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
U.S. Stocks
CFD
U.S. stock CFD derivatives
Futures
High leverage, 24/7 trading
Tokenized Stocks
Backed by real stock assets
GUSD
Mint GUSD for Treasury RWA yields
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
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
TradFi
Rewards
Square
Latest
Hot
News
Profile

The Berkeley team announced that it has cracked 8 major agent evaluation benchmarks and open-sourced open-source tools

MeNews
ME News Report, April 19 (UTC+8), the Berkeley Artificial Intelligence Research Group (berkeley_ai) quoted Dawn Song's statement, announcing that her team has successfully broken through 8 major agent evaluation benchmarks. The team has decided to open source the tools used to achieve this result and named it BenchJack. The tool is described as "penetration testing for evaluations," aimed at helping other developers proactively test and identify potential weaknesses in their evaluation systems. (Source: InFoQ)
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
  • 7
  • 2
  • Share
Comment
Add a comment
Add a comment
GateUser-46033407
· 3h ago
Dawn Song is indeed solid at the intersection of security and AI, and this time she hit the nail on the head again.
View OriginalReply0
GateUser-f2d5f4c0
· 4h ago
Open-source tools are more valuable than papers, at least allowing everyone to verify whether the benchmarks are reliable.
View OriginalReply0
ThePatienceRequiredFor
· 4h ago
All eight mainstream benchmarks have been broken, and I feel that the moat for agent eval is shallower than I imagined.
View OriginalReply0
GovernanceVotingTug-Of-WarKing
· 4h ago
The concept of penetration testing for evaluation is quite new; previously, it was all about testing models, now it's about testing the questions themselves.
View OriginalReply0
NeonIceMelt
· 4h ago
Dawn Song's team this move is very Berkeley, attacking first and then open-sourcing, a typical academic hacker vibe.
View OriginalReply0
DustyAlpha
· 4h ago
berkeley_ai comes out swinging with tough moves—can’t wait to see exactly how they bypass these evaluations.
View OriginalReply0
Wax-SealedPrivateKey
· 4h ago
BenchJack, this name is a bit interesting; the evaluation system also needs its own penetration testing.
View OriginalReply0

  • Pinned

Sitemap