C1754955896/Lifebenchv2.0

🧠 LifeBench 2.0

A life-logging benchmark dataset for long-term memory evaluation

LifeBench 2.0 collects a full year (2025-01-01 ~ 2025-12-31) of life data for 10 virtual users, covering personas, daily life events, event trees, and 9 types of mobile-phone data (SMS, calls, calendar, notes, photos, push notifications, fitness & health, contacts, and agent chat), together with thousands of question-answer pairs annotated with evidence and score points, for evaluating the long-term memory, temporal reasoning, and multi-hop QA capabilities of large language models. The dataset provides Chinese (data/) and English (data_en/) versions, as well as a standardized LoCoMo conversational format (locomo_format/).

📊 Dataset Statistics

Metric	Value
👥 Users	10
🌐 Languages	Chinese (data/) · English (data_en/)
📅 Time span	2025-01-01 ~ 2025-12-31 (full year)
🗓️ Daily life events	57,486
📱 Phone data records	47,409 (9 data sources)
❓ QA pairs (QA_all / locomo)	3,380
💾 Data size	~423 MB

🏆 Leaderboard

Accuracy (%) of memory systems on LifeBench 2.0 and LoCoMo. LifeBench Micro is the accuracy over all questions, while Macro is the arithmetic mean of the nine category accuracies. Gold Evidence† feeds the annotated supporting evidence directly to the answer model (a reader under perfect retrieval) and is a reference upper bound — it is not a memory system and is excluded from the per-column best. Bold marks the best among memory systems in each column. LoCoMo excludes adversarial questions; — = not reported.

Base LLM	Memory System	SH	MH	TR	ND	KU	CR	CD	HI	UA	Micro	Macro	LoCoMo
DeepSeek-V4-Flash	Mem0	78.09	41.41	32.45	48.48	75.84	52.55	78.07	32.47	55.73	61.54	55.01	85.32
	Cognee	71.69	37.07	27.36	45.44	72.49	43.16	68.63	30.93	92.14	63.17	54.32	81.12
	Hindsight	83.43	53.98	44.34	58.75	85.09	67.83	80.66	49.48	75.56	71.98	66.57	82.83
	MemU	60.34	23.87	17.92	23.19	47.30	30.29	62.26	23.20	90.77	52.75	42.13	80.26
	MemOS	72.94	32.91	26.98	33.65	71.47	40.21	68.63	35.57	88.89	61.51	52.36	79.40
	EverMemOS	71.76	48.55	38.30	54.75	81.23	59.25	67.45	34.02	85.30	66.04	60.07	80.69
	MindMemOS	79.83	42.04	42.45	35.36	69.67	38.07	84.67	26.80	84.10	67.37	55.89	86.70
	Zep	77.78	49.46	46.23	43.92	73.78	53.89	83.25	39.69	54.53	63.85	58.06	88.84
	GraphRAG	22.10	12.93	12.83	19.77	16.97	15.01	11.08	10.31	88.38	30.50	23.26	82.72
	Gold Evidence†	98.14	94.85	94.91	93.73	96.14	94.10	98.11	93.30	100.00	97.28	95.92	—
GLM-5.2	Hindsight	85.60	53.62	45.85	56.65	83.55	66.76	80.90	40.72	84.79	74.41	66.49	—
	EverMemOS	79.64	58.95	59.25	58.75	84.83	57.64	78.54	40.72	71.45	70.95	65.53	—
Qwen-3.8-MAX	Hindsight	83.30	50.72	38.87	54.94	83.29	64.88	82.08	44.85	85.30	72.31	65.36	—
	EverMemOS	73.49	55.06	55.09	54.37	86.38	55.50	72.17	43.81	86.84	69.32	64.75	—

SH Single-hop · MH Multi-hop · TR Temporal · ND Non-declarative · KU Knowledge update · CR Causal · CD Conflict detection · HI Hidden information · UA Unanswerable · Micro micro-average · Macro macro-average (across the 9 categories). † = Gold Evidence reference (perfect retrieval).

📊 Interactive version — sort and explore the full results at C1754955896/Lifebench-Leaderboard.

📁 Directory Structure

version2/
├── README.md                         # This file
├── data/                             # Chinese multi-source data (10 users)
│   └── {user}/                       # One folder per user
│       ├── persona.json              # User profile
│       ├── daily_event.json          # Daily life events
│       ├── event_tree.json           # Event tree structure
│       ├── daily_draft.json          # Daily outline (organized by month)
│       ├── phone_data/               # Mobile-phone data (9 types)
│       │   ├── sms.json              # SMS
│       │   ├── call.json             # Call logs
│       │   ├── calendar.json         # Calendar
│       │   ├── note.json             # Notes
│       │   ├── photo.json            # Photos
│       │   ├── push.json             # Push notifications
│       │   ├── fitness_health.json   # Fitness & health
│       │   ├── contact.json          # Contacts
│       │   └── agent_chat.json       # Agent chat
│       └── QA_all/
│           └── QA.json               # QA pairs (with evidence & score points)
├── data_en/                          # English multi-source data (same structure as data/)
└── locomo_format/                    # LoCoMo conversational format
    ├── lifebench_locomo_conversation_format_v2.0_3380QA.json      # Chinese
    └── lifebench_locomo_conversation_format_v2.0_3380QA_en.json   # English

