𝗛𝗼𝗻𝗲𝘀𝘁𝗟𝗟𝗠
- Introduces 𝙃𝙊𝙉𝙀𝙎𝙀𝙏, a dataset with 930 queries in six categories to evaluate LLM honesty
- Proposes curiosity-driven prompting and two-stage fine-tuning for improving honesty and helpfulness
- Demonstrates up to 124.7% honesty and helpfulness improvement in models like Mistral-7b
- Introduces 𝙃𝙊𝙉𝙀𝙎𝙀𝙏, a dataset with 930 queries in six categories to evaluate LLM honesty
- Proposes curiosity-driven prompting and two-stage fine-tuning for improving honesty and helpfulness
- Demonstrates up to 124.7% honesty and helpfulness improvement in models like Mistral-7b
December 6, 2024 at 9:06 PM
𝗛𝗼𝗻𝗲𝘀𝘁𝗟𝗟𝗠
- Introduces 𝙃𝙊𝙉𝙀𝙎𝙀𝙏, a dataset with 930 queries in six categories to evaluate LLM honesty
- Proposes curiosity-driven prompting and two-stage fine-tuning for improving honesty and helpfulness
- Demonstrates up to 124.7% honesty and helpfulness improvement in models like Mistral-7b
- Introduces 𝙃𝙊𝙉𝙀𝙎𝙀𝙏, a dataset with 930 queries in six categories to evaluate LLM honesty
- Proposes curiosity-driven prompting and two-stage fine-tuning for improving honesty and helpfulness
- Demonstrates up to 124.7% honesty and helpfulness improvement in models like Mistral-7b
Reposted by Marek Suppa
This is my tiny hill I will die on.
November 29, 2024 at 9:10 PM
This is my tiny hill I will die on.
Multimodal Large Language Models Make Text-to-Image Generative Models Align Better
- VisionPrefer datset captures diverse preferences (prompt-following, aesthetic, fidelity, harmlessness) using multimodal LLMs
- VP-Score model matches human accuracy in preference prediction, guiding model tuning
- VisionPrefer datset captures diverse preferences (prompt-following, aesthetic, fidelity, harmlessness) using multimodal LLMs
- VP-Score model matches human accuracy in preference prediction, guiding model tuning
December 5, 2024 at 10:28 PM
Multimodal Large Language Models Make Text-to-Image Generative Models Align Better
- VisionPrefer datset captures diverse preferences (prompt-following, aesthetic, fidelity, harmlessness) using multimodal LLMs
- VP-Score model matches human accuracy in preference prediction, guiding model tuning
- VisionPrefer datset captures diverse preferences (prompt-following, aesthetic, fidelity, harmlessness) using multimodal LLMs
- VP-Score model matches human accuracy in preference prediction, guiding model tuning
The Super Weight in Large Language Models
Setting as few as a single weight to zero will make various LLMs go from generating coherent text to outputting gibberish.
arxiv.org/abs/2411.07191
Setting as few as a single weight to zero will make various LLMs go from generating coherent text to outputting gibberish.
arxiv.org/abs/2411.07191
The Super Weight in Large Language Models
Recent works have shown a surprising result: a small fraction of Large Language Model (LLM) parameter outliers are disproportionately important to the quality of the model. LLMs contain billions of pa...
arxiv.org
November 28, 2024 at 9:16 AM
The Super Weight in Large Language Models
Setting as few as a single weight to zero will make various LLMs go from generating coherent text to outputting gibberish.
arxiv.org/abs/2411.07191
Setting as few as a single weight to zero will make various LLMs go from generating coherent text to outputting gibberish.
arxiv.org/abs/2411.07191