Finisky Garden
NLP, Software Engineering, Product Design
Home
NLP/AI
MongoDB
Tags
Categories
Archives
Links
About
0%
Theme NexT works best with JavaScript enabled
NLP
Category
2026
05-12
NV-Embed: Training a Decoder-only Embedding Model with Latent Attention Pooling
05-12
ICLR 2025 Outstanding Paper: Shallow Safety Alignment Only Modifies the First Few Tokens
05-10
NeurIPS 2025 Best Paper Gated Attention: Non-linearity, Sparsity, and Attention Sink
05-09
MRL: Nest Embeddings During Training, Truncate at Inference
05-06
GritLM: A Single LLM for Both Embedding and Generation
05-06
HyDE: Retrieval via LLM-Generated Hypothetical Documents
05-03
LLM2Vec-Gen: Embeddings That Encode the Model's Answer, Not the Query
05-02
ICLR 2026 Outstanding Paper: LLMs Get Lost in Multi-Turn Conversation
04-30
Differentiable Search Index: A Brief Reading
04-29
Transformer Succinctness: The Other Side of Expressiveness
04-25
LLM2Vec: Turning Decoder LLMs into Embedding Models
04-22
Thoughts on Training Data Depletion and Quality Degradation
04-22
Claude Code: 1.6% AI, 98.4% Scaffolding
04-17
The Hivemind of Language Models
04-16
From RAG to Knowledge Compilation
04-15
Theoretical Ceiling of Vector Retrieval
04-13
How Claude Dreams: Background Memory Defragmentation
04-09
Three Evolutions of Agent Engineering
04-07
Context Management in Claude Code vs OpenClaw
04-07
Foundation Models Plateau, Applications Take Off
1
2