Papers with Code 아카이브
LAMBADA
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 0
- 평가표
- 1
- 보존된 평가 기록
- 2
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
AdaGC: Improving Training Stability for Large Language Model Pretraining
Explaining and Improving Contrastive Decoding by Extrapolating the Probabilities of a Huge and Hypothetical LM
Beyond Autoregression: Fast LLMs via Self-Distillation Through Time
Spectra: Surprising Effectiveness of Pretraining Ternary Language Models at Scale
SymBa: Symbolic Backward Chaining for Structured Natural Language Reasoning
PIXAR: Auto-Regressive Language Modeling in Pixel Space
Concise and Organized Perception Facilitates Reasoning in Large Language Models
Headless Language Models: Learning without Predicting with Contrastive Weight Tying
Stay on topic with Classifier-Free Guidance
Inconsistencies in Masked Language Models
LAMBADA: Backward Chaining for Automated Reasoning in Natural Language
Training Compute-Optimal Large Language Models
Scaling Language Models: Methods, Analysis & Insights from Training Gopher
CoreLM: Coreference-aware Language Model Fine-Tuning
The Stability-Efficiency Dilemma: Investigating Sequence Length Warmup for Training GPT Models
E.T.: Entity-Transformers. Coreference augmented Neural Language Model for richer mention representations via Entity-Transformer blocks
Residual Shuffle-Exchange Networks for Fast Processing of Long Sequences
Attending to Entities for Better Text Understanding
Not Enough Data? Deep Learning to the Rescue!
Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
Neural Shuffle-Exchange Networks -- Sequence Processing in O(n log n) Time
Entity Tracking Improves Cloze-style Reading Comprehension
Universal Transformers
Neural Models for Reasoning over Multiple Mentions using Coreference
Linguistic Knowledge as Memory for Recurrent Neural Networks
Broad Context Language Modeling as Reading Comprehension
The LAMBADA dataset: Word prediction requiring a broad discourse context
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.