Zobrazeno 1 - 10
of 780
pro vyhledávání: '"Singh, Sameer"'
Uncertainty expressions such as ``probably'' or ``highly unlikely'' are pervasive in human language. While prior work has established that there is population-level agreement in terms of how humans interpret these expressions, there has been little i
Externí odkaz:
http://arxiv.org/abs/2407.15814
Gender bias research has been pivotal in revealing undesirable behaviors in large language models, exposing serious gender stereotypes associated with occupations, and emotions. A key observation in prior work is that models reinforce stereotypes as
Externí odkaz:
http://arxiv.org/abs/2405.00588
Content Warning: This paper contains examples of misgendering and erasure that could be offensive and potentially triggering. Misgendering, the act of incorrectly addressing someone's gender, inflicts serious harm and is pervasive in everyday technol
Externí odkaz:
http://arxiv.org/abs/2404.14695
Autor:
Nottingham, Kolby, Majumder, Bodhisattwa Prasad, Mishra, Bhavana Dalvi, Singh, Sameer, Clark, Peter, Fox, Roy
Large language models (LLMs) have recently been used for sequential decision making in interactive environments. However, leveraging environment reward signals for continual LLM actor improvement is not straightforward. We propose Skill Set Optimizat
Externí odkaz:
http://arxiv.org/abs/2402.03244
Amidst growing concerns of large language models (LLMs) being misused for generating misinformation or completing homework assignments, watermarking has emerged as an effective solution for distinguishing human-written and LLM-generated text. A promi
Externí odkaz:
http://arxiv.org/abs/2311.09816
Autor:
Elazar, Yanai, Paranjape, Bhargavi, Peng, Hao, Wiegreffe, Sarah, Raghavi, Khyathi, Srikumar, Vivek, Singh, Sameer, Smith, Noah A.
The inevitable appearance of spurious correlations in training datasets hurts the generalization of NLP models on unseen data. Previous work has found that datasets with paired inputs are prone to correlations between a specific part of the input (e.
Externí odkaz:
http://arxiv.org/abs/2311.09605
Autor:
Elazar, Yanai, Bhagia, Akshita, Magnusson, Ian, Ravichander, Abhilasha, Schwenk, Dustin, Suhr, Alane, Walsh, Pete, Groeneveld, Dirk, Soldaini, Luca, Singh, Sameer, Hajishirzi, Hanna, Smith, Noah A., Dodge, Jesse
Large text corpora are the backbone of language models. However, we have a limited understanding of the content of these corpora, including general statistics, quality, social factors, and inclusion of evaluation data (contamination). In this work, w
Externí odkaz:
http://arxiv.org/abs/2310.20707
Language models are achieving impressive performance on various tasks by aggressively adopting inference-time prompting techniques, such as zero-shot and few-shot prompting. In this work, we introduce EchoPrompt, a simple yet effective approach that
Externí odkaz:
http://arxiv.org/abs/2309.10687
Bias amplification is a phenomenon in which models exacerbate biases or stereotypes present in the training data. In this paper, we study bias amplification in the text-to-image domain using Stable Diffusion by comparing gender ratios in training vs.
Externí odkaz:
http://arxiv.org/abs/2308.00755
Autor:
Nottingham, Kolby, Razeghi, Yasaman, Kim, Kyungmin, Lanier, JB, Baldi, Pierre, Fox, Roy, Singh, Sameer
Large language models (LLMs) are being applied as actors for sequential decision making tasks in domains such as robotics and games, utilizing their general world knowledge and planning abilities. However, previous work does little to explore what en
Externí odkaz:
http://arxiv.org/abs/2307.11922