Claim search

At the High tier, MARCO surpasses unbounded recall (0.925 vs 0.880) at 42% lower cost.

The multi-modal parser achieves 0.962 entity F1 across 50 seeds spanning five modalities.

The paper 'The Last Human-Written Paper: Agent-Native Research Artifacts' (arXiv:2604.24658) reports: Scientific publication compresses a branching, iterative research process into a linear narrative, discarding the majority of what was discovered along the way. This compilation imposes two structural costs: a Storytelling Tax, where failed experiments, rejected hypotheses, and the branching exploration process are discarded to fit a linear narrative; and an Engineering Tax, where the gap between reviewer-sufficient prose and agent-sufficient specification leaves critical implementation…

The paper 'aiXiv: A Next-Generation Open Access Ecosystem for Scientific Discovery Generated by AI Scientists' (arXiv:2508.15126) reports: Recent advances in large language models (LLMs) have enabled AI agents to autonomously generate scientific proposals, conduct experiments, author papers, and perform peer reviews. Yet this flood of AI-generated research content collides with a fragmented and largely closed publication ecosystem.

The paper 'The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search' (arXiv:2504.08066) reports: AI is increasingly playing a pivotal role in transforming how scientific discoveries are made. We introduce The AI Scientist-v2, an end-to-end agentic system capable of producing the first entirely AI generated peer-review-accepted workshop paper.

The paper 'Paper2Agent: Reimagining Research Papers As Interactive and Reliable AI Agents' (arXiv:2509.06917) reports: We introduce Paper2Agent, an automated framework that converts research papers into AI agents. Paper2Agent transforms research output from passive artifacts into active systems that can accelerate downstream use, adoption, and discovery.

The paper 'Kosmos: An AI Scientist for Autonomous Discovery' (arXiv:2511.02824) reports: Data-driven scientific discovery requires iterative cycles of literature search, hypothesis generation, and data analysis. Substantial progress has been made towards AI agents that can automate scientific research, but all such agents remain limited in the number of actions they can take before losing coherence, thus limiting the depth of their findings.

The paper 'OmniScientist: Toward a Co-evolving Ecosystem of Human and AI Scientists' (arXiv:2511.16931) reports: With the rapid development of Large Language Models (LLMs), AI agents have demonstrated increasing proficiency in scientific tasks, ranging from hypothesis generation and experimental design to manuscript writing. Such agent systems are commonly referred to as "AI Scientists." However, existing AI Scientists predominantly formulate scientific discovery as a standalone search or optimization problem, overlooking the fact that scientific research is inherently a social and…

The paper 'FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness' (arXiv:2205.14135) reports: Transformers are slow and memory-hungry on long sequences, since the time and memory complexity of self-attention are quadratic in sequence length. Approximate attention methods have attempted to address this problem by trading off model quality to reduce the compute complexity, but often do not achieve wall-clock speedup.

The paper 'FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning' (arXiv:2307.08691) reports: Scaling Transformers to longer sequence lengths has been a major problem in the last several years, promising to improve performance in language modeling and high-resolution image understanding, as well as to unlock new applications in code, audio, and video generation. The attention layer is the main bottleneck in scaling to longer sequences, as its runtime and memory increase quadratically in the sequence length.

The paper 'FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision' (arXiv:2407.08608) reports: Attention, as a core layer of the ubiquitous Transformer architecture, is the bottleneck for large language models and long-context applications. FlashAttention elaborated an approach to speed up attention on GPUs through minimizing memory reads/writes.

The paper 'Efficient Memory Management for Large Language Model Serving with PagedAttention' (arXiv:2309.06180) reports: High throughput serving of large language models (LLMs) requires batching sufficiently many requests at a time. However, existing systems struggle because the key-value cache (KV cache) memory for each request is huge and grows and shrinks dynamically.

The paper 'GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints' (arXiv:2305.13245) reports: Multi-query attention (MQA), which only uses a single key-value head, drastically speeds up decoder inference. However, MQA can lead to quality degradation, and moreover it may not be desirable to train a separate model just for faster inference.

The paper 'Fast Transformer Decoding: One Write-Head is All You Need' (arXiv:1911.02150) reports: Multi-head attention layers, as used in the Transformer neural sequence model, are a powerful alternative to RNNs for moving information across and between sequences. While training these layers is generally fast and simple, due to parallelizability across the length of the sequence, incremental inference (where such paralleization is impossible) is often slow, due to the memory-bandwidth cost of repeatedly loading the large "keys" and "values" tensors.

