AI Research

AI & ML Research Papers

A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.

All Papers

17 papers

Preprint2406

Sparser is faster and less is more: Efficient sparse attention for long-range transformers

Unknown

This paper introduces a novel sparse attention mechanism that improves efficiency and performance for long-range transformers.

0Jan 1, 2406TransformersAttention Mechanisms

arXiv

Preprint2021

Efficient content-based sparse attention with routing transformers

Unknown

This paper introduces a content-based sparse attention mechanism that combines the benefits of content-based and local temporal sparse attention for improved efficiency in Transformer models.

0Jan 1, 2021TransformersAttention Mechanisms

Preprint2024

Unlocking efficiency in large language model inference: A comprehensive survey of speculative decoding

Unknown

A comprehensive survey of speculative decoding techniques for efficient large language model inference.

0Jan 1, 2024Large Language Models

Preprint2402

Lora+: Efficient low rank adaptation of large models

Unknown

This paper identifies and addresses suboptimal fine-tuning in LoRA for wide models by proposing Lora+, a method that improves parameter efficiency and performance.

0Jan 1, 2402

arXiv

Preprint2026

Moving Alphabet: A Controlled Study of Training Data for Text-to-Video Generation

Amber Yijia Zheng, Lu Liu, Raymond A. Yeh, et al.

A controlled study using a procedural testbed reveals that data distribution balance and caption quality critically impact text-to-video model generalization and training efficiency.

0Jul 21, 2026Attention MechanismsFine Tuning

arXiv

Preprint2406

Longrag: Enhancing retrieval-augmented generation with long-context llms

Unknown

This paper proposes LongRAG, a framework that leverages long-context LLMs to improve the efficiency and effectiveness of retrieval-augmented generation.

0Jan 1, 2406Retrieval Augmented Generation

arXiv

Preprint2024

Lightrag: Simple and fast retrieval-augmented generation

Unknown

Lightrag introduces a graph-based indexing approach to enhance efficiency in retrieval-augmented generation for LLMs.

0Jan 1, 2024Retrieval Augmented Generation

Preprint2024

Unveiling encoder-free vision-language models

Unknown

This paper proposes encoder-free vision-language models that directly process visual features without a separate vision encoder, simplifying architecture and improving efficiency.

0Jan 1, 2024Large Language ModelsComputer Vision

Preprint2603

Task-specific efficiency analysis: When small language models outperform large language models

Unknown

This paper establishes a quantitative foundation showing that small language models can outperform large ones on task-specific efficiency metrics.

0Jan 1, 2603Large Language Models

arXiv

Preprint2026

Shorterbetter: Guiding reasoning models to find optimal inference length for efficient reasoning

Unknown

This paper introduces ShorterBetter, a method to train reasoning models to dynamically determine optimal inference length, improving efficiency without sacrificing accuracy.

0Jan 1, 2026Reasoning

Preprint2025

Adaptthink: Reasoning models can learn when to think

Unknown

AdaptThink uses reinforcement learning to teach reasoning models when to engage in extended thinking, improving efficiency and accuracy.

0Jan 1, 2025Reasoning

PreprintarXiv.org2024

Personal llm agents: Insights and survey about the capability, efficiency and security

Yuanchun Li, Hao Wen, Weijun Wang, et al.

This paper surveys and analyzes Personal LLM Agents, focusing on their capability, efficiency, and security to guide future development.

356Jan 1, 2024Computer Vision

arXiv