabels, i.e., heuristic labels
Externí odkaz:
http://arxiv.org/abs/2306.01669
We introduce an adaptive method with formal quality guarantees for weak supervision in a non-stationary setting. Our goal is to infer the unknown labels of a sequence of data by using weak supervision sources that provide independent noisy signals of
Externí odkaz:
http://arxiv.org/abs/2306.01658