DASH-Q uses a stable diagonal curvature estimate and weighted least squares to achieve robust ultra-low-bit post-training quantization of LLMs, improving zero-shot accuracy by 7% on average over baselines.
Winogrande: An adversarial winograd schema challenge at scale
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
dataset 2
citation-polarity summary
roles
dataset 2polarities
use dataset 2representative citing papers
A progressive training scheme with binary-aware initialization and dual-scaling allows pre-trained LLMs to be converted to high-performance 1-bit models without training from scratch.
An analytical post-training method restructures FFNs into MoE by partitioning neurons based on activation patterns and building a router from statistics, achieving 1.17x speedup with minimal resources.
H2O evicts non-heavy-hitter tokens from the KV cache using a dynamic submodular policy, retaining recent and frequent-co-occurrence tokens to reduce memory while preserving accuracy.
GPT-4 as an LLM judge achieves over 80% agreement with human preferences on MT-Bench and Chatbot Arena, matching human agreement levels and providing a scalable evaluation method.
Mistral 7B is a 7B-parameter LLM that outperforms Llama 2 13B across benchmarks via grouped-query attention and sliding-window attention while remaining efficient.
NeuronMLP applies SVD-based compression and Trainium-specific tiling and caching to MLP layers, delivering 1.35x kernel speedup and 1.21x end-to-end inference speedup at 0.05 compression ratio versus AWS NKI baseline.
citing papers explorer
-
Robust Ultra Low-Bit Post-Training Quantization via Stable Diagonal Curvature Estimate
DASH-Q uses a stable diagonal curvature estimate and weighted least squares to achieve robust ultra-low-bit post-training quantization of LLMs, improving zero-shot accuracy by 7% on average over baselines.
-
Rethinking 1-bit Optimization Leveraging Pre-trained Large Language Models
A progressive training scheme with binary-aware initialization and dual-scaling allows pre-trained LLMs to be converted to high-performance 1-bit models without training from scratch.
-
Analytical FFN-to-MoE Restructuring via Activation Pattern Analysis
An analytical post-training method restructures FFNs into MoE by partitioning neurons based on activation patterns and building a router from statistics, achieving 1.17x speedup with minimal resources.
-
H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models
H2O evicts non-heavy-hitter tokens from the KV cache using a dynamic submodular policy, retaining recent and frequent-co-occurrence tokens to reduce memory while preserving accuracy.
-
Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
GPT-4 as an LLM judge achieves over 80% agreement with human preferences on MT-Bench and Chatbot Arena, matching human agreement levels and providing a scalable evaluation method.
-
Mistral 7B
Mistral 7B is a 7B-parameter LLM that outperforms Llama 2 13B across benchmarks via grouped-query attention and sliding-window attention while remaining efficient.
-
NeuronMLP: Efficient LLM Inference via Singular Value Decomposition Compression and Tiling on AWS Trainium
NeuronMLP applies SVD-based compression and Trainium-specific tiling and caching to MLP layers, delivering 1.35x kernel speedup and 1.21x end-to-end inference speedup at 0.05 compression ratio versus AWS NKI baseline.