Phillip Wallis

Reinforcement Learning with Backtracking Feedback

Bilgehan Sel, Vaishakh Keshava, Phillip Wallis, Lukas Rutishauser et autres

Addressing the critical need for robust safety in Large Language Models (LLMs), particularly against adversarial attacks and in-distribution errors, we introduce Reinforcement Learning with Backtracking Feedback (RLBF). This framework advances upon prior methods, such as BSAFE, by primarily leveraging a Reinforcement Learning …

Reinforcement Learning with Backtracking Feedback

Bilgehan Sel, Vaishakh Keshava, Phillip Wallis, Lukas Rutishauser et autres

Addressing the critical need for robust safety in Large Language Models (LLMs), particularly against adversarial attacks and in-distribution errors, we introduce Reinforcement Learning with Backtracking Feedback (RLBF). This framework advances upon prior methods, such as BSAFE, by primarily leveraging a Reinforcement Learning …

us (code pays fourni par la source)

CausalArmor: Efficient Indirect Prompt Injection Guardrails via Causal Attribution

Minbeom Kim, Mihir Parmar, Phillip Wallis, Lesly Miculicich et autres

AI agents equipped with tool-calling capabilities are susceptible to Indirect Prompt Injection (IPI) attacks. In this attack scenario, malicious commands hidden within untrusted content trick the agent into performing unauthorized actions. Existing defenses can reduce attack success but often suffer from the …

CausalArmor: Efficient Indirect Prompt Injection Guardrails via Causal Attribution

Minbeom Kim, Mihir Parmar, Phillip Wallis, Lesly Miculicich et autres

AI agents equipped with tool-calling capabilities are susceptible to Indirect Prompt Injection (IPI) attacks. In this attack scenario, malicious commands hidden within untrusted content trick the agent into performing unauthorized actions. Existing defenses can reduce attack success but often suffer from the …

Adversarial Reinforcement Learning for Large Language Model Agent Safety

Zizhao Wang, Dingcheng Li, Vaishakh Keshava, Phillip Wallis et autres

Large Language Model (LLM) agents can leverage tools such as Google Search to complete complex tasks. However, this tool usage introduces the risk of indirect prompt injections, where malicious instructions hidden in tool outputs can manipulate the agent, posing security risks like …

Backtracking for Safety

Bilgehan Sel, Dingcheng Li, Phillip Wallis, Vaishakh Keshava et autres

Large language models (LLMs) have demonstrated remarkable capabilities across various tasks, but ensuring their safety and alignment with human values remains crucial. Current safety alignment methods, such as supervised fine-tuning and reinforcement learning-based approaches, can exhibit vulnerabilities to adversarial attacks and often …

Efficient Fine-Tuning of Deep Neural Networks with Effective Parameter Allocation

Phillip Wallis, Xubo B. Song

It’s commonplace in modern deep learning to achieve SOTA performance by fine-tuning a large, pretrained base model. Recent successes in natural language processing, attributed in part to knowledge transfer from large, pretrained, transformer-based language models, have sparked a similar revolution in computer …

us (code pays fourni par la source)

LoRA Fine-Tuning of a 3B Code LLM for Algorithmic Efficiency

J. Edward Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu et autres

An important paradigm of natural language processing consists of large-scale pre-training on general domain data and adaptation to particular tasks or domains. As we pre-train larger models, full fine-tuning, which retrains all model parameters, becomes less feasible. Using GPT-3 175B as an …

Automatic Event Detection of REM Sleep Without Atonia From Polysomnography Signals Using Deep Neural Networks

Phillip Wallis, Daniel B. Yaeger, Alexander B. Kain, Xubo B. Song et autres

Rapid eye movement (REM) sleep behavior disorder (RBD) is a sleep disorder that features loss of atonia, or REM sleep without atonia (RSWA). RBD and RSWA are early manifestations of degenerative neurological diseases such as Parkinson’s and Lewy Body Dementia. Accurate diagnosis …

us (code pays fourni par la source)

Differential Equation Units: Learning Functional Forms of Activation Functions from Data

MohamadAli Torkamani, Shiv Shankar Shankar, Amirmohammad Rooshenas, Phillip Wallis

Most deep neural networks use simple, fixed activation functions, such as sigmoids or rectified linear units, regardless of domain or network structure. We introduce differential equation units (DEUs), an improvement to modern neural networks, which enables each neuron to learn a particular …

de, us, gb (code pays fourni par la source)

Differential Equation Units: Learning Functional Forms of Activation Functions from Data

MohamadAli Torkamani, Amirmohammad Rooshenas, Phillip Wallis

Most deep neural networks use simple, fixed activation functions, such as sigmoids or rectified linear units, regardless of domain or network structure. We introduce differential equation units (DEUs), an improvement to modern neural networks, which enables each neuron to learn a particular …

de, us (code pays fourni par la source)

Opportunities for improving NRM governance in Victoria

Moragh Mackay, Ross Colliver, Phillip Wallis, Ray L. Ison et autres