{"ok":true,"snapshot":{"date":"2026-08-30","capturedAt":"2026-08-30T11:00:07.288Z","total_papers":30,"queries":["large language model","transformer neural network","reinforcement learning from human feedback","AI agents","diffusion model"],"raw_count":50,"papers":[{"paperId":"83b90f4a0ae4cc214eb3cc140ccfef9cd99fac05","title":"Efficient Memory Management for Large Language Model Serving with PagedAttention","abstract":"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. When managed inefficiently, this memory can be significantly wasted by fragmentation and redundant duplication, limiting the batch size. To address this problem, we propose PagedAttention, an attention algorithm inspired by the classical virtual memory and paging techniques in operating systems. On top of it, we build vLLM, an LLM serving system that achieves (1) near-zero waste in KV cache memory and (2) flexible sharing of KV cache within and across requests to further reduce memory usage. Our evaluations show that vLLM improves the throu…","authors":["Woosuk Kwon","Zhuohan Li","Siyuan Zhuang","Ying Sheng","Lianmin Zheng","C. Yu","Joseph E. Gonzalez","Haotong Zhang"],"year":2023,"venue":"Symposium on Operating Systems Principles","citationCount":7915,"url":"https://www.semanticscholar.org/paper/83b90f4a0ae4cc214eb3cc140ccfef9cd99fac05","publicationDate":"2023-09-12","arxivId":"2309.06180","doi":"10.1145/3600006.3613165","fieldsOfStudy":["Computer Science"]},{"paperId":"28c6ac721f54544162865f41c5692e70d61bccab","title":"A survey on large language model based autonomous agents","abstract":"Autonomous agents have long been a research focus in academic and industry communities. Previous research often focuses on training agents with limited knowledge within isolated environments, which diverges significantly from human learning processes, and makes the agents hard to achieve human-like decisions. Recently, through the acquisition of vast amounts of Web knowledge, large language models (LLMs) have shown potential in human-level intelligence, leading to a surge in research on LLM-based autonomous agents. In this paper, we present a comprehensive survey of these studies, delivering a systematic review of LLM-based autonomous agents from a holistic perspective. We first discuss the construction of LLM-based autonomous agents, proposing a unified framework that encompasses much of…","authors":["Lei Wang","Chengbang Ma","Xueyang Feng","Zeyu Zhang","Hao-ran Yang","Jingsen Zhang","Zhi-Yang Chen","Jiakai Tang"],"year":2023,"venue":"Frontiers of Computer Science","citationCount":3697,"url":"https://www.semanticscholar.org/paper/28c6ac721f54544162865f41c5692e70d61bccab","publicationDate":"2023-08-22","arxivId":"2308.11432","doi":"10.1007/s11704-024-40231-1","fieldsOfStudy":["Computer Science"]},{"paperId":"0c72450890a54b68d63baa99376131fda8f06cf9","title":"The Rise and Potential of Large Language Model Based Agents: A Survey","abstract":"For a long time, humanity has pursued artificial intelligence (AI) equivalent to or surpassing the human level, with AI agents considered a promising vehicle for this pursuit. AI agents are artificial entities that sense their environment, make decisions, and take actions. Many efforts have been made to develop intelligent agents, but they mainly focus on advancement in algorithms or training strategies to enhance specific capabilities or performance on particular tasks. Actually, what the community lacks is a general and powerful model to serve as a starting point for designing AI agents that can adapt to diverse scenarios. Due to the versatile capabilities they demonstrate, large language models (LLMs) are regarded as potential sparks for Artificial General Intelligence (AGI), offering …","authors":["Zhiheng Xi","Wenxiang Chen","Xin Guo","Wei He","Yiwen Ding","Boyang Hong","Ming Zhang","Junzhe Wang"],"year":2023,"venue":"arXiv.org","citationCount":2023,"url":"https://www.semanticscholar.org/paper/0c72450890a54b68d63baa99376131fda8f06cf9","publicationDate":"2023-09-14","arxivId":"2309.07864","doi":"10.48550/arXiv.2309.07864","fieldsOfStudy":["Computer