Open-vocabulary object detectors (OVODs) unify vision and language to detect arbitrary object categories based on text prompts, enabling strong zero-shot generalization to novel concepts. As these models gain traction in high-stakes applications such as robotics, autonomous driving, and surveillance, understanding their security risks becomes crucial. In this work, we conduct the first study of backdoor attacks on OVODs and reveal a new attack surface introduced by prompt tuning. We propose TrAP (Trigger-Aware Prompt tuning), a multi-modal backdoor injection strategy that jointly optimizes prompt parameters in both image and text modalities along with visual triggers. TrAP enables the attacker to implant malicious behavior using lightweight, learnable prompt tokens without retraining the base model weights, thus preserving generalization while embedding a hidden backdoor. We adopt a curriculum-based training strategy that progressively shrinks the trigger size, enabling effective backdoor activation using small trigger patches at inference. Experiments across multiple datasets show that TrAP achieves high attack success rates for both object misclassification and object disappearance attacks, while also improving clean image performance on downstream datasets compared to the zero-shot setting.

Backdoor attacks embed a hidden functionality into deep neural networks, causing the network to display anomalous behavior when activated by a predetermined pattern in the input (Trigger), while behaving well otherwise on public test data. Recent works have shown that backdoored face recognition (FR) systems can respond to natural-looking triggers like a particular pair of sunglasses. Such attacks pose a serious threat to the applicability of FR systems in high-security applications. We propose a novel technique to (1) detect whether an FR network is compromised with a natural, physically realizable trigger, and (2) identify such triggers given a compromised network. We demonstrate the effectiveness of our methods with a compromised FR network, where we are able to identify the trigger (e.g. green-sunglasses or redbowtie) with a top-5 accuracy of 74%, whereas a naïve brute force baseline achieves 56% accuracy.

The presented work focuses on automatic recognition of object classes while ensuring near real-time training required for recognizing a new object not seen previously. This is achieved by proposing a two-stage hierarchical deep learning framework which first learns object categories using a Nearest Class Mean (NCM) classifier applied directly to CNN features and then, uses a two-layer artificial neural network to learn the object labels within each category. In order to recognize a new object not seen earlier, the category is identified first and then the second stage neural network is incrementally trained with the features of the new object without forgetting previously learnt labels. The proposed hierarchical framework is shown to provide comparable recognition accuracy with significant reduction in overall computational time in recognizing new objects compared to methods that use end-to-end re-training. The efficacy of the approach is demonstrated through comparison with existing state-of-the-art methods on the publicly available CORe50 dataset.

One of the key challenges in G.fast is to minimize the power consumption at distribution points. G.fast standards define Discontinuous Operation modes that provide avenues for power reduction by allowing intermittent transmission of users along time slots. In this paper, we formulate a power efficiency problem as a user-slot assignment problem, where we schedule users to time slots during discontinuous operation such that the total power consumption is minimized. Since the general user-slot assignment is NP hard, we propose a weighted$A^{*}$algorithm based solution that achieves reasonable performance with limited computational resources. Further, the proposed user-slot assignment is also G.fast standards compliant and therefore can be implemented in practice. We also explore different precoding as well as user grouping strategies that can be employed while performing user-slot allocation. Finally, the main insight of this work is that on using our user-slot assignment algorithm and by only precoding during the normal operations, we achieve energy efficiency levels comparable to those achieved when precoding is applied during the discontinuous operations without the suggested user-slot allocation.

To enable energy efficiency, G.fast standards define discontinuous operations (DO) where a set of users can remain inactive while others transmit during a time domain duplex (TDD) frame. In this work, we investigate energy efficient discontinuous operations (DO) by scheduling users to time slots, such that the total energy consumption is minimized while satisfying the individual data rate constraints. Since the user-slot assignment problem is NP complete in nature, we propose the use of weighted A* algorithm that achieves reasonable performance with limited computational resources. The main insight of this work is that on using our user-slot assignment algorithm and by only precoding during the normal operations, we achieve the same energy efficiency as achieved by precoding strategies like discontinuous vectoring [1] while satisfying the provisions of the G.fast standard.

In this work we address the problem of recovering sparse solutions to non-linear inverse problems. We look at two variants of the basic problem - the synthesis prior problem when the solution is sparse and the analysis prior problem where the solution is co-sparse in some linear basis. For the first problem, we propose non-linear variants of the Orthogonal Matching Pursuit (OMP) and CoSamp algorithms, for the second problem we propose a non-linear variant of the Greedy Analysis Pursuit (GAP) algorithm. We empirically test the success rates of our algorithms on exponential and logarithmic functions.