Papers with being-trained model
CuSINeS: Curriculum-driven Structure Induced Negative Sampling for Statutory Article Retrieval (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods to assess Statutory Article Retrieval (SAR) are vague and underspecified, resulting in a lack of clarity and a gap between legal expertise and public comprehension. |
| Approach: | They propose a negative sampling approach to enhance the performance of Statutory Article Retrieval (SAR) it employs a curriculum-based negative sampling strategy guiding the model to focus on easier negatives initially and progressively tackle more difficult ones. |
| Outcome: | The proposed approach surpasses static methods and can be used to assess the difficulty of the model. |
Reduce Catastrophic Forgetting of Dense Retrieval Training with Teleportation Negatives (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent research shows strong empirical advantages of dense retrieval in various information access scenarios, such as openQA. |
| Approach: | They propose a method which accumulates momentum negatives from past iterations and approximates future iteration with lookahead negatives as "teleportations" on web search and OpenQA, ANCE-Tele outperforms previous state-of-the-art systems of similar size and eliminates the dependency on sparse retrieval negatives. |
| Outcome: | The proposed method outperforms previous state-of-the-art systems on web search and OpenQA and is competitive among systems with significantly more parameters. |