Science"]},{"paperId":"1f2a20a6efaf83214861dddae4a38a83ae18fe32","title":"DeepSeek-Coder: When the Large Language Model Meets Programming - The Rise of Code Intelligence","abstract":"The rapid development of large language models has revolutionized code intelligence in software development. However, the predominance of closed-source models has restricted extensive research and development. To address this, we introduce the DeepSeek-Coder series, a range of open-source code models with sizes from 1.3B to 33B, trained from scratch on 2 trillion tokens. These models are pre-trained on a high-quality project-level code corpus and employ a fill-in-the-blank task with a 16K window to enhance code generation and infilling. Our extensive evaluations demonstrate that DeepSeek-Coder not only achieves state-of-the-art performance among open-source code models across multiple benchmarks but also surpasses existing closed-source models like Codex and GPT-3.5. Furthermore, DeepSeek…","authors":["Daya Guo","Qihao Zhu","Dejian Yang","Zhenda Xie","Kai Dong","Wentao Zhang","Guanting Chen","Xiao Bi"],"year":2024,"venue":"arXiv.org","citationCount":1862,"url":"https://www.semanticscholar.org/paper/1f2a20a6efaf83214861dddae4a38a83ae18fe32","publicationDate":"2024-01-25","arxivId":"2401.14196","doi":"10.48550/arXiv.2401.14196","fieldsOfStudy":["Computer Science"]},{"paperId":"7bf72a3b5fbac8bc0f461780810fbc781c28ef53","title":"CAMEL: Communicative Agents for \"Mind\" Exploration of Large Language Model Society","abstract":"The rapid advancement of chat-based language models has led to remarkable progress in complex task-solving. However, their success heavily relies on human input to guide the conversation, which can be challenging and time-consuming. This paper explores the potential of building scalable techniques to facilitate autonomous cooperation among communicative agents, and provides insight into their\"cognitive\"processes. To address the challenges of achieving autonomous cooperation, we propose a novel communicative agent framework named role-playing. Our approach involves using inception prompting to guide chat agents toward task completion while maintaining consistency with human intentions. We showcase how role-playing can be used to generate conversational data for studying the behaviors and c…","authors":["G. Li","Hasan Hammoud","Hani Itani","Dmitrii Khizbullin","Bernard Ghanem"],"year":2023,"venue":"Neural Information Processing Systems","citationCount":1800,"url":"https://www.semanticscholar.org/paper/7bf72a3b5fbac8bc0f461780810fbc781c28ef53","publicationDate":"2023-03-31","arxivId":"2303.17760","doi":"10.52202/075280-2264","fieldsOfStudy":["Computer Science"]},{"paperId":"38115e80d805fb0fb8f090dc88ced4b24be07878","title":"CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis","abstract":"Program synthesis strives to generate a computer program as a solution to a given problem specification, expressed with input-output examples or natural language descriptions. The prevalence of large language models advances the state-of-the-art for program synthesis, though limited training resources and data impede open access to such models. To democratize this, we train and release a family of large language models up to 16.1B parameters, called CODEGEN, on natural language and programming language data, and open source the training library JAXFORMER. We show the utility of the trained model by demonstrating that it is competitive with the previous state-of-the-art on zero-shot Python code generation on HumanEval. We further investigate the multi-step paradigm for program synthesis, w…","authors":["Erik Nijkamp","Bo Pang","Hiroaki Hayashi","Lifu Tu","Haiquan Wang","Yingbo Zhou","S. Savarese","Caiming Xiong"],"year":2022,"venue":"International Conference on Learning Representations","citationCount":1576,"url":"https://www.semanticscholar.org/paper/38115e80d805fb0fb8f090dc88ced4b24be07878","publicationDate":"2022-03-25","arxivId":"2203.13474","doi":null,"fieldsOfStudy":["Computer Science"]},{"paperId":"7d8905a1fd288068f12c8347caeabefd36d0dd6c","title":"Gorilla: Large Language