5 citations · 6 across the 6 of their papers we have counts for
7 papers
Do VLMs Share Safety Neurons Across Modalities?
Jiaxuan Li, Jiahao Zhang, Duc Minh Vo +3
Vision-language models (VLMs) can comply with harmful requests delivered through images, even when their LLM backbones would refuse the same content in text. While prior work chara…
Predict, Don't React: Value-Based Safety Forecasting for LLM Streaming
Pride Kavumba, Koki Wataoka, Huy H. Nguyen +2
In many practical LLM deployments, a single guardrail is used for both prompt and response moderation. Prompt moderation operates on fully observed text, whereas streaming response…
Rubrik's Cube: Testing a New Rubric for Evaluating Explanations on the CUBE dataset
Diana Galvan-Sosa, Gabrielle Gaudeau, Pride Kavumba +5
The performance and usability of Large-Language Models (LLMs) are driving their use in explanation generation tasks. However, despite their widespread adoption, LLM explanations ha…
Are Prompt-based Models Clueless?
Pride Kavumba, Ryo Takahashi, Yusuke Oda
Finetuning large pre-trained language models with a task-specific head has advanced the state-of-the-art on many natural language understanding benchmarks. However, models with a t…
COPA-SSE: Semi-structured Explanations for Commonsense Reasoning
Ana Brassard, Benjamin Heinzerling, Pride Kavumba +1
We present Semi-Structured Explanations for COPA (COPA-SSE), a new crowdsourced dataset of 9,747 semi-structured, English common sense explanations for Choice of Plausible Alternat…
Learning to Learn to be Right for the Right Reasons
Pride Kavumba, Benjamin Heinzerling, Ana Brassard +1
Improving model generalization on held-out data is one of the core objectives in commonsense reasoning. Recent work has shown that models trained on the dataset with superficial cu…
When Choosing Plausible Alternatives, Clever Hans can be Clever
Pride Kavumba, Naoya Inoue, Benjamin Heinzerling +3
Pretrained language models, such as BERT and RoBERTa, have shown large improvements in the commonsense reasoning benchmark COPA. However, recent work found that many improvements i…