BerkeleyRDI Profile Banner
UC Berkeley RDI Profile
UC Berkeley RDI

@BerkeleyRDI

Followers
2K
Following
212
Media
117
Statuses
333

UC Berkeley's campus-wide, cross-disciplinary Center for Responsible, Decentralized Intelligence - RDI

Berkeley, CA
Joined December 2021
Don't wanna be here? Send us removal request.
@BerkeleyRDI
UC Berkeley RDI
28 days
🚨 Early Bird Sale Ends June 30! 🚨. Don't miss your chance to join the premier gathering of 1500+ in-person attendees for the #AgenticAI community at the Agentic AI Summit 2025 – hosted on August 2 at @UCBerkeley!.Early bird sale through June 30:. 🎓 Student Pass: $25. 🚀
Tweet media one
1
1
15
@BerkeleyRDI
UC Berkeley RDI
19 days
RT @dawnsongtweets: 1/ 🔥 AI agents are reaching a breakthrough moment in cybersecurity. In our latest work:. 🔓 CyberGym: AI agents discov….
0
138
0
@BerkeleyRDI
UC Berkeley RDI
1 month
🚨 FINAL CALL: #AgentX submissions close in 3 days! 🚨.⏰ Deadline: May 31st, PM PT.💰 Over $150,000 in prizes across both tracks from our amazing sponsors including @amazon, @auth0, @okta, @GroqInc, @huggingface, @Google, @LambdaAPI, @mlfoundry, @MistralAI, NobelEra Group,
Tweet media one
0
1
2
@BerkeleyRDI
UC Berkeley RDI
1 month
🚀Missed our NobelEra Group Frontier Innovators Club (fka Nobel FutureTech Genius Club) info session on 5/15? We've got you covered!.📹Watch Dr. Jiadi Zhu share his inspiring journey from MIT PhD to founding CDimension, plus how the Frontier Innovators network amplified his
Tweet media one
0
0
1
@BerkeleyRDI
UC Berkeley RDI
2 months
RT @dawnsongtweets: 🔐 Frontier AI is reshaping cybersecurity, raising critical new questions:.🔍 What is its current impact?.⚖️ Who stands t….
0
25
0
@BerkeleyRDI
UC Berkeley RDI
2 months
RT @dawnsongtweets: 🌟 Excited to announce our esteemed panel of judges for the #AgentX competition by @BerkeleyRDI @UCBerkeley. Huge thanks….
0
29
0
@BerkeleyRDI
UC Berkeley RDI
2 months
🚀 Exciting opportunity! Join Nobel FutureTech Group & Berkeley RDI for an exclusive virtual fireside chat on Thur 5/15 at 12PM PT. Meet Jiadi Zhu, MIT PhD & CEO of CDimension, who's developing cutting-edge materials and semiconductor infrastructure for AI and high-performance
Tweet media one
0
0
1
@BerkeleyRDI
UC Berkeley RDI
2 months
🔐Secure Your AI Agents with @Auth0 — Special AgentX Workshop! 📷.Join our exclusive workshop (Wed 5/7 10am PT) in collaboration with @Auth0 @okta to master robust authentication for your AI-powered agents!.📚Learn:. ✅Best practices for implementing authentication in LLM-powered
Tweet media one
0
0
2
@BerkeleyRDI
UC Berkeley RDI
2 months
RT @dawnsongtweets: Excited to speak at @RSAConference 2025! Join my session "Safety and Security of LLM Agents: Challenges and Future Dire….
0
14
0
@BerkeleyRDI
UC Berkeley RDI
2 months
10/10 Next Steps: Tackling proof‐data scarcity, strengthening verifiability, enabling open-ended conjecturing, and scaling to vast search spaces and experimental workflows. 🚀🔧. ▶️ Dive into Lecture 11 with Swarat Chaudhuri and witness the future of AI‐driven discovery!.
0
0
0
@BerkeleyRDI
UC Berkeley RDI
2 months
9/10 Beyond Math: Agents extend to experiment design, scaling‐law identification, and self‐evolving visual concept libraries for empirical science. 🔬🔭
Tweet media one
1
1
1
@BerkeleyRDI
UC Berkeley RDI
2 months
8/10 Concept‐Guided Evolution (LaSR): LLM‐guided concept abstraction accelerates discovery—finding Coulomb’s Law in fewer steps and simpler forms. ⚡🔬
Tweet media one
1
0
0
@BerkeleyRDI
UC Berkeley RDI
2 months
7/10 Symbolic Regression (PySR): Evolutionary search over expression trees uncovers interpretable laws (e.g., Kepler’s Third Law) directly from data. 🌌📈.
1
0
0
@BerkeleyRDI
UC Berkeley RDI
2 months
6/10 Formal Verification Demo: See how an LLM agent invents lemmas and constructs a complete compiler‐correctness proof in Lean, showcasing real‐world impact. 💻🔒.
1
0
0
@BerkeleyRDI
UC Berkeley RDI
2 months
5/10 In-Context Agents (Copra): A hierarchical pipeline—prompt synthesis, tactic parsing, lemma retrieval, backtracking—solves formal goals without massive curated corpora. 📝🔨
Tweet media one
1
0
0
@BerkeleyRDI
UC Berkeley RDI
2 months
4/10 RL‐Powered Proofs (AlphaProof): Inspired by AlphaZero, agents use perfect formal feedback to learn proof strategies via self-play, approaching IMO‐level performance. ♟️🔄.
1
0
0