Model Connected with Massive APIs","abstract":"Large Language Models (LLMs) have seen an impressive wave of advances recently, with models now excelling in a variety of tasks, such as mathematical reasoning and program synthesis. However, their potential to effectively use tools via API calls remains unfulfilled. This is a challenging task even for today's state-of-the-art LLMs such as GPT-4, largely due to their inability to generate accurate input arguments and their tendency to hallucinate the wrong usage of an API call. We release Gorilla, a finetuned LLaMA-based model that surpasses the performance of GPT-4 on writing API calls. When combined with a document retriever, Gorilla demonstrates a strong capability to adapt to test-time document changes, enabling flexible user updates or version changes. It also substantially mitigates…","authors":["Shishir G. Patil","Tianjun Zhang","Xin Wang","Joseph E. Gonzalez"],"year":2023,"venue":"Neural Information Processing Systems","citationCount":1563,"url":"https://www.semanticscholar.org/paper/7d8905a1fd288068f12c8347caeabefd36d0dd6c","publicationDate":"2023-05-24","arxivId":"2305.15334","doi":"10.52202/079017-4020","fieldsOfStudy":["Computer Science"]},{"paperId":"83edcfbb206ddad38a971d605da09390604248ea","title":"BloombergGPT: A Large Language Model for Finance","abstract":"The use of NLP in the realm of financial technology is broad and complex, with applications ranging from sentiment analysis and named entity recognition to question answering. Large Language Models (LLMs) have been shown to be effective on a variety of tasks; however, no LLM specialized for the financial domain has been reported in literature. In this work, we present BloombergGPT, a 50 billion parameter language model that is trained on a wide range of financial data. We construct a 363 billion token dataset based on Bloomberg's extensive data sources, perhaps the largest domain-specific dataset yet, augmented with 345 billion tokens from general purpose datasets. We validate BloombergGPT on standard LLM benchmarks, open financial benchmarks, and a suite of internal benchmarks that most …","authors":["Shijie Wu","Ozan Irsoy","Steven Lu","Vadim Dabravolski","Mark Dredze","Sebastian Gehrmann","P. Kambadur","David Stuart Rosenberg"],"year":2023,"venue":"arXiv.org","citationCount":1472,"url":"https://www.semanticscholar.org/paper/83edcfbb206ddad38a971d605da09390604248ea","publicationDate":"2023-03-30","arxivId":"2303.17564","doi":null,"fieldsOfStudy":["Computer Science","Economics"]},{"paperId":"383c598625110e0a4c60da4db10a838ef822fbcf","title":"A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Ugly","abstract":"Large Language Models (LLMs), such as ChatGPT and Bard, have revolutionized natural language understanding and generation. They possess deep language comprehension, human-like text generation capabilities, contextual awareness, and robust problem-solving skills, making them invaluable in various domains (e.g., search engines, customer support, translation). In the meantime, LLMs have also gained traction in the security community, revealing security vulnerabilities and showcasing their potential in security-related tasks. This paper explores the intersection of LLMs with security and privacy. Specifically, we investigate how LLMs positively impact security and privacy, potential risks and threats associated with their use, and inherent vulnerabilities within LLMs. Through a comprehensive …","authors":["Yifan Yao","Jinhao Duan","Kaidi Xu","Yuanfang Cai","Eric Sun","Yue Zhang"],"year":2023,"venue":"High Confid. Comput.","citationCount":1323,"url":"https://www.semanticscholar.org/paper/383c598625110e0a4c60da4db10a838ef822fbcf","publicationDate":"2023-12-04","arxivId":"2312.02003","doi":"10.1016/j.hcc.2024.100211","fieldsOfStudy":["Computer Science"]},{"paperId":"8f070e301979732e0dd73f6aa6170309cf73aa7d","title":"Large Language Model based Multi-Agents: A Survey of Progress and Challenges","abstract":"Large Language Models (LLMs) have achieved remarkable success across a wide array of tasks. Due to their notable capabilities in