MoEITS is an information-theoretic algorithm for pruning experts in MoE-LLMs that produces models with higher accuracy and greater size reduction than prior state-of-the-art methods on Mixtral 8x7B, Qwen1.5-2.7B, and DeepSeek-V2-Lite.
Think you have solved question answering? try arc, the ai2 reasoning challenge
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 1
dataset 1
citation-polarity summary
representative citing papers
WizardLM uses LLM-driven iterative rewriting to generate complex instruction data and fine-tunes LLaMA to reach over 90% of ChatGPT capacity on 17 of 29 evaluated skills.
LLaMA-Adapter turns frozen LLaMA 7B into a capable instruction follower using only 1.2M new parameters and zero-init attention, matching Alpaca while extending to image-conditioned reasoning on ScienceQA and COCO.
STS repurposes draft-model attention scores from speculative decoding to build token-and-head-wise sparsity masks, delivering 2.67x speedup at ~90% sparsity on NarrativeQA with negligible accuracy loss.
FineWeb is a curated 15T-token web dataset that produces stronger LLMs than prior open collections, while its educational subset sharply improves performance on MMLU and ARC benchmarks.
DPOP is a new loss function that prevents DPO from lowering preferred response likelihoods and outperforms standard DPO on diverse datasets, MT-Bench, and enables Smaug-72B to exceed 80% on the Open LLM Leaderboard.
Quant.npu provides a fully static quantization pipeline for on-device LLMs on NPUs by combining rotation matrices, bit-width-aware initialization, two-stage selective optimization, and adaptive mixed precision.
STELLA aligns ESM3 bimodal sequence-structure encodings with Llama-3.1-8B text modeling to claim state-of-the-art results on protein functional description prediction and enzyme-catalyzed reaction prediction.
LoRA-FA freezes LoRA's A matrix and trains only B with gradient corrections to approximate full fine-tuning gradients more closely.
citing papers explorer
-
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention
LLaMA-Adapter turns frozen LLaMA 7B into a capable instruction follower using only 1.2M new parameters and zero-init attention, matching Alpaca while extending to image-conditioned reasoning on ScienceQA and COCO.