My current research centers around:
- LLM agents and agentic RLs, including environment and task constructions, specialized model training, harness and evaluation designs
- Agent Safety and Security, including lost-of-control alignment, monitoring, red-teaming and guardrails
At Meta, I contribute to post-training and evaluation of the Muse models on cybersecurity and coding, as well as system safety for the Muse-related products.
My previous works studied the trustworthiness of DRL and DNN via statistical modeling (e.g., Deep GP, Gaussian process
models) and learning (e.g., variational inference, MCMC, empirical bayes),
and designing DNNs for security applications with a focus on noisy learning and OOD.
- Academic Senate Faculty Research Award, UCSB, 2025
- Google
ML and Systems Junior Faculty Award, Google, 2025
- Berkeley RDI AI & Decentralization Innovation Award, UC Berkeley, 2025
- Amazon Research Award (with Christopher Kruegel), Amazon, 2024
- FAR AI Research Award, FAR AI, 2024
- IBM
Fellowship Award, IBM, 2020
- Facebook
Fellowship Finalist, Facebook, 2020
- Baidu AI Fellowship Finalist, Baidu, 2020
-
CCS Outstanding Paper Award, ACM, 2018
- Black Hat Student Scholarship, Black Hat, 2018-2020
- Multiple Conference Travel Grants (e.g., USENIX Security, CCS, NeurIPS, ICML), 2018-2021
-
Terminal-Bench: OpenSage ranked as the top one agent (Feb 2026)
-
DARPA AIxCC
Top 5, August, 2025 (
Press release: Washington post
)
-
SWE-bench-Verfied: PatchPilot ranked in the top five open source
tools; PatchPilot+co-PatcheR ranked second in the open weighted model (May 2025)
-
Google SBFT Fuzzing Tool Competition
Top 1, April, 2024
-
Geekpwn Competition on Data Tracing CTF
Top 10, August, 2018
-
Geekpwn Competition on Advesarial Attacks and Defenses CTF
Finalist (Top 6 worldwide), August, 2018
-
Kanxue AI CTF
Competition Question Maker, June, 2018
Copyright © Wenbo Guo
