Fixing reference to scoring icons: humanoid robot vs. computer joystick.
Loading data from Talking to Chatbots Dataset reddgr/talking-to-chatbots-chats …
😎
[image: file-zcycJpZnzzoK345HJDLSceQM]
🤖 Here’s an ALT text for the image:😎 the icons used for scoring are these 🤖🕹️. They are not exactly positive and neutral faces. 🤖 🤖 (a humanoid robot) is meant to symbolized a higher level of “intelligence” than 🕹️ (a computer joystick). Don’t extend on this explanations (humans understand it perfectly), just fix your reference to face icons
A screenshot of a comparison table titled “Chatbot Battle: Web Browsing Chatbots”. The table lists five chatbots: ChatGPT, Copilot, HuggingChat, Leo, and Bard. ChatGPT and Copilot are ranked as “Winner” with a trophy icon, HuggingChat and Leo are “Runner-up” with a medal icon, and Bard is a “Contender”. Each chatbot is evaluated on four criteria: Specificity, Coherency, Brevity, and Novelty, represented by icons of faces with varying expressions. All chatbots have a mix of positive and neutral face icons for each criterion. There is a “Link” column on the far right with the text “View Chat” under each chatbot name.
🤖 Here’s an ALT text for the image:
A screenshot of a comparison table titled “Chatbot Battle: Web Browsing Chatbots”. The table lists five chatbots: ChatGPT, Copilot, HuggingChat, Leo, and Bard. ChatGPT and Copilot are ranked as “Winner” with a trophy icon, HuggingChat and Leo are “Runner-up” with a medal icon, and Bard is a “Contender”. Each chatbot is evaluated on four criteria: Specificity, Coherency, Brevity, and Novelty, represented by icons of humanoid robots 🤖 for higher levels and computer joysticks 🕹️ for lower levels. Each criterion for all chatbots has a combination of these icons. There is a “Link” column on the far right with the text “View Chat” under each chatbot name.
Hugging Face Dataset Metrics
All the conversation prompts, responses, and metrics are available to download and explore on Hugging Face dataset reddgr/talking-to-chatbots-chats: