Search Results for author:
Found 12 papers, 2 papers with code
LLM Misalignment via Adversarial RLHF Platforms
Given the growing adoption of RLHF and open-source RLHF frameworks, we investigate the trustworthiness of these systems and their potential impact on behavior of LLMs.
OverThink: Slowdown Attacks on Reasoning LLMs
We evaluated our attack across closed-(OpenAI o1, o1-mini, o3-mini) and open-(DeepSeek R1) weights reasoning models on the FreshQA and SQuAD datasets.
Riddle Me This! Stealthy Membership Inference for Retrieval-Augmented Generation
Retrieval-Augmented Generation (RAG) enables Large Language Models (LLMs) to generate grounded responses by leveraging external knowledge databases without altering model parameters.
Synthetic Data Can Mislead Evaluations: Membership Inference as Machine Text Detection
Recent work shows membership inference attacks (MIAs) on large language models (LLMs) produce inconclusive results, partly due to difficulties in creating non-member datasets without temporal shifts.
Backdooring Bias into Text-to-Image Models
Furthermore, we show how the current state-of-the-art generative models make this attack both cheap and feasible for any adversary, with costs ranging between $12-$18.
Iteratively Prompting Multimodal LLMs to Reproduce Natural and AI-Generated Images
With the digital imagery landscape rapidly evolving, image stocks and AI-generated image marketplaces have become central to visual media.
Diffence: Fencing Membership Privacy With Diffusion Models
A unique feature of DIFFENCE is that it works on input samples only, without modifying the training or inference phase of the target model.
Understanding (Un)Intended Memorization in Text-to-Image Generative Models
Multimodal machine learning, especially text-to-image models like Stable Diffusion and DALL-E 3, has gained significance for transforming text into detailed images.
Memory Triggers: Unveiling Memorization in Text-To-Image Generative Models through Word-Level Duplication
Diffusion-based models, such as the Stable Diffusion model, have revolutionized text-to-image synthesis with their ability to produce high-quality, high-resolution images.
Stealing the Decoding Algorithms of Language Models
A key component of generating text from modern language models (LM) is the selection and tuning of decoding algorithms.
On Free Energy for Deformed JT Gravity
In this paper, we study a particular deformation of the Jackiw-Teitelboim gravity recently considered by Maxfield, Turiaci and independently by Witten.
High Energy Physics - Theory General Relativity and Quantum Cosmology
On the first law of holographic complexity
In this paper, we examine the proposed first law of holographic complexity through studying different perturbations around various spacetime backgrounds.
High Energy Physics - Theory General Relativity and Quantum Cosmology
