Mistral 7B
Emerging5papers using it
2024first seen
'Mistral-7B' is a large language model benchmark used to evaluate the performance of compression methods like SoLA on various language modeling and downstream tasks.
Papers using Mistral 7B (5)
- SoLA: Leveraging Soft Activation Sparsity and Low-Rank Decomposition for Large Language Model CompressionLatentGuard: Controllable Latent Steering for Robust Refusal of Attacks and Reliable Response GenerationTensorized Clustered LoRA Merging for Multi-Task InterferenceGradient-Adaptive Policy Optimization: Towards Multi-Objective Alignment of Large Language ModelsTODO: Enhancing LLM Alignment with Ternary Preferences