Papers with Code 아카이브
Knowledge Distillation
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 2
- 평가표
- 11
- 보존된 평가 기록
- 90
Knowledge distillation is the process of transferring knowledge from a large model to a smaller one. While large models (such as very deep neural networks or ensembles of many models) have higher knowledge capacity than small models, this capacity might not be fully utilized.
코드
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
https://github.com/hkuds/lightgnn
공식 코드https://github.com/hkuds/promptmm
공식 코드PromptMM: Multi-Modal Knowledge Distillation for Recommendation with Prompt-Tuning
https://github.com/hkuds/simrec
공식 코드https://github.com/hkunlp/progen
공식 코드ProGen: Progressive Zero-shot Dataset Generation via In-context Feedback
https://github.com/hkust-knowcomp/abspyramid
공식 코드AbsPyramid: Benchmarking the Abstraction Ability of Language Models with a Unified Entailment Graph
https://github.com/hkust-knowcomp/cskb-population
공식 코드Benchmarking Commonsense Knowledge Base Population with an Effective Evaluation Dataset
https://github.com/hkust-knowcomp/efo-1-qa-benchmark
공식 코드Benchmarking the Combinatorial Generalizability of Complex Query Answering on Knowledge Graphs
https://github.com/hkust-knowcomp/mars
공식 코드https://github.com/hkust-longgroup/awesome-mllm-benchmarks
공식 코드A Survey on Multimodal Benchmarks: In the Era of Large AI Models
https://github.com/hkust-nlp/agentboard
공식 코드AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents
https://github.com/hkust-nlp/felm
공식 코드FELM: Benchmarking Factuality Evaluation of Large Language Models
https://github.com/hkustdial/deepfund
공식 코드Time Travel is Cheating: Going Live with DeepFund for Real-Time Fund Investment Benchmarking
https://github.com/hkzhang91/cabvit
공식 코드Fcaformer: Forward Cross Attention in Hybrid Vision Transformer
https://github.com/hkzhang91/fcaformer
공식 코드Fcaformer: Forward Cross Attention in Hybrid Vision Transformer
https://github.com/hl-hanlin/smkd
공식 코드Supervised Masked Knowledge Distillation for Few-Shot Transformers
https://github.com/hlt-mt/fbk-neutr-eval
공식 코드Hi Guys or Hi Folks? Benchmarking Gender-Neutral Machine Translation with the GeNTE Corpus
https://github.com/hlt-mt/speech-massive
공식 코드Speech-MASSIVE: A Multilingual Speech Dataset for SLU and Beyond
https://github.com/hltcoe/colbert-x
공식 코드Translate-Distill: Learning Cross-Language Dense Retrieval by Translation and Distillation
https://github.com/hly1998/brcd
공식 코드Bit-mask Robust Contrastive Knowledge Distillation for Unsupervised Semantic Hashing
https://github.com/hnuzhy/ssda-yolo
공식 코드SSDA-YOLO: Semi-supervised Domain Adaptive YOLO for Cross-Domain Object Detection
https://github.com/hoiliu-0801/monotakd
공식 코드MonoTAKD: Teaching Assistant Knowledge Distillation for Monocular 3D Object Detection
https://github.com/holistic-ai/SAGED-Bias
공식 코드https://github.com/holistic-ai/THaMES
공식 코드THaMES: An End-to-End Tool for Hallucination Mitigation and Evaluation in Large Language Models
https://github.com/homzer/MCC-KD
공식 코드https://github.com/hongw579/agkd-bml
공식 코드https://github.com/hoqolo/sdstrack
공식 코드https://github.com/horsepurve/StyleSegor
공식 코드Neural Style Transfer Improves 3D Cardiovascular MR Image Segmentation on Inconsistent Data
https://github.com/howiehwong/awareness-in-llm
공식 코드I Think, Therefore I am: Benchmarking Awareness of Large Language Models Using AwareBench
https://github.com/howiehwong/unigen
공식 코드UniGen: A Unified Framework for Textual Dataset Generation Using Large Language Models
https://github.com/hpai-bsc/prompt_engine
공식 코드https://github.com/hpc-lab-koreatech/tie-kd
공식 코드TIE-KD: Teacher-Independent and Explainable Knowledge Distillation for Monocular Depth Estimation
https://github.com/hpclab/efficient_nn_for_ltr
공식 코드https://github.com/hpi-xnor/bnext
공식 코드Join the High Accuracy Club on ImageNet with A Binary Neural Network Ticket
https://github.com/hpnair/18663_Project_FBNet
공식 코드https://github.com/hq-deng/RD4AD
공식 코드Anomaly Detection via Reverse Distillation from One-Class Embedding
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.