Abstract: While AI has achieved human-level performance in specific tasks, it often fails when deployed in real-world applications — like failing to detect a large truck in front of a Tesla. Why do such errors occur, and how long will it be before AI is widely and reliably deployed in everyday life? This talk will
Abstract: Abstract: In a world increasingly shaped by AI technologies, addressing their technical, ethical, and fairness challenges becomes imperative. This talk will address the criticality of identifying ethical, responsible, and safe requirements for large language models, featuring published and ongoing research at Precog Lab https://precog.iiit.ac.in/. We’ll discuss representation steering to make interpretable changes to language
Abstract: This talk will discuss the challenges associated with tracing data and attributing it to the input or output of a model. Tracing of data to the input, referred to as membership inference, involves determining whether a specific data point was part of the model’s training set. Successfully attributing data to the training set of
Abstract: With blockchain technology at its core, cryptocurrencies have captured widespread attention both in academic research and beyond in recent years. As these systems oversee increasing stakes, reaching into the billions of US dollars, they have become attractive attack targets. Notably, several attacks have focused on their network layer, which crucially provides the communication foundation for innovations in the consensus or
Abstract: The focus of applied cryptography is the security of cryptographic systems used in practice. This includes analyzing cryptographic protocols and primitives used in the wild, and designing and deploying secure systems. Unfortunately, this is a challenging task. Cryptography is highly brittle and small design or implementation mistakes can have devastating effects on a system
Abstract: This seminar spotlights the latest research aimed at boosting aviation security. As the geopolitical landscape shifts, it has exposed key vulnerabilities, especially in digital communications and navigation systems. We will delve into ongoing research efforts focused on strengthening aviation security by pinpointing these vulnerabilities and crafting effective countermeasures to address various potential threats. Join
Abstract: The success of deep learning in many application domains has been nothing short of dramatic. This has brought the spotlight onto security and privacy concerns with machine learning (ML). One such concern is the threat of model theft. I will discuss our work on exploring the threat of model theft, especially in the form
Abstract: The Arm Advanced eXtensible Interface (AXI) protocols are widely used in bus implementations that connect processors, accelerators, memories, and other IP cores. Any bugs in these implementations can pose a security risk to the correctness of the connected IPs. Malicious or third-party IPs can use such implementation bugs to bypass security mechanisms employed at