planning and reasoning, LLMs have been utilized as autonomous agents for the automatic execution of various tasks. Recently, LLM-based agent systems have rapidly evolved from single-agent planning or decision-making to operating as multi-agent systems, enhancing their ability in complex problem-solving and world simulation. To offer an overview of this dynamic field, we present this survey to offer an in-depth discussion on the essential aspects and challenges of LLM-based multi-agent (LLM-MA) systems. Our objective is to provide readers with an in-depth understanding of these key points: the domains and settings where LLM-MA systems operate or simulate; the pro…","authors":["Taicheng Guo","Xiuying Chen","Yaqi Wang","Ruidi Chang","Shichao Pei","N. Chawla","Olaf Wiest","Xiangliang Zhang"],"year":2024,"venue":"International Joint Conference on Artificial Intelligence","citationCount":1136,"url":"https://www.semanticscholar.org/paper/8f070e301979732e0dd73f6aa6170309cf73aa7d","publicationDate":"2024-01-21","arxivId":"2402.01680","doi":"10.48550/arXiv.2402.01680","fieldsOfStudy":["Computer Science"]},{"paperId":"7d645a3fd276918374fd9483fd675c28e46506d1","title":"Galactica: A Large Language Model for Science","abstract":"Information overload is a major obstacle to scientific progress. The explosive growth in scientific literature and data has made it ever harder to discover useful insights in a large mass of information. Today scientific knowledge is accessed through search engines, but they are unable to organize scientific knowledge alone. In this paper we introduce Galactica: a large language model that can store, combine and reason about scientific knowledge. We train on a large scientific corpus of papers, reference material, knowledge bases and many other sources. We outperform existing models on a range of scientific tasks. On technical knowledge probes such as LaTeX equations, Galactica outperforms the latest GPT-3 by 68.2% versus 49.0%. Galactica also performs well on reasoning, outperforming Chi…","authors":["Ross Taylor","Marcin Kardas","Guillem Cucurull","Thomas Scialom","A. Hartshorn","Elvis Saravia","Andrew Poulton","Viktor Kerkez"],"year":2022,"venue":"arXiv.org","citationCount":1084,"url":"https://www.semanticscholar.org/paper/7d645a3fd276918374fd9483fd675c28e46506d1","publicationDate":"2022-11-16","arxivId":"2211.09085","doi":null,"fieldsOfStudy":["Computer Science","Mathematics"]},{"paperId":"a1f8082505c7e90b0a033e1b9da0a97d67aad66c","title":"Accelerating Large Language Model Decoding with Speculative Sampling","abstract":"We present speculative sampling, an algorithm for accelerating transformer decoding by enabling the generation of multiple tokens from each transformer call. Our algorithm relies on the observation that the latency of parallel scoring of short continuations, generated by a faster but less powerful draft model, is comparable to that of sampling a single token from the larger target model. This is combined with a novel modified rejection sampling scheme which preserves the distribution of the target model within hardware numerics. We benchmark speculative sampling with Chinchilla, a 70 billion parameter language model, achieving a 2-2.5x decoding speedup in a distributed setup, without compromising the sample quality or making modifications to the model itself.","authors":["Charlie Chen","Sebastian Borgeaud","G. Irving","Jean-Baptiste Lespiau","L. Sifre","J. Jumper"],"year":2023,"venue":"arXiv.org","citationCount":1050,"url":"https://www.semanticscholar.org/paper/a1f8082505c7e90b0a033e1b9da0a97d67aad66c","publicationDate":"2023-02-02","arxivId":"2302.01318","doi":"10.48550/arXiv.2302.01318","fieldsOfStudy":["Computer Science"]},{"paperId":"ad113d8b6f8f6dace7abd12dc88af520aaaf3fc7","title":"LISA: Reasoning Segmentation via Large Language Model","abstract":"Although perception systems have made remarkable ad-vancements in recent years, they still rely on explicit human instruction or pre-defined categories to identify the target objects before executing visual recognition tasks. Such systems cannot actively reason and comprehend