7 papers
When Weak LLMs Speak with Confidence, Preference Alignment Gets Stronger
Amirabbas Afzali, Myeongho Jeon, Maria Brbic
Preference alignment is an essential step in adapting large language models (LLMs) to human values, but existing approaches typically depend on costly human annotations or large-sc…
ULTra: Unveiling Latent Token Interpretability in Transformer-Based Understanding and Segmentation
Hesam Hosseini, Ghazal Hosseini Mighan, Amirabbas Afzali +2
Transformers have revolutionized Computer Vision (CV) through self-attention mechanisms. However, their complexity makes latent token representations difficult to interpret. We int…
LORE: Lagrangian-Optimized Robust Embeddings for Visual Encoders
Borna Khodabandeh, Amirabbas Afzali, Amirhossein Afsharrad +4
Visual encoders have become fundamental components in modern computer vision pipelines. However, ensuring robustness against adversarial perturbations remains a critical challenge.…
Controlling Gender Bias in Retrieval via a Backpack Architecture
Amirabbas Afzali, Amirreza Velae, Iman Ahmadi +1
The presence of social biases in large language models (LLMs) has become a significant concern in AI research. These biases, often embedded in training data, can perpetuate harmful…
One Goal, Many Challenges: Robust Preference Optimization Amid Content-Aware and Multi-Source Noise
Amirabbas Afzali, Amirhossein Afsharrad, Seyed Shahabeddin Mousavi +1
Large Language Models (LLMs) have made significant strides in generating human-like responses, largely due to preference alignment techniques. However, these methods often assume u…
Aligning Visual Contrastive learning models via Preference Optimization
Amirabbas Afzali, Borna Khodabandeh, Ali Rasekh +3
Contrastive learning models have demonstrated impressive abilities to capture semantic similarities by aligning representations in the embedding space. However, their performance c…