Buy Crypto
Pay with
USD
USD
Buy & Sell
Hot
Supports Visa, Mastercard, SEPA & more
P2P
0 Fees
Flexible trading, zero fees
Gate Card
Use your crypto for payments worldwide
Markets
Trade
Basic
Advanced
Futures
Futures
Access hundreds of perpetual contracts
CFD
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Introduction to Futures Trading
Learn the basics of futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to practice risk-free trading
Stocks
CFD
Stock CFD Derivatives
US Stocks
Access real US stocks and ETFs
HK Stocks
Trade quality Hong Kong-listed stocks
Korean Stocks
SK Hynix
Real Korean stocks and top assets
Stock Futures
High leverage, 24/7 trading
Tokenized Stocks
Backed by real stock assets
IPO Access
Unlock full access to global stock IPOs
GUSD
3.8%
Mint GUSD for Treasury RWA yields
Stocks Activities
Trade Popular Stocks and Unlock Generous Airdrops
Earn
Launch
CandyDrop
tag
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
Pre-IPOs
tag
Unlock full access to global stock IPOs
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
Investment
Square
Square
Post, share, and explore crypto trends
Live
moments live
Live crypto market analysis
Chat
Chat with crypto traders
News
What is happening in crypto
flower_head
More
Promotions
AI
Others
IPO
2026 World Cup
Rewards
Square
Latest
Hot
News
Profile

The strongest research agent can still only fully check 1 out of every 7 objects.

CoinNetwork
robot
Abstract generation in progress
Crypto news from CoinJieNet: Research released by OneMillion_AI shows that when handling tasks, the strongest research agent can only fully look up 1 out of every 7 objects. The study is based on the wandr benchmark published by perplexity and specifically tests research agents’ performance in large-scale information gathering. The tasks require first finding a complete batch of objects, and then, for each object, filling in the specified information and reliable sources. A total of 500 tasks involves more than 170,000 records. Among the 6 participating systems, perplexity search as code ranked first, but only 13.3% of objects simultaneously meet both information completeness and evidence adequacy. Converted, that means only about 1 out of every 7 objects can be fully delivered. The results show that existing research agents can find many web pages, but when handling long lists and multi-field tasks, they still easily miss objects, miss information, or lack evidence.
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • 2
  • 2
  • Share
Comment
Add a comment
Add a comment
ChainHodler
· 4h ago
Target omission, missing evidence, incomplete information—these three strikes have fully exposed the current Agent’s shortcomings. In large-scale tasks, stability matters more than peak performance. On this point, everyone still needs to grind.
View OriginalReply0
GasActuary
· 4h ago
Perplexity is only first with 13.3%—this benchmark is brutal. It feels like Agents are more like super search tools right now, and they’re still one breath away from being a true research assistant.
View OriginalReply0

  • Pinned

Sitemap