6 papers
Diverging Preferences: When do Annotators Disagree and do Models Know?
Michael JQ Zhang, Zhilin Wang, Jena D. Hwang +6
We examine diverging preferences in human-labeled preference datasets. We develop a taxonomy of disagreement sources spanning ten categories across four high-level classes and find…
Let Them Down Easy! Contextual Effects of LLM Guardrails on User Perceptions and Preferences
Mingqian Zheng, Wenjia Hu, Patrick Zhao +5
Current LLMs are trained to refuse potentially harmful input queries regardless of whether users actually had harmful intents, causing a tradeoff between safety and user experience…
DataDecide: How to Predict Best Pretraining Data with Small Experiments
Ian Magnusson, Nguyen Tai, Ben Bogin +10
Because large language models are expensive to pretrain on different datasets, using smaller-scale experiments to decide on data is crucial for reducing costs. Which benchmarks and…
Intentionally Unintentional: GenAI Exceptionalism and the First Amendment
David Atkinson, Jena D. Hwang, Jacob Morrison
This paper challenges the assumption that courts should grant First Amendment protections to outputs from large generative AI models, such as GPT-4 and Gemini. We argue that becaus…
Semantic and Expressive Variation in Image Captions Across Languages
Andre Ye, Sebastin Santy, Jena D. Hwang +2
Computer vision often treats human perception as homogeneous: an implicit assumption that visual stimuli are perceived similarly by everyone. This assumption is reflected in the wa…
Tulu 3: Pushing Frontiers in Open Language Model Post-Training
Nathan Lambert, Jacob Morrison, Valentina Pyatkin +20
Language model post-training is applied to refine behaviors and unlock new skills across a wide range of recent language models, but open recipes for applying these techniques lag…