Zzong's Notes

최근 작성한 노트

노트 37개

4개

  1. Elasticsearch 검색 과정
    retrieval/search_engine/Elasticsearch 검색 과정
  2. Embedding Dimension Rule of Thumb
    machine_learning/feature_engineering/embedding dimension rule of thumb
  3. VLM 임베딩 추출 방법
    retrieval/dense/VLM 임베딩 추출 방법
  4. Focal Loss와 Gradient Blending
    machine_learning/loss/Focal Loss와 Gradient Blending

4개

  1. Bi-Encoder vs Cross-Encoder
    retrieval/concepts/Bi-Encoder vs Cross-Encoder
  2. Entire Space Multi-Task Model - An Effective Approach for Estimating Post-Click Conversion Rate
    papers/advertisement/Entire Space Multi-Task Model - An Effective Approach for Estimating Post-Click Conversion Rate
  3. Geo Search
    retrieval/search_engine/Geo Search
  4. Prefix Search
    retrieval/search_engine/Prefix Search

1개

  1. Progressive Alignment of Recommender Foundation Model through Multi-Phase Post-Training
    papers/recommender_system/Progressive Alignment of Recommender Foundation Model through Multi-Phase Post-Training

4개

  1. Berkeley Function Calling Leaderboard (BFCL)
    generative_ai/agent/Berkeley Function Calling Leaderboard
  2. BrowseComp
    generative_ai/agent/BrowseComp
  3. LLM 에이전트·도구 사용 벤치마크 비교
    generative_ai/agent/LLM Agent Benchmarks
  4. τ-bench
    generative_ai/agent/tau-bench

2개

  1. Permutations
    algorithms/problems/Permutations
  2. Generate Parentheses
    algorithms/problems/Generate Parentheses

2개

  1. Brier loss
    machine_learning/loss/Brier loss
  2. Jev와 RLCD
    generative_ai/LLM/Jev

1개

  1. 분류 모델의 학습과 평가
    machine_learning/metrics/분류 모델의 학습과 평가

1개

  1. Decoupled Clip and Dynamic sAmpling Policy Optimization (DAPO)
    generative_ai/post_training/DAPO

2개

  1. Hash-Based Embedding
    recommendation_system/hash-based embedding
  2. On-Policy Distillation (OPD)
    generative_ai/post_training/on-policy distillation

1개

  1. Generalized Knowledge Distillation (GKD)
    generative_ai/post_training/GKD

2개

  1. OneSearch - End-to-End Generative E-commerce Search
    papers/e-commerce/OneSearch - End-to-End Generative E-commerce Search
  2. Prompt Compression Trends - From Token Pruning to Context Engineering
    generative_ai/inference/Prompt Compression Trends

1개

  1. LLMLingua - Prompt Compression for LLM Inference
    generative_ai/inference/LLMLingua

1개

  1. TabFM
    machine_learning/deep_learning/TabFM

5개

  1. KaLM-Reranker-V1 - Fast but Not Late Interaction for Compressed Document Reranking
    papers/retrieval/KaLM-Reranker-V1 - Fast but Not Late Interaction for Compressed Document Reranking
  2. Ko-WideSearch - A Korean Breadth-Search Benchmark for Exhaustive Set Enumeration by Web Agents
    papers/language_model/Ko-WideSearch - Korean Breadth-Search Benchmark
  3. Prefill
    generative_ai/inference/Prefill
  4. FP8 Quantization
    generative_ai/inference/FP8 Quantization
  5. Group Sequence Policy Optimization (GSPO)
    generative_ai/post_training/GSPO

3개

  1. Apache Flink
    data_engineering/Apache Flink
  2. TLPO - Token-Level Policy Optimization
    papers/language_model/TLPO - Token-Level Policy Optimization for Mitigating Language Confusion in Large Language Models
  3. Qwen-AgentWorld - Language World Models for General Agents
    papers/language_model/Qwen-AgentWorld - Language World Models for General Agents

1개

  1. 자연스러운 한국어를 위한 LLM Post-Training
    generative_ai/post_training/LLM Post-Training for Natural Korean

1개

  1. Semantic Retrieval for Product Search in E-Commerce
    papers/e-commerce/Semantic Retrieval for Product Search in E-Commerce

1개

  1. Trinity, 여러 LLM을 조율하는 진화된 코디네이터
    papers/language_model/Trinity - An Evolved LLM Coordinator