implicit user intention. In this work, we propose a new segmentation task - reasoning segmentation. The task is designed to output a segmentation mask given a complex and implicit query text. Furthermore, we establish a benchmark comprising over one thousand image-instruction-mask data samples, incorporating intricate reasoning and world knowledge for evaluation purposes. Finally, we present LISA: large Language Instructed Segmentation Assistant, which inherits the language generation capabilities of multimodal Large…","authors":["Xin Lai","Zhuotao Tian","Yukang Chen","Yanwei Li","Yuhui Yuan","Shu Liu","Jiaya Jia"],"year":2023,"venue":"Computer Vision and Pattern Recognition","citationCount":1017,"url":"https://www.semanticscholar.org/paper/ad113d8b6f8f6dace7abd12dc88af520aaaf3fc7","publicationDate":"2023-08-01","arxivId":"2308.00692","doi":"10.1109/CVPR52733.2024.00915","fieldsOfStudy":["Computer Science"]},{"paperId":"ae736662f64d56f3ab1894fbd9c45f8f37251843","title":"OpenAssistant Conversations - Democratizing Large Language Model Alignment","abstract":"Aligning large language models (LLMs) with human preferences has proven to drastically improve usability and has driven rapid adoption as demonstrated by ChatGPT. Alignment techniques such as supervised fine-tuning (SFT) and reinforcement learning from human feedback (RLHF) greatly reduce the required skill and domain knowledge to effectively harness the capabilities of LLMs, increasing their accessibility and utility across various domains. However, state-of-the-art alignment techniques like RLHF rely on high-quality human feedback data, which is expensive to create and often remains proprietary. In an effort to democratize research on large-scale alignment, we release OpenAssistant Conversations, a human-generated, human-annotated assistant-style conversation corpus consisting of 161,44…","authors":["Andreas Kopf","Yannic Kilcher","Dimitri von Rutte","Sotiris Anagnostidis","Zhi Rui Tam","K. Stevens","A. Barhoum","Nguyen Minh Duc"],"year":2023,"venue":"Neural Information Processing Systems","citationCount":919,"url":"https://www.semanticscholar.org/paper/ae736662f64d56f3ab1894fbd9c45f8f37251843","publicationDate":"2023-04-14","arxivId":"2304.07327","doi":"10.48550/arXiv.2304.07327","fieldsOfStudy":["Computer Science"]},{"paperId":"72f77a393079431e4207b3afe678ee80b420e6f8","title":"DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Serving","abstract":"DistServe improves the performance of large language models (LLMs) serving by disaggregating the prefill and decoding computation. Existing LLM serving systems colocate the two phases and batch the computation of prefill and decoding across all users and requests. We find that this strategy not only leads to strong prefill-decoding interferences but also couples the resource allocation and parallelism plans for both phases. LLM applications often emphasize individual latency for each phase: time to first token (TTFT) for the prefill phase and time per output token (TPOT) of each request for the decoding phase. In the presence of stringent latency requirements, existing systems have to prioritize one latency over the other, or over-provision compute resources to meet both. DistServe assign…","authors":["Yinmin Zhong","Shengyu Liu","Junda Chen","Jianbo Hu","Yibo Zhu","Xuanzhe Liu","Xin Jin","Hao Zhang"],"year":2024,"venue":"USENIX Symposium on Operating Systems Design and Implementation","citationCount":912,"url":"https://www.semanticscholar.org/paper/72f77a393079431e4207b3afe678ee80b420e6f8","publicationDate":"2024-01-18","arxivId":"2401.09670","doi":"10.48550/arXiv.2401.09670","fieldsOfStudy":["Computer Science"]},{"paperId":"3487c12512fa41d3a4d64f00cb842525a8590ad3","title":"TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation","abstract":"Large Language Models (LLMs) have demonstrated remarkable performance across diverse domains, thereby prompting researchers to explore their potential for use in recommendation systems. Initial attempts have leveraged the exceptional capabilities of LLMs, such as rich knowledge and strong generalization