Knowledge catalogue
arXiv:2605.13734v1 Announce Type: cross Abstract: LLMs are widely adopted in production, pushing inference systems to their limits. Disaggregated LLM serving (e.g., PD separation and KV state disaggre
arXiv:2605.11026v1 Announce Type: cross Abstract: Defenses against indirect prompt injection (IPI) in tool-using LLM agents share two structural weaknesses. First, they all attempt to prevent attacks
arXiv:2601.21351v3 Announce Type: replace Abstract: Attentio-FFN disaggregation (AFD) is an emerging architecture for LLM decoding that separates state-heavy, KV-cache-dominated Attention computation
arXiv:2601.05752v3 Announce Type: replace Abstract: We introduce AutoMonitor-Bench, the first benchmark designed to systematically evaluate the reliability of LLM-based misbehavior monitors across div
arXiv:2605.11235v1 Announce Type: new Abstract: In LLM Reinforcement Fine-Tuning (RFT), curriculum learning drives both efficiency and performance. Yet, current methods externalize curriculum judgment
Databases have introduced new AI-powered SQL functions which take natural language instructions as input and are evaluated using LLMs. They leverage the power of LLMs to answer new kinds of queries: W
arXiv:2605.10531v1 Announce Type: new Abstract: This work investigates whether knowledge-driven large language model (LLM)-based storytelling can support purposeful narrative interaction with a digita
arXiv:2605.09678v1 Announce Type: new Abstract: While extremely powerful and versatile at various tasks, the thinking capabilities of large language models (LLMs) are often put under scrutiny as they
arXiv:2605.09760v1 Announce Type: new Abstract: A reliable resume-job matching system helps a company find suitable candidates from a pool of resumes and helps a job seeker find relevant jobs from a l
arXiv:2605.08442v1 Announce Type: cross Abstract: Persistent memory attacks against LLM agents achieve high attack success rates against open-source models. In these attacks, malicious instructions in
arXiv:2605.09777v1 Announce Type: cross Abstract: Gradient-based preference optimization methods for large language model (LLM) alignment suffer from preference collapse, converging to narrow behavior
arXiv:2603.04474v2 Announce Type: replace-cross Abstract: Large Language Model-based Multi-Agent Systems (LLM-MAS) are increasingly applied to complex collaborative scenarios. However, their collabora
arXiv:2605.08327v1 Announce Type: cross Abstract: In structured decision-making workflows such as form filling, compliance checking, and maintenance reporting, LLM outputs must be locally correct, glo
arXiv:2503.14434v3 Announce Type: replace-cross Abstract: Automated feature engineering plays a critical role in improving predictive model performance for tabular learning tasks. Traditional automate
arXiv:2605.08448v1 Announce Type: new Abstract: Semi-supervised learning approaches have been investigated as a means to enhance the analysis of social media data in disaster management contexts. In t
arXiv:2605.09781v1 Announce Type: cross Abstract: Large Language Models exhibit mode collapse, producing homogeneous outputs that fail to explore valid solution spaces. We present QD-LLM, a framework
arXiv:2604.01532v2 Announce Type: replace Abstract: LLM agents are beginning to invoke industrial asset-management tools through the Model Context Protocol (MCP), yet whether they can act reliably on
arXiv:2605.10614v1 Announce Type: new Abstract: Multi-agent LLM systems introduce a security risk in which sensitive information accessed by one agent can propagate through shared context and reappear
arXiv:2605.09808v1 Announce Type: new Abstract: User simulators are increasingly leveraged to build interactive AI assistants, yet how to measure the quality of these simulators remains an open questi
arXiv:2510.20036v2 Announce Type: replace Abstract: Large language model (LLM) agents rely on external tools to solve complex tasks, but real-world toolsets often contain redundant tools with overlapp
arXiv:2504.02343v2 Announce Type: replace Abstract: Recent advancements in Large Language Models (LLMs) and the proliferation of Text-Attributed Graphs (TAGs) across various domains have positioned LL
arXiv:2605.10405v1 Announce Type: new Abstract: Selecting the best large language model (LLM) for a fixed benchmark is often expensive, since exhaustive evaluation requires running every model on ever
arXiv:2605.09461v1 Announce Type: new Abstract: Automated vulnerability detection is a fundamental task in software security, yet existing learning-based methods still struggle to capture the structur
arXiv:2605.07926v1 Announce Type: new Abstract: As LLM-based agents increasingly rely on external tools, it is important to evaluate their ability to sustain tool-grounded reasoning beyond familiar wo
arXiv:2605.07063v1 Announce Type: cross Abstract: Data selection methods address a critical challenge in LLM post-training: effectively leveraging scarce, high-fidelity target data alongside abundant
arXiv:2605.06840v1 Announce Type: new Abstract: Large language models (LLMs), especially reasoning models, generate extended chain-of-thought (CoT) reasoning that often contains explicit deliberation
arXiv:2605.07442v1 Announce Type: new Abstract: LLM-based game generation promises to turn natural-language specifications into executable games, but progress is limited by the lack of reliable automa
arXiv:2605.06957v1 Announce Type: new Abstract: We present a dynamic policy-learning approach that combines generalized planning and hierarchical task decomposition for LLM-based agents. Our method, H
arXiv:2605.06940v1 Announce Type: new Abstract: Annotation automation via Large Language Models (LLMs) is the core approach for scaling NLP datasets; however, LLM behavior with respect to closed-set i
🧵 Slime: The Most Elegant & Comfortable RL Training Framework Ever A deep dive into why Slime redefines LLM RL training with clean architecture & production-grade engineering ✨ Insights from Zhihu con