👥 User List

#	User ID	Chinese Name	English Name
1	fenghaoran	冯浩然	Feng Haoran
2	leimingxuan	雷铭轩	Lei Mingxuan
3	lumingqiang	卢明强	Lu Mingqiang
4	maxiulan	马秀兰	Ma Xiulan
5	songyajing	宋雅静	Song Yajing
6	sunyuwei	孙雨薇	Sun Yuwei
7	yemingxuan	叶铭轩	Ye Mingxuan
8	yinhao	尹浩	Yin Hao
9	yuxiaowei	于晓薇	Yu Xiaowei
10	yuxiaowen	于晓雯	Yu Xiaowen

📄 File Descriptions

Each user folder (data/{user}/ and data_en/{user}/) contains the following files:

File	Description
persona.json	👤 User profile: name, age, occupation, family, personality (MBTI), hobbies, etc.
daily_event.json	📅 Daily life events: a chronological record of the user's life over a year
event_tree.json	🌳 Event tree: hierarchical decomposition of complex events (decompose/subevent)
daily_draft.json	🗂️ Daily outline: organized by month, with date attributes (weather/holiday/weekday) and daily overview
phone_data/	📱 9 types of mobile-phone data (see table below)
QA_all/QA.json	❓ QA pairs: question, answer, evidence chain, score points, question type

📱 Phone Data Sources (phone_data/)

File	Description	Records (10 users total)
agent_chat.json	🤖 Agent chat	8,177
calendar.json	📅 Calendar	6,011
call.json	📞 Call logs	3,947
contact.json	👤 Contacts	244
fitness_health.json	🏃 Fitness & health	3,641
note.json	📝 Notes	7,836
photo.json	📷 Photos	5,604
push.json	🔔 Push notifications	6,026
sms.json	💬 SMS	5,923

❓ QA Data

The dataset provides two representations of the QA pairs:

1. QA_all/QA.json (per user, 3,380 in total): individual QA pairs with the fields question, answer, evidence (evidence chain), score_points, question_type (e.g. Single_hop), required_events_id, and ask_time.

2. locomo_format/ (3,380 QA, 10 samples): converted to the LoCoMo conversational format; each sample contains sample_id, conversation (two-speaker dialogue), and qa.

Each user has 326–354 QA pairs; see the individual user folders for details.

🧾 Data Format Examples

👤 persona.json (user profile)

{
  "name": "Yu Xiaowen",
  "birth": "1997-12-29",
  "age": 24,
  "nationality": "Han",
  "gender": "Female",
  "education": "Undergraduate degree (formal higher education).",
  "job": "Resident physician",
  "occupation": "Zhengzhou Central Hospital",
  "salary": 120000.0,
  "body": { "height": 163, "weight": 50.0, "BMI": 18.8 },
  "personality": { "mbti": "ISFJ", "traits": ["Benevolence", "Social responsibility orientation", "Personal growth orientation"] },
  "hobbies": ["city walk", "Read books/newspapers/magazines", "Listen to European classical music", "Badminton", "Made pottery (hand building)."]
}

📅 daily_event.json (daily event)

{
  "event_id": "1",
  "name": "New Year's Eve dinner and New Year's goal sharing",
  "date": ["2025-01-01 00:00:00 to 2025-01-01 02:30:00"],
  "type": "Relationships",
  "description": "In the early morning, she had a New Year's Eve dinner at home with close friends Zhang Jing and Sun Yue...",
  "participant": [
    { "name": "Yu Xiaowen", "relation": "self" },
    { "name": "Zhang Jing", "relation": "close friend" }
  ],
  "location": "No. 89, Jingsan Road, Jinshui District, Zhengzhou City, Henan Province...",
  "atomic_id": ["1-1"]
}

🌳 event_tree.json (event tree)

{
  "name": "Systematic fund investment and financial planning adjustments.",
  "date": ["2025-06-10"],
  "type": "Finance",
  "event_id": 229,
  "participant": [{ "name": "self", "relation": "self" }],
  "location": "unknown",
  "decompose": 1,
  "subevent": [
    { "event_id": "229-2", "name": "Execute systematic fund investment", "type": "Finance", "decompose": 0 }
  ]
}

❓ QA_all/QA.json (QA pair)

{
  "question": "At noon on the day my dad had a stroke, I remember I had just taken out my lunchbox and had only eaten a few bites when my mom called... How long exactly did that call last?",
  "answer": "2 minutes (12:13-12:15)",
  "required_events_id": ["248", "248"],
  "ask_time": "2025-06-20",
  "question_type": ["Single_hop"],
  "score_points": [
    { "description": "Identified the call record with mother Li Xiuying (12:13-12:15).", "score": 4 },
    { "description": "Correctly calculated the call duration as 2 minutes.", "score": 3 }
  ],
  "evidence": [
    { "type": "call", "phoneNumber": "+8618739081234", "contactName": "Li Xiuying", "datetime": "2025-01-15 12:13:00" }
  ]
}

💬 locomo_format (LoCoMo conversational format)

{
  "sample_id": "Yu Xiaowei",
  "conversation": {
    "speaker_a": "Yu Xiaowei",
    "speaker_b": "Yu Xiaowei's Assistant",
    "session_1": [
      { "speaker": "Yu Xiaowei", "dia_id": "2025-01-01_agent_chat0", "text": "..." }
    ]
  },
  "qa": [ ... ]
}

📥 Loading

from datasets import load_dataset

# Load the entire dataset (including the JSON files under data/ and data_en/)
ds = load_dataset("C1754955896/Lifebenchv2.0")

# Or read a single JSON file directly
import json
with open("data/yuxiaowen/persona.json", encoding="utf-8") as f:
    persona = json.load(f)

📌 Notes

• Two QA representations: QA_all/QA.json (per user) and locomo_format/ (merged conversational format) both contain 3,380 QA pairs — the former is the original annotated format, the latter is the LoCoMo conversational format.
• License: Apache-2.0.