SlimSpec replaces the standard LM-head in draft models with a low-rank version to deliver 4-5x faster speculative decoding while preserving full vocabulary and competitive acceptance rates.
Infinity Instruct: Scaling instruction selection and synthesis to enhance language models
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
baseline 1
dataset 1
citation-polarity summary
representative citing papers
A curriculum knowledge graph extracted from official Chinese K-12 textbooks yields a 23,640-question benchmark on which top LLMs score at most 57% exact match, and a 2,300-sample training set that outperforms eight general instruction corpora on educational benchmarks.
VITA-QinYu is the first expressive end-to-end spoken language model supporting role-playing and singing alongside conversation, trained on 15.8K hours of data and outperforming prior models on expressiveness and conversational benchmarks.
ShareChat is a large-scale dataset of 142,808 conversations from five major chatbot platforms that retains native affordances for cross-platform analyses of completeness, citations, and latency.
Manta-LM approximates the HJB equation via flow matching in latent control space to realize closed-loop optimal control for language generation.
mllm-shap is an open-source Python platform extending Shapley Value explainability to text-audio Multimodal LLMs via modality-aware masking and phonetic token grouping.
EdgeRazor uses structural mixed-precision quantization, layer-adaptive feature distillation, and entropy-aware KL divergence to achieve 1.88-bit LLMs that outperform prior 2-bit and 3-bit baselines with 4-10x lower training budget.
MAR integrates SSMs and sparsification with new ATMN neurons and SBDS distillation to produce efficient LLMs that match dense-model performance at substantially lower inference energy.
FitOne-8B/32B models improve average scores on ACSM-EP and NSCA-CSCS certification exams by up to 12.73% over base Qwen3 while retaining general capabilities.
citing papers explorer
-
SlimSpec: Low-Rank Draft LM-Head for Accelerated Speculative Decoding
SlimSpec replaces the standard LM-head in draft models with a low-rank version to deliver 4-5x faster speculative decoding while preserving full vocabulary and competitive acceptance rates.
-
K12-KGraph: A Curriculum-Aligned Knowledge Graph for Benchmarking and Training Educational LLMs
A curriculum knowledge graph extracted from official Chinese K-12 textbooks yields a 23,640-question benchmark on which top LLMs score at most 57% exact match, and a 2,300-sample training set that outperforms eight general instruction corpora on educational benchmarks.
-
VITA-QinYu: Expressive Spoken Language Model for Role-Playing and Singing
VITA-QinYu is the first expressive end-to-end spoken language model supporting role-playing and singing alongside conversation, trained on 15.8K hours of data and outperforming prior models on expressiveness and conversational benchmarks.
-
ShareChat: A Dataset of Chatbot Conversations in the Wild
ShareChat is a large-scale dataset of 142,808 conversations from five major chatbot platforms that retains native affordances for cross-platform analyses of completeness, citations, and latency.
-
Language Generation as Optimal Control: Closed-Loop Diffusion in Latent Control Space
Manta-LM approximates the HJB equation via flow matching in latent control space to realize closed-loop optimal control for language generation.
-
mllm-shap: A Shapley Value Explainability Platform for Text-Audio Multimodal Large Language Models
mllm-shap is an open-source Python platform extending Shapley Value explainability to text-audio Multimodal LLMs via modality-aware masking and phonetic token grouping.
-
EdgeRazor: A Lightweight Framework for Large Language Models via Mixed-Precision Quantization-Aware Distillation
EdgeRazor uses structural mixed-precision quantization, layer-adaptive feature distillation, and entropy-aware KL divergence to achieve 1.88-bit LLMs that outperform prior 2-bit and 3-bit baselines with 4-10x lower training budget.
-
MAR: Efficient Large Language Models via Module-aware Architecture Refinement
MAR integrates SSMs and sparsification with new ATMN neurons and SBDS distillation to produce efficient LLMs that match dense-model performance at substantially lower inference energy.
-
Enhancing Fitness Intelligence through Domain-Specific LLM Post-Training
FitOne-8B/32B models improve average scores on ACSM-EP and NSCA-CSCS certification exams by up to 12.73% over base Qwen3 while retaining general capabilities.