mohsenfayyaz / GlobEnc
[NAACL 2022] GlobEnc: Quantifying Global Token Attribution by Incorporating the Whole Encoder Layer in Transformers
☆21Updated last year
Related projects ⓘ
Alternatives and complementary repositories for GlobEnc
- Measuring the Mixing of Contextual Information in the Transformer☆25Updated last year
- DecompX: Explaining Transformers Decisions by Propagating Token Decomposition☆15Updated last year
- EMNLP 2021 - Frustratingly Simple Pretraining Alternatives to Masked Language Modeling☆31Updated 2 years ago
- Code of NAACL 2022 "Efficient Hierarchical Domain Adaptation for Pretrained Language Models" paper.☆32Updated last year
- Influence Experiments☆36Updated last year
- ☆57Updated 2 years ago
- DEMix Layers for Modular Language Modeling☆53Updated 3 years ago
- ☆32Updated 2 years ago
- Rationales for Sequential Predictions☆40Updated 2 years ago
- ☆37Updated 3 years ago
- ☆23Updated 2 years ago
- This repository accompanies our paper “Do Prompt-Based Models Really Understand the Meaning of Their Prompts?”☆84Updated 2 years ago
- [EMNLP 2020] Collective HumAn OpinionS on Natural Language Inference Data☆33Updated 2 years ago
- ☆42Updated 4 years ago
- Research code for the paper "How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models"☆26Updated 3 years ago
- Automatic metrics for GEM tasks☆61Updated 2 years ago
- ☆22Updated 3 years ago
- Code for ACL 2022 paper "Expanding Pretrained Models to Thousands More Languages via Lexicon-based Adaptation"☆31Updated 2 years ago
- A library for parameter-efficient and composable transfer learning for NLP with sparse fine-tunings.☆70Updated 3 months ago
- This is a repository with the code for the EMNLP 2020 paper "Information-Theoretic Probing with Minimum Description Length"☆69Updated 2 months ago
- ☆77Updated 6 months ago
- No Parameters Left Behind: Sensitivity Guided Adaptive Learning Rate for Training Large Transformer Models (ICLR 2022)☆29Updated 2 years ago
- Debiasing Methods in Natural Language Understanding Make Bias More Accessible: Code and Data☆13Updated 2 years ago
- ☆42Updated 3 years ago
- The official implemetation of "Evidentiality-guided Generation for Knowledge-Intensive NLP Tasks" (NAACL 2022).☆43Updated last year
- ☆13Updated 2 years ago
- Query-focused summarization data☆41Updated last year
- ☆34Updated 5 months ago
- Code for paper "Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language?"☆20Updated 4 years ago
- The official code of EMNLP 2022, "SCROLLS: Standardized CompaRison Over Long Language Sequences".☆68Updated 10 months ago