datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
chatbot_arena_conversations
Chatbot Arena Conversations Dataset
This dataset contains 33K cleaned conversations with pairwise human preferences.
It is collected from 13K unique IP addresses on the Chatbot Arena from April to June 2023.
Each sample includes a question ID, two model names, their full conversation text in OpenAI API JSON format, the user vote, the anonymized user ID, the detected language tag, the OpenAI moderation API tag, the additional toxic tag, and the timestamp.
To ensure the safe release… See the full description on the dataset page: https://huggingface.co/datasets/lmsys/chatbot_arena_conversations.chatbot-arena-elo
LMSYS Chatbot Arena ELO Scores
This dataset is a datasets-friendly version of Chatbot Arena ELO scores,
updated daily from the leaderboard API at
https://huggingface.co/spaces/lmarena-ai/chatbot-arena-leaderboard.
Updated: 20250717
Loading Data
from datasets import load_dataset
dataset = load_dataset("mathewhe/chatbot-arena-elo", split="train")
The main branch of this dataset will always be updated to the latest ELO and
leaderboard version. If you need a fixed dataset… See the full description on the dataset page: https://huggingface.co/datasets/mathewhe/chatbot-arena-elo.lmsys-chatbot_arena_conversations
Dataset Card for "lmsys-chatbot_arena_conversations"
More Information needed
chatbot-arena-spoken-voicestw_chatbot_arena
TW Chatbot Arena 資料集說明
概述
TW Chatbot Arena 資料集是一個開源資料集，旨在促進台灣聊天機器人競技場 https://arena.twllm.com/ 的人類回饋強化學習資料（RLHF）。這個資料集包含英文和中文的對話資料，主要聚焦於繁體中文，以支援語言模型的開發和評估。
資料集摘要
授權: Apache-2.0
語言: 主要為繁體中文
規模: 3.6k 筆資料（2024/08/02）
內容: 使用者與聊天機器人的互動，每筆互動都根據回應品質標記為被選擇或被拒絕。
贊助
本計畫由「【g0v 零時小學校】繁體中文AI 開源實踐計畫」（https://sch001.g0v.tw/dash/brd/2024TC-AI-OS-Grant/list）贊助。
資料集結構
資料集包含以下欄位：
question_id: 每次互動的唯一隨機識別碼。
model_a: 左側模型的名稱。
model_b: 右側模型的名稱。
winner:… See the full description on the dataset page: https://huggingface.co/datasets/aigrant/tw_chatbot_arena.chatbot_arena_completionslmsys_chatbot_arena_conversationsdatasource: https://colab.research.google.com/drive/1KdwokPjirkTmpO_P1WByFNFiqxWQquwH
chatbot-arena-llm-judges
Chatbot-Arena
https://www.kaggle.com/competitions/lmsys-chatbot-arena/data
Single-turn data: https://huggingface.co/datasets/potsawee/chatbot-arena-llm-judges
#examples = 49938
split: A_win = 17312 (34.67%), B_win = 16985 (34.01%), tie = 15641 (31.32%)
#2-way only examples = 34297 (68.68%)
This repository
train.single-turn.json: data extracted from the train file from LMSys on Kaggle
each example has attributes - id, model_[a, b], winne_model_[a, b, tie], question… See the full description on the dataset page: https://huggingface.co/datasets/potsawee/chatbot-arena-llm-judges.lmsys_chatbot_arena_conversations
Dataset Card for "lmsys_chatbot_arena_conversations"
More Information needed
cje-chatbot-arena
CJE Chatbot Arena Dataset
Dataset from Causal Judge Evaluation experiments on Chatbot Arena data.
Dataset Structure
cje_dataset.jsonl - Main dataset with judge scores and oracle labels (4,961 prompts)
prompts.jsonl - Original Chatbot Arena prompts
responses/ - Model responses for each policy variant
logprobs/ - Token logprobs for importance sampling estimators
Policies
5 system prompt variants evaluated:
base - No system prompt
clone - "Respond exactly as… See the full description on the dataset page: https://huggingface.co/datasets/elandy/cje-chatbot-arena.WSDM-Cup-Multilingual-Chatbot-Arenachatbot_arena_conversations_hinglishThe dataset is created by translating "lmsys/chatbot_arena_conversations" dataset.
link to original datset - https://huggingface.co/datasets/lmsys/chatbot_arena_conversations
Original dataset contain two conversation from model_a and model_b and also given winner model between these two model conversation.
