Memory-Efficient Incremental Clustering for Long-Text Coreference Resolution
Published:Dec 31, 2025 08:26
•1 min read
•ArXiv
Analysis
This paper addresses the challenge of coreference resolution in long texts, a crucial area for LLMs. It proposes MEIC-DT, a novel approach that balances efficiency and performance by focusing on memory constraints. The dual-threshold mechanism and SAES/IRP strategies are key innovations. The paper's significance lies in its potential to improve coreference resolution in resource-constrained environments, making LLMs more practical for long documents.
Key Takeaways
- •Proposes MEIC-DT, a novel approach for memory-efficient incremental clustering.
- •Employs a dual-threshold constraint mechanism to manage Transformer input scale.
- •Introduces SAES for intelligent cache management.
- •Implements IRP to condense clusters and preserve semantic integrity.
- •Achieves competitive performance under memory constraints.
Reference
“MEIC-DT achieves highly competitive coreference performance under stringent memory constraints.”