Srihari Thyagarajan

Haleshot

AI & ML interests

AI, ML, DL, CV, RecSys

Recent Activity

liked a Space 8 days ago
franciszzj/Leffa
reacted to singhsidhukuldeep's post with πŸš€ 13 days ago
Exciting breakthrough in AI Recommendation Systems! Just read a fascinating paper from Meta AI and UW-Madison researchers on unifying generative and dense retrieval methods for recommendations. The team introduced LIGER (LeveragIng dense retrieval for GEnerative Retrieval), a novel hybrid approach that combines the best of both worlds: Key Technical Innovations: - Integrates semantic ID-based generative retrieval with dense embedding methods - Uses a T5 encoder-decoder architecture with 6 layers, 6 attention heads, and 128-dim embeddings - Processes item attributes through sentence-T5-XXL for text representations - Employs a dual-objective training approach combining cosine similarity and next-token prediction - Implements beam search with size K for candidate generation - Features an RQ-VAE with 3-layer MLP for semantic ID generation Performance Highlights: - Significantly outperforms traditional methods on cold-start recommendations - Achieves state-of-the-art results on major benchmark datasets (Amazon Beauty, Sports, Toys, Steam) - Reduces computational complexity from O(N) to O(tK) where t is semantic ID count - Maintains minimal storage requirements while improving recommendation quality The most impressive part? LIGER effectively solves the cold-start problem that has long plagued recommendation systems while maintaining computational efficiency. This could be a game-changer for e-commerce platforms and content recommendation systems! What are your thoughts on hybrid recommendation approaches?
reacted to singhsidhukuldeep's post with πŸ‘ 13 days ago
Exciting breakthrough in AI Recommendation Systems! Just read a fascinating paper from Meta AI and UW-Madison researchers on unifying generative and dense retrieval methods for recommendations. The team introduced LIGER (LeveragIng dense retrieval for GEnerative Retrieval), a novel hybrid approach that combines the best of both worlds: Key Technical Innovations: - Integrates semantic ID-based generative retrieval with dense embedding methods - Uses a T5 encoder-decoder architecture with 6 layers, 6 attention heads, and 128-dim embeddings - Processes item attributes through sentence-T5-XXL for text representations - Employs a dual-objective training approach combining cosine similarity and next-token prediction - Implements beam search with size K for candidate generation - Features an RQ-VAE with 3-layer MLP for semantic ID generation Performance Highlights: - Significantly outperforms traditional methods on cold-start recommendations - Achieves state-of-the-art results on major benchmark datasets (Amazon Beauty, Sports, Toys, Steam) - Reduces computational complexity from O(N) to O(tK) where t is semantic ID count - Maintains minimal storage requirements while improving recommendation quality The most impressive part? LIGER effectively solves the cold-start problem that has long plagued recommendation systems while maintaining computational efficiency. This could be a game-changer for e-commerce platforms and content recommendation systems! What are your thoughts on hybrid recommendation approaches?
View all activity

Organizations

Hugging Face Discord Community's profile picture Nerdy Face's profile picture

Haleshot's activity

upvoted an article 2 months ago
view article
Article

Fine-tuning LLMs to 1.58bit: extreme quantization made easy

β€’ 213
upvoted an article 3 months ago
view article
Article

Introducing the Open FinLLM Leaderboard

β€’ 66
upvoted an article 3 months ago
view article
Article

ColPali: Efficient Document Retrieval with Vision Language Models πŸ‘€

By manu β€’
β€’ 181
upvoted an article 4 months ago
upvoted an article 5 months ago
view article
Article

SmolLM - blazingly fast and remarkably powerful

β€’ 292