I have selected winner conversation and converted that user query and assistant answer into hinglish language using Gemini pro
chatbot-arena-conversations-Embeddings
Chatbot Arena Conversations Embeddings
Embeddings of agie-ai/lmsys-chatbot_arena_conversations, produced with amkdg/Qwen3-Embedding-8B-NVFP4 — 4096-d,
L2-normalized float16 (cosine = dot product).
65,960 conversations → 65,960 vectors
emb.npy — float16 [65960, 4096]
meta.parquet — one row per vector, aligned with emb.npy: id, uuid, tag, chunk, n_chunks, count, source_ref
manifest.json — counts and provenance
Usage
import numpy as np, pyarrow.parquet as pq
emb… See the full description on the dataset page: https://huggingface.co/datasets/amkdg/chatbot-arena-conversations-Embeddings.chatbot-arena-llm-refusal
Hand-Labeled Refusal Dataset for Chatbot Arena Responses
Dataset Overview
This dataset extends the Chatbot Arena: Human Preference 55K dataset by providing manual annotations of LLM responses with respect to refusal behaviors. The labels classifies if models refuse to answer a prompt due to ethical concerns or technical/capability limitations.
The Dataset contains labels for 1,750 response pairs, i.e. 3,500 model responses.
Dataset Details
This dataset adds… See the full description on the dataset page: https://huggingface.co/datasets/Human-CentricAI/chatbot-arena-llm-refusal.llm-jp-chatbot-arena-conversations
LLM-jp Chatbot Arena Conversations Dataset
This dataset contains approximately 1,000 conversations with pairwise human preferences, most of which are in Japanese.
The data was collected during the trial phase of the LLM-jp Chatbot Arena (January–February 2025), where users compared responses from two different models in a head-to-head format.
Each sample includes a question ID, the names of the two models, their conversation transcripts, the user's vote, an anonymized user ID, a… See the full description on the dataset page: https://huggingface.co/datasets/llm-jp/llm-jp-chatbot-arena-conversations.chatbot-arena-elo-scores
Dataset Card for "chatbot-arena-elo-scores"
More Information needed
RubricARM-ChatbotArena-ARMJudge-Correctchatbot-arena-ja-karakuri-lm-8x7b-chat-v0.1-awqchatbot-arena-ja-calm2-7b-chatをフィルタリングし、karakuri-lm-8x7b-chat-v0.1-awqでchosenを生成しました
chatbot_arena_personalized_0125chatbot-arena-ja-elo-rating
Leaderboard
ChatBotArena-jaのランキングです｡
2時間に1回程度のペースで更新されます｡
chatbot_arena_completionsRubricARM-ChatbotArena-Llama-Correctchatbot-arena-ja-calm2-7b-chat-experimental_deduped_add_generated_textchatbotarena-spoken-all-7824
ChatbotArena-Spoken Dataset
Based on ChatbotArena, we employ GPT-4o-mini to select dialogue turns that are well-suited to spoken-conversation analysis, yielding 7824 data points. To obtain audio, we synthesize every utterance, user prompt and model responses, using one of 12 voices from KokoroTTS (v0.19), chosen uniformly at random. Because the original human annotations assess only lexical content in text, we keep labels unchanged and treat them as ground truth for the spoken… See the full description on the dataset page: https://huggingface.co/datasets/potsawee/chatbotarena-spoken-all-7824.chatbot_arena_personalized_1023chatbot-arena-dbchatbot_arena_binarizedchatbot_arenachatbot_arena_rm_train_persona_3b_0521chatbot-arena-ja-calm2-7b-chat-experimental_dedupedchatbot-arena-ja-calm2-7b-chatからpromptが一致するデータを削除したデータセットです。