Impact of FlowKV and SnapKV Integration on Llama-3-8B Retrieval Performance in BEIR Under Dynamic Workloads

Given the large number of Hindi speakers worldwide, there is a pressing need for robust and efficient information retrieval systems for Hindi. Despite ongoing research, there is a lack of comprehensive benchmark for evaluating retrieval models in Hindi. To address this gap, we introduce the Hindi version of the BEIR benchmark, which includes a subset of English BEIR datasets translated to Hindi, existing Hindi retrieval datasets, and synthetically created datasets for retrieval. The benchmark is comprised of \$15\$ datasets spanning across \$8\$ distinct tasks. We evaluate state-of-the-art multili

Research goal: How does the integration of FlowKV versus SnapKV with Llama-3-8B impact memory-efficient retrieval performance on the BEIR benchmark when evaluated under dynamic workload conditions (e.g., varying query complexity)?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 7.8/10.