14 papers
Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks
Nobin Sarwar, Shubhashis Roy Dipta, Zheyuan Liu +1
With the growing adoption of VLMs, DMs, LLMs, and AFMs, these multimodal foundation models can inadvertently encode sensitive, copyrighted, biased, or unsafe cross-modal associatio…
Flash-BoN: Instant Drafts for Inference-Time Scaling in Diffusion Models
Ruchit Rawal, Reza Shirkavand, Sayak Paul +5
Inference-time scaling for text-to-image generation has progressed from simple Best-of- (BoN) sampling to guided search methods that verify and steer candidate trajectories at i…
Capability Self-Assessment: Teaching LLMs to Know Their Limits
Haoyan Yang, Reza Shirkavand, Yukai Jin +3
The ability to recognize one's own limitations and decide whether to solve a problem or delegate is fundamental for reliable intelligent systems. Yet we show that modern large lang…
Privacy-Preserving LLMs Routing
Xidong Wu, Yukuan Zhang, Yuqiong Ji +3
Large language model (LLM) routing has emerged as a critical strategy to balance model performance and cost-efficiency by dynamically selecting services from various model provider…
Catalog-Native LLM: Speaking Item-ID Dialect with Less Entanglement for Recommendation
Reza Shirkavand, Xiaokai Wei, Chen Wang +3
While collaborative filtering delivers predictive accuracy and efficiency, and Large Language Models (LLMs) enable expressive and generalizable reasoning, modern recommendation sys…
ToMoE: Converting Dense Large Language Models to Mixture-of-Experts through Dynamic Structural Pruning
Shangqian Gao, Ting Hua, Reza Shirkavand +10
Large Language Models (LLMs) have demonstrated remarkable abilities in tackling a wide range of complex tasks. However, their huge computational and memory costs raise significant…