Bo Li

Probing Human Visual Robustness with Neurally-Guided Deep Neural Networks

Any-Depth Alignment: Unlocking Innate Safety Alignment of LLMs to Any-Depth

ARMs: Adaptive Red-Teaming Agent against Multimodal Models with Plug-and-Play Attacks

Reasoning over Boundaries: Enhancing Specification Alignment via Test-time Delibration

Guarding a Needle in the Haystack: A Real-Time Policy-Following Streaming Video Guardrail

BlueCodeAgent: A Blue Teaming Agent Enabled by Automated Red Teaming for CodeGen AI

Scaling Agent Learning via Experience Synthesis

Cer-Eval: Certifiable and Cost-Efficient Evaluation Framework for LLMs

TrustGen: A Platform of Dynamic Benchmarking on the Trustworthiness of Generative Foundation Models

GraphQ-LM: Scalable Graph Representation for Large Language Models via Residual Vector Quantization

RedCodeAgent: Automatic Red-teaming Agent against Diverse Code Agents

Extrapolating Large Models from the Small: Optimal Learning of Scaling Laws

See, Think, Hallucinate: Interpreting Reasoning and Hallucinations Beyond the First Hop in Vision-Language Models

Mitigating Indirect Prompt Injection via Instruction-Following Intent Analysis

SoSBench: Benchmarking Safety Alignment on Six Scientific Domains

SAFEVISION: Efficient Image Guardrail with Robust Policy Adherence and Explainability

ArchPilot: A Proxy-Guided Multi-Agent Approach for Machine Learning Engineering

AIR-BENCH 2024: A Safety Benchmark based on Regulation and Policies Specified Risk Categories

$R^2$-Guard: Robust Reasoning Enabled LLM Guardrail via Knowledge-Enhanced Logical Reasoning

Reliable and Efficient Amortized Model-based Evaluation

AutoDAN-Turbo: A Lifelong Agent for Strategy Self-Exploration to Jailbreak LLMs

SafeWatch: An Efficient Safety-Policy Following Video Guardrail Model with Transparent Explanations

MMDT: Decoding the Trustworthiness and Safety of Multimodal Foundation Models

Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment

AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integration

C-SafeGen: Certified Safe LLM Generation with Claim-Based Streaming Guardrails

SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal

EIA: ENVIRONMENTAL INJECTION ATTACK ON GENERALIST WEB AGENTS FOR PRIVACY LEAKAGE

Reliable and Efficient Amortized Model-based Evaluation

Underestimated Privacy Risks for Minority Populations in Large Language Model Unlearning

AdvAgent: Controllable Blackbox Red-teaming on Web Agents

ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning

SafeAuto: Knowledge-Enhanced Safe Autonomous Driving with Multimodal Foundation Models

GuardAgent: Safeguard LLM Agents via Knowledge-Enabled Reasoning

GuardAgent: Safeguard LLM Agent by a Guard Agent via Knowledge-Enabled Reasoning

Tamper-Resistant Safeguards for Open-Weight LLMs

Can Watermarks be Used to Detect LLM IP Infringement For Free?

Underestimated Privacy Risks for Minority Populations in Large Language Model Unlearning

KnowHalu: Multi-Form Knowledge Enhanced Hallucination Detection

AdvWave: Stealthy Adversarial Jailbreak Attack against Large Audio-Language Models

KnowData: Knowledge-Enabled Data Generation for Improving Multimodal Models

SafeVision: Efficient Image Guardrail with Robust Policy Adherence and Explainability

Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Dataset

Assessing the Knowledge-intensive Reasoning Capability of Large Language Models with Realistic Benchmarks Generated Programmatically at Scale

On Memorization of Large Language Models in Logical Reasoning

SCOPE: Scalable and Adaptive Evaluation of Misguided Safety Refusal in LLMs

MultiTrust: Enhancing Safety and Trustworthiness of Large Language Models from Multiple Perspectives

SecCodePLT: A Unified Platform for Evaluating the Security of Code GenAI

UDora: A Unified Red Teaming Framework against LLM Agents by Dynamically Hijacking Their Own Reasoning

RedCodeAgent: Automatic Red-teaming Agent against Code Agents

AdvWeb: Controllable Black-box Attacks on VLM-powered Web Agents

AutoHijacker: Automatic Indirect Prompt Injection Against Black-box LLM Agents

SafeAuto: Knowledge-Enhanced Safe Autonomous Driving with Multimodal Foundation Models

AutoRedTeamer: An Autonomous Red Teaming Agent Against Language Models

Which Network is Trojaned? Increasing Trojan Evasiveness for Model-Level Detectors

IDS-Agent: An LLM Agent for Explainable Intrusion Detection in IoT Networks