CoolFace
11 shown

datasets

Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.

Clear all
01microsoft /ChatBench Dataset Card for ChatBench This is the dataset from the paper, "ChatBench: From Static Benchmarks to Human-AI Evaluation", by Serina Chang, Ashton Anderson, and Jake Hofman. Data Summary ChatBench contains data from our user study on Prolific and our automated AI-alone experiments, enabling comparison of AI-alone, user-AI, and user-alone answers for the same set of MMLU benchmark questions (Hendrycks et al., 2021). User study. Our user study consists of two phases. In… See the full description on the dataset page: https://huggingface.co/datasets/microsoft/ChatBench.13 likes280 downloads1y agoHugging Face02GabeA /chatbench-specific-detail Specific Detail Part of ChatBench: a benchmark for evaluating embedding models on chat/conversational retrieval tasks. Task Description Given a query about a specific detail, find the conversation containing it. Dataset Statistics Split Queries Corpus Documents test 547 1595 Usage from datasets import load_dataset # Load corpus corpus = load_dataset("GabeA/chatbench-specific-detail", "corpus", split="test") # Load queries queries =… See the full description on the dataset page: https://huggingface.co/datasets/GabeA/chatbench-specific-detail.texttext-retrieval1K<n<10K0 likes58 downloads7mo agoHugging Face03GabeA /chatbench-thread-discrimination Thread Discrimination Part of ChatBench: a benchmark for evaluating embedding models on chat/conversational retrieval tasks. Task Description Distinguish between semantically similar conversations. Dataset Statistics Split Queries Corpus Documents test 505 1595 Usage from datasets import load_dataset # Load corpus corpus = load_dataset("GabeA/chatbench-thread-discrimination", "corpus", split="test") # Load queries queries =… See the full description on the dataset page: https://huggingface.co/datasets/GabeA/chatbench-thread-discrimination.texttext-retrieval1K<n<10K0 likes48 downloads7mo agoHugging Face04GabeA /chatbench-cross-channel Cross Channel Part of ChatBench: a benchmark for evaluating embedding models on chat/conversational retrieval tasks. Task Description Find related conversations across different channels. Dataset Statistics Split Queries Corpus Documents test 554 1595 Usage from datasets import load_dataset # Load corpus corpus = load_dataset("GabeA/chatbench-cross-channel", "corpus", split="test") # Load queries queries =… See the full description on the dataset page: https://huggingface.co/datasets/GabeA/chatbench-cross-channel.texttext-retrieval1K<n<10K0 likes37 downloads7mo agoHugging Face05GabeA /chatbench-conversation-similarity Conversation Similarity Part of ChatBench: a benchmark for evaluating embedding models on chat/conversational retrieval tasks. Task Description Given a conversation, find other conversations about similar topics. Dataset Statistics Split Queries Corpus Documents test 82 1595 Usage from datasets import load_dataset # Load corpus corpus = load_dataset("GabeA/chatbench-conversation-similarity", "corpus", split="test") # Load queries… See the full description on the dataset page: https://huggingface.co/datasets/GabeA/chatbench-conversation-similarity.texttext-retrieval1K<n<10K0 likes33 downloads7mo agoHugging Face06GabeA /chatbench-topic-retrieval Topic Retrieval Part of ChatBench: a benchmark for evaluating embedding models on chat/conversational retrieval tasks. Task Description Given a topic description, find relevant conversations. Dataset Statistics Split Queries Corpus Documents test 513 1595 Usage from datasets import load_dataset # Load corpus corpus = load_dataset("GabeA/chatbench-topic-retrieval", "corpus", split="test") # Load queries queries =… See the full description on the dataset page: https://huggingface.co/datasets/GabeA/chatbench-topic-retrieval.texttext-retrieval1K<n<10K0 likes33 downloads7mo agoHugging Face07GabeA /chatbench-cross-platform-transfer Cross-Platform Transfer (slack) Part of ChatBench: a benchmark for evaluating embedding models on chat/conversational retrieval tasks. Task Description Thread retrieval evaluated on a held-out chat platform. Dataset Statistics Split Queries Corpus Documents test 113 1232 Usage from datasets import load_dataset # Load corpus corpus = load_dataset("GabeA/chatbench-cross-platform-transfer", "corpus", split="test") # Load queries… See the full description on the dataset page: https://huggingface.co/datasets/GabeA/chatbench-cross-platform-transfer.texttext-retrieval1K<n<10K0 likes23 downloads7mo agoHugging Face08GabeA /chatbench-thread-retrieval Thread Retrieval Part of ChatBench: a benchmark for evaluating embedding models on chat/conversational retrieval tasks. Task Description Given a message from a chat conversation, retrieve the thread it belongs to. Dataset Statistics Split Queries Corpus Documents test 500 1232 Usage from datasets import load_dataset # Load corpus corpus = load_dataset("GabeA/chatbench-thread-retrieval", "corpus", split="test") # Load queries queries… See the full description on the dataset page: https://huggingface.co/datasets/GabeA/chatbench-thread-retrieval.texttext-retrieval1K<n<10K0 likes20 downloads7mo agoHugging Face09GabeA /chatbench-response-retrieval Response Retrieval Part of ChatBench: a benchmark for evaluating embedding models on chat/conversational retrieval tasks. Task Description Given the beginning of a conversation, retrieve the continuation. Dataset Statistics Split Queries Corpus Documents test 500 1173 Usage from datasets import load_dataset # Load corpus corpus = load_dataset("GabeA/chatbench-response-retrieval", "corpus", split="test") # Load queries queries =… See the full description on the dataset page: https://huggingface.co/datasets/GabeA/chatbench-response-retrieval.texttext-retrieval1K<n<10K0 likes17 downloads7mo agoHugging Face10GabeA /chatbench-resultstabularn<1K0 likes11 downloads7mo agoHugging Face11GabeA /chat-benchtextn<1K0 likes7 downloads7mo agoHugging Face

Listings come live from the Hugging Face Hub API. CoolFace does not host these files.