arXiv:2605.07068v1 Announce Type: cross Abstract: The LLM Wiki pattern, to compile and provide domain knowledge into a persistent artifact and serve it to LLMs via KV cache inference, promises context
Signals V2 is an LLM-free analyzer tool that evaluates live agent execution trajectories by scoring them as OpenTelemetry spans, enabling developers to monitor and trace AI agent behavior without rely
Excited to speak at PyCon US @pycon about production LLM inference systems, runtime optimization, and next-gen engine design — including some of the new ideas behind the recently open-sourced project
arXiv:2511.02230v4 Announce Type: replace-cross Abstract: KV cache management is essential for efficient LLM inference. To maximize utilization, existing inference engines evict finished requests' KV
arXiv:2605.04903v1 Announce Type: cross Abstract: Large language models (LLMs) show strong potential for neural architecture generation, yet existing approaches produce complete model implementations
arXiv:2605.03101v1 Announce Type: new Abstract: Symbolic regression (SR), the task of discovering mathematical expressions that best describe a given dataset, remains a fundamental challenge in scient
arXiv:2605.02897v1 Announce Type: cross Abstract: LLM assistant personalities play a critical role in user experience and perceived response quality. We present a large-scale experiment of frontier LL
This post discusses upgrade recommendations for improving local LLM inference performance on a gaming laptop with an RTX 4080 (12GB VRAM) and 16GB system RAM. Community members likely shared options f
arXiv:2605.03310v1 Announce Type: cross Abstract: Multi-agent LLM systems fail in production at rates between 41% and 87%, mostly due to coordination defects rather than base-model capability. Existin
arXiv:2603.00977v2 Announce Type: replace-cross Abstract: Large language model (LLM) agents have recently demonstrated strong capabilities in interactive decision-making, yet they remain fundamentally
arXiv:2605.03228v1 Announce Type: cross Abstract: As large language model (LLM)-powered agents are increasingly deployed to perform complex, real-world tasks, they face a growing class of attacks that
arXiv:2605.02240v1 Announce Type: new Abstract: We introduce PhysicianBench, a benchmark for evaluating LLM agents on physician tasks grounded in real clinical setting within electronic health record
arXiv:2601.08517v2 Announce Type: replace Abstract: Channel-configuration search, the optimization of layer specifications such as channel widths in deep neural networks, presents a combinatorial chal
arXiv:2602.15823v2 Announce Type: replace Abstract: A central challenge in large language model (LLM) editing is capability preservation: methods that successfully change targeted behavior can quietly
Release: datasette-llm 0.1a7 Mechanism for configuring default options for specific models. Part of Datasette's evolving support mechanism for plugins that use LLMs. It's now possible to configure a m
arXiv:2605.02789v1 Announce Type: cross Abstract: Modern fuzzers increasingly use Large Language Models (LLMs) to generate structured inputs, but LLM-driven fuzzing is sensitive to prompt initializati
arXiv:2605.02801v1 Announce Type: new Abstract: As large language model (LLM) agents evolve from isolated tool users into coordinated teams, reinforcement learning (RL) must optimize not only individu
arXiv:2605.00519v2 Announce Type: cross Abstract: The operational landscape of local Large Language Model (LLM) inference has shifted from lightweight models to datacenter-class weights exceeding 70B
arXiv:2605.00197v1 Announce Type: cross Abstract: Studies attempting to simulate human behavior with extit{Silicon Societies} grow in numbers while LLM-only social networks have started appearing outs
arXiv:2605.00410v1 Announce Type: new Abstract: A multi-agent pipeline with N agents typically issues N LLM calls per run. Merging agents into fewer calls (compound execution) promises token savings,
arXiv:2508.06361v4 Announce Type: replace Abstract: Large Language Models (LLMs) are widely deployed in reasoning, planning, and decision-making tasks, making their trustworthiness critical. A signifi
arXiv:2507.14201v3 Announce Type: replace-cross Abstract: We present ExCyTIn-Bench, the first benchmark to Evaluate an LLM agent X on the task of Cyber Threat Investigation through security questions
arXiv:2605.00358v1 Announce Type: new Abstract: LLM parameter editing methods commonly rely on computing an ideal target hidden-state at a target layer (referred as anchor point) and distributing the
arXiv:2509.24496v3 Announce Type: replace Abstract: The explosive growth of large language models (LLMs) has created a vast but opaque landscape: millions of models exist, yet their evolutionary relat
arXiv:2604.27089v1 Announce Type: new Abstract: Large-language-models (LLMs) demonstrate enormous utility in long-context tasks which require processing prompts that consist of tens to hundreds of tho
arXiv:2604.27006v1 Announce Type: cross Abstract: Context: Study screening in systematic literature reviews is costly, inconsistency-prone, and risk-asymmetric, since false negatives can compromise va
arXiv:2604.27747v1 Announce Type: cross Abstract: Large language model (LLM)-based generative list-wise recommendation has advanced rapidly, but decoding remains sequential and thus latency-prone. To
arXiv:2604.27228v1 Announce Type: new Abstract: Democratic discourse analysis systems increasingly rely on multi-agent LLM pipelines in which distinct evaluator models are assigned adversarial roles t
arXiv:2601.21459v4 Announce Type: replace-cross Abstract: LLM role-playing, i.e., using LLMs to simulate specific personas, has emerged as a key capability in various applications, such as companionsh
arXiv:2602.20426v2 Announce Type: replace Abstract: While most efforts to improve LLM-based tool-using agents focus on the agent itself - through larger models, better prompting, or fine-tuning - agen