The paper 'AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration' (arXiv:2306.00978) reports: Large language models (LLMs) have transformed numerous AI applications. On-device LLM is becoming increasingly important: running LLMs locally on edge devices can reduce the cloud computing cost and protect users' privacy.

The paper 'GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers' (arXiv:2210.17323) reports: Generative Pre-trained Transformer models, known as GPT or OPT, set themselves apart through breakthrough performance across complex language modelling tasks, but also by their extremely high computational and storage costs. Specifically, due to their massive size, even inference for large, highly-accurate GPT models may require multiple performant GPUs, which limits the usability of such models.

The paper 'SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models' (arXiv:2211.10438) reports: Large language models (LLMs) show excellent performance but are compute- and memory-intensive. Quantization can reduce memory and accelerate inference.

The paper 'KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache' (arXiv:2402.02750) reports: Efficiently serving large language models (LLMs) requires batching of many requests to reduce the cost per request. Yet, with larger batch sizes and longer context lengths, the key-value (KV) cache, which stores attention keys and values to avoid re-computations, significantly increases memory demands and becomes the new bottleneck in speed and memory usage.

The paper 'H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models' (arXiv:2306.14048) reports: Large Language Models (LLMs), despite their recent impressive accomplishments, are notably cost-prohibitive to deploy, particularly for applications involving long-content generation, such as dialogue systems and story writing. Often, a large amount of transient state information, referred to as the KV cache, is stored in GPU memory in addition to model parameters, scaling linearly with the sequence length and batch size.

The paper 'Efficient Streaming Language Models with Attention Sinks' (arXiv:2309.17453) reports: Deploying Large Language Models (LLMs) in streaming applications such as multi-round dialogue, where long interactions are expected, is urgently needed but poses two major challenges. Firstly, during the decoding stage, caching previous tokens' Key and Value states (KV) consumes extensive memory.

The paper 'DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model' (arXiv:2405.04434) reports: We present DeepSeek-V2, a strong Mixture-of-Experts (MoE) language model characterized by economical training and efficient inference. It comprises 236B total parameters, of which 21B are activated for each token, and supports a context length of 128K tokens.

The paper 'Fast Inference from Transformers via Speculative Decoding' (arXiv:2211.17192) reports: Inference from large autoregressive models like Transformers is slow - decoding K tokens takes K serial runs of the model. In this work we introduce speculative decoding - an algorithm to sample from autoregressive models faster without any changes to the outputs, by computing several tokens in parallel.

The paper 'Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads' (arXiv:2401.10774) reports: Large Language Models (LLMs) employ auto-regressive decoding that requires sequential computation, with each step reliant on the previous one's output. This creates a bottleneck as each step necessitates moving the full model parameters from High-Bandwidth Memory (HBM) to the accelerator's cache.

The paper 'SGLang: Efficient Execution of Structured Language Model Programs' (arXiv:2312.07104) reports: Large language models (LLMs) are increasingly used for complex tasks that require multiple generation calls, advanced prompting techniques, control flow, and structured inputs/outputs. However, efficient systems are lacking for programming and executing these applications.

The paper 'Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving' (arXiv:2407.00079) reports: Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI. It features a KVCache-centric disaggregated architecture that separates the prefill and decoding clusters.

The paper 'KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization' (arXiv:2401.18079) reports: LLMs are seeing growing use for applications which require large context windows, and with these large context windows KV cache activations surface as the dominant contributor to memory consumption during inference. Quantization is a promising approach for compressing KV cache activations; however, existing solutions fail to represent activations accurately in sub-4-bit precision.

The paper 'Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs' (arXiv:2310.01801) reports: In this study, we introduce adaptive KV cache compression, a plug-and-play method that reduces the memory footprint of generative inference for Large Language Models (LLMs). Different from the conventional KV cache that retains key and value vectors for all context tokens, we conduct targeted profiling to discern the intrinsic structure of attention modules.

The paper 'Fast Distributed Inference Serving for Large Language Models' (arXiv:2305.05920) reports: Large language models (LLMs) power a new generation of interactive AI applications exemplified by ChatGPT. The interactive nature of these applications demands low latency for LLM inference.

The paper's own code (https://github.com/beir-cellar/beir) reproduces ndcg@10 = 0.65 for 'BEIR: BM25 (Anserini) nDCG@10 on SciFact'.

The paper's own code (https://github.com/hexiangnan/neural_collaborative_filtering) reproduces ml1m_hr_at_10 = 0.73 for 'Neural Collaborative Filtering (NeuMF) — MovieLens-1M HR@10'.