| Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | ✓ Link | 70.7 | | CoA | 2024-03-26 |
| Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | ✓ Link | 64.7 | | CoA w/o actions | 2024-03-26 |
| DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines | ✓ Link | 59.4 | | DSP | 2023-10-05 |
| Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | ✓ Link | 59.4 | | DSP | 2024-03-26 |
| FiE: Building a Global Probability Space by Leveraging Early Fusion in Encoder for Open-Domain Question Answering | | 56.3 | | FiE+PAQ | 2022-11-18 |
| FiE: Building a Global Probability Space by Leveraging Early Fusion in Encoder for Open-Domain Question Answering | | 52.4 | | FiE | 2022-11-18 |
| FiDO: Fusion-in-Decoder optimized for stronger performance and faster inference | | 51.1 | | FiDO | 2022-12-15 |
| Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks | ✓ Link | 45.2 | | RAG | 2020-05-22 |
| Language Models are Few-Shot Learners | ✓ Link | 44.7 | | Few-shot | 2020-05-28 |
| Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | ✓ Link | 44.7 | | Few-shot | 2024-03-26 |
| PaLM: Scaling Language Modeling with Pathways | ✓ Link | 43.5 | | PaLM-540B (Few-Shot) | 2022-04-05 |
| Language Models are Unsupervised Multitask Learners | ✓ Link | 43 | | Zero-shot | 2019-02-14 |
| Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | ✓ Link | 43 | | Zero-shot | 2024-03-26 |
| Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer | ✓ Link | 42.8 | | T5.1.1-XXL+SSM | 2019-10-23 |
| Chain-of-Thought Prompting Elicits Reasoning in Large Language Models | ✓ Link | 42.5 | | CoT | 2022-01-28 |
| Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | ✓ Link | 42.5 | | CoT | 2024-03-26 |
| Dense Passage Retrieval for Open-Domain Question Answering | ✓ Link | 42.4 | | DPR | 2020-04-10 |
| Language Models are Few-Shot Learners | ✓ Link | 41.5 | | GPT-3-175B (Few-Shot) | 2020-05-28 |
| REALM: Retrieval-Augmented Language Model Pre-Training | ✓ Link | 40.7 | | REALM | 2020-02-10 |
| ReAct: Synergizing Reasoning and Acting in Language Models | ✓ Link | 38.3 | | React | 2022-10-06 |
| Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | ✓ Link | 38.3 | | React | 2024-03-26 |
| Latent Retrieval for Weakly Supervised Open Domain Question Answering | ✓ Link | 36.4 | | ORQA | 2019-06-01 |
| Measuring and Narrowing the Compositionality Gap in Language Models | ✓ Link | 31.1 | | Self-Ask | 2022-10-07 |
| Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | ✓ Link | 31.1 | | Self-Ask | 2024-03-26 |
| PaLM 2 Technical Report | ✓ Link | 28.2 | | PaLM 2-L (one-shot) | 2023-05-17 |
| PaLM 2 Technical Report | ✓ Link | 26.9 | | PaLM 2-M (one-shot) | 2023-05-17 |
| Tree of Thoughts: Deliberate Problem Solving with Large Language Models | ✓ Link | 26.3 | | ToT | 2023-05-17 |
| Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | ✓ Link | 26.3 | | ToT | 2024-03-26 |
| Language Models are Few-Shot Learners | ✓ Link | 25.3 | | GPT-3-175B (One-Shot) | 2020-05-28 |
| PaLM: Scaling Language Modeling with Pathways | ✓ Link | 22.6 | | PaLM-540B (One-Shot) | 2022-04-05 |
| PaLM 2 Technical Report | ✓ Link | 21.8 | | PaLM 2-S (one-shot) | 2023-05-17 |
| GLaM: Efficient Scaling of Language Models with Mixture-of-Experts | | 15.5 | | GLaM 62B/64E (Zero-Shot) | 2021-12-13 |
| Language Models are Few-Shot Learners | ✓ Link | 14.4 | | GPT-3-175B (Zero-Shot) | 2020-05-28 |
| PaLM: Scaling Language Modeling with Pathways | ✓ Link | 10.6 | | PaLM-540B (Zero-Shot) | 2022-04-05 |
| Large-scale Simple Question Answering with Memory Networks | ✓ Link | | 42.2% | Memory Networks (ensemble) | 2015-06-05 |
| Question Answering with Subgraph Embeddings | ✓ Link | | 39.2% | Subgraph embeddings | 2014-06-14 |
| Open Question Answering with Weakly Supervised Embedding Models | | | 29.7% | Weakly Supervised Embeddings | 2014-04-16 |