Papers with sparse methods
ReSCORE: Label-free Iterative Retriever Training for Multi-hop Question Answering with Relevance-Consistency Supervision (2025.acl-long)
Copied to clipboard
| Challenge: | Multi-hop question answering requires reasoning across multiple documents to answer complex questions. |
| Approach: | They propose a method for training dense retrievers for multi-hop question answering . they leverage large language models to measure document-question relevance with answer consistency . their results lead to state-of-the-art Exact Match and F1 scores for MHQA . |
| Outcome: | Evaluated on three MHQA benchmarks, the proposed method improves retrieval performance . it leads to state-of-the-art Exact Match and F1 scores for the proposed technique . |
Salient Phrase Aware Dense Retrieval: Can a Dense Retriever Imitate a Sparse One? (2022.findings-emnlp)
Copied to clipboard
Xilun Chen, Kushal Lakhotia, Barlas Oguz, Anchit Gupta, Patrick Lewis, Stan Peshterliev, Yashar Mehdad, Sonal Gupta, Wen-tau Yih
| Challenge: | Existing sparse retrievers lack the ability to match salient phrases and rare entities in the query. |
| Approach: | They introduce a dense Lexical Model that can be trained to imitate a sparse one. |
| Outcome: | The proposed model outperforms sparse retrievers on a range of tasks including five question answering datasets and the MS MARCO passage retrieval. |
ClusterFormer: Neural Clustering Attention for Efficient and Effective Transformer (2022.acl-long)
Copied to clipboard
| Challenge: | Existing sparse attention methods use fixed patterns to select words without considering similarities between words. |
| Approach: | They propose a neural clustering method which integrates into the Self-Attention Mechanism in Transformer and integrates it into the target task. |
| Outcome: | The proposed method outperforms two typical sparse attention methods on translation, text classification, and text matching tasks while having a comparable or even better time and memory efficiency. |
Latent-Condensed Transformer for Efficient Long Context Modeling (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches address these bottlenecks separately: Multi-head Latent Attention (MLA) reduces the KV cache by projecting tokens into a low-dimensional latent space, while sparse attention reduces computation. |
| Approach: | They propose a Latent-Condensed Attention mechanism that performs structured context condensation directly within MLA's latent space. |
| Outcome: | The proposed approach reduces KV cache size and attention cost without adding parameters. |