Tootfinder

MAPEX: A Multi-Agent Pipeline for Keyphrase Extraction
Liting Zhang, Shiwan Zhao, Aobo Kong, Qicheng Li
https://arxiv.org/abs/2509.18813 https://arxiv.org/…

MAPEX: A Multi-Agent Pipeline for Keyphrase Extraction
Keyphrase extraction is a fundamental task in natural language processing. However, existing unsupervised prompt-based methods for Large Language Models (LLMs) often rely on single-stage inference pipelines with uniform prompting, regardless of document length or LLM backbone. Such one-size-fits-all designs hinder the full exploitation of LLMs' reasoning and generation capabilities, especially given the complexity of keyphrase extraction across diverse scenarios. To address these challenges, we…

Living Off the LLM: How LLMs Will Change Adversary Tactics
Sean Oesch, Jack Hutchins, Luke Koch, Kevin Kurian
https://arxiv.org/abs/2510.11398 https://arxi…

Living Off the LLM: How LLMs Will Change Adversary Tactics
In living off the land attacks, malicious actors use legitimate tools and processes already present on a system to avoid detection. In this paper, we explore how the on-device LLMs of the future will become a security concern as threat actors integrate LLMs into their living off the land attack pipeline and ways the security community may mitigate this threat.

Reasoning by Exploration: A Unified Approach to Retrieval and Generation over Graphs
Haoyu Han, Kai Guo, Harry Shomer, Yu Wang, Yucheng Chu, Hang Li, Li Ma, Jiliang Tang
https://arxiv.org/abs/2510.07484

Reasoning by Exploration: A Unified Approach to Retrieval and Generation over Graphs
Reasoning over structured graphs remains a fundamental challenge for Large Language Models (LLMs), particularly when scaling to large graphs. Existing approaches typically follow the retrieval-augmented generation (RAG) paradigm: first retrieving subgraphs relevant to the query and then generating answers conditioned on the retrieved subgraphs. However, such two-phase pipelines often struggle to faithfully incorporate graph structure, since the generation process is ultimately constrained by th…

Omni-Captioner: Data Pipeline, Models, and Benchmark for Omni Detailed Perception
Ziyang Ma, Ruiyang Xu, Zhenghao Xing, Yunfei Chu, Yuxuan Wang, Jinzheng He, Jin Xu, Pheng-Ann Heng, Kai Yu, Junyang Lin, Eng Siong Chng, Xie Chen
https://arxiv.org/abs/2510.12720

Omni-Captioner: Data Pipeline, Models, and Benchmark for Omni Detailed Perception
Fine-grained perception of multimodal information is critical for advancing human-AI interaction. With recent progress in audio-visual technologies, Omni Language Models (OLMs), capable of processing audio and video signals in parallel, have emerged as a promising paradigm for achieving richer understanding and reasoning. However, their capacity to capture and describe fine-grained details remains limited explored. In this work, we present a systematic and comprehensive investigation of omni de…

eXplainable Artificial Intelligence for RL-based Networking Solutions
Yeison Stiven Murcia, Oscar Mauricio Caicedo, Daniela Maria Casas, Nelson Luis Saldanha da Fonseca
https://arxiv.org/abs/2509.21649

eXplainable Artificial Intelligence for RL-based Networking Solutions
Reinforcement Learning (RL) agents have been widely used to improve networking tasks. However, understanding the decisions made by these agents is essential for their broader adoption in networking and network management. To address this, we introduce eXplaNet - a pipeline grounded in explainable artificial intelligence - designed to help networking researchers and practitioners gain deeper insights into the decision-making processes of RL-based solutions. We demonstrate how eXplaNet can be app…

Nova: Real-Time Agentic Vision-Language Model Serving with Adaptive Cross-Stage Parallelization
Yuhang Xu, Shengzhong Liu, Dong Zhang, Bingheng Yan, Fan Wu, Guihai Chen
https://arxiv.org/abs/2509.21301

Nova: Real-Time Agentic Vision-Language Model Serving with Adaptive Cross-Stage Parallelization
This paper presents Nova, a real-time scheduling framework for serving agentic vision-language models (VLMs) on a single GPU with balanced per-request latency and overall request process throughput. Our design begins by enabling effective pipelining across vision encode, LLM prefill, and LLM decode stages of VLMs, by exploiting their heterogeneous resource demands during execution and incorporating elastic GPU spatial partitioning among stages to maximally utilize the compute and memory resourc…