Krum Arnaudov
krumeto
AI & ML interests
None yet
Recent Activity
reacted
to
singhsidhukuldeep's
post
with 👍
28 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?
new activity
about 2 months ago
INSAIT-Institute/BgGPT-Gemma-2-27B-IT-v1.0-GGUF:Information about performance of different quantisation options
liked
a model
about 2 months ago
INSAIT-Institute/BgGPT-Gemma-2-27B-IT-v1.0-GGUF
Organizations
None yet
krumeto's activity
Information about performance of different quantisation options
#1 opened about 2 months ago
by
krumeto
Reason for `trust_remote_code`?
2
#7 opened 9 months ago
by
krumeto
Update README.md with correct model name in "Direct use for inference"
#3 opened about 1 year ago
by
krumeto
Intuition for quality decrease after quantization
4
#23 opened about 1 year ago
by
krumeto
Strategies for long documents - document-level context?
1
#2 opened about 1 year ago
by
krumeto
Context Length
2
#7 opened about 1 year ago
by
mrfakename
Is the context length same as Mistral (8k)?
2
#1 opened about 1 year ago
by
krumeto
Performance and latency vs. GPTQ
1
#3 opened about 1 year ago
by
krumeto
Question on the cc-by-nc-sa-4.0 Licence
3
#1 opened about 1 year ago
by
krumeto
Question on the cc-by-nc-sa-4.0 Licence
3
#1 opened about 1 year ago
by
krumeto