We study whether safety directions learned in one language model can transfer across representation spaces to steer heterogeneous visual generators, reducing unsafe generations through benign-only alignment while preserving generation utility.

CounterVid introduces an agentic framework combining VLM reasoning, image editing, and video diffusion to generate counterfactual training pairs that mitigate object, action, and temporal hallucinations. The work was developed at Amazon Prime Video and accepted at EMNLP 2026.

We propose the prefilling attack, a structured natural-language prefix prepended to the model output, to steer the model to respond with a clean, valid option in multiple-choice question answering tasks, significantly improving accuracy and calibration.

HySAC, Hyperbolic Safety-Aware CLIP, models hierarchical safety relations to enable effective retrieval of unsafe content, dynamically redirecting it to safer alternatives for enhanced content moderation.

Vision-and-language model designed to mitigate the risks associated with NSFW content in AI applications. Safe-CLIP is fine-tuned to serve the association between linguistic and visual concepts, ensuring safer outputs in text-to-image and image-to-text retrieval and generation tasks.

A study of shortcut bias induced by ArUco markers in RGB-based 6-DoF object pose estimation, proposing a generative data augmentation pipeline that removes markers and synthesizes realistic backgrounds to improve robustness in ArUco-free settings.

A study on aligning Large Language Models (LLMs) for offensive content removal, introducing the CAiA dataset and Safe-CLIP to ensure ethical and responsible AI-driven content moderation.

A deep learning pipeline for DeepFake detection, leveraging custom datasets and an optimized Xception-based CNN to identify manipulated facial images with high accuracy.

A comprehensive study of data labeling tools and techniques, evaluating their efficiency in person and traffic sign detection, as well as 3D pose estimation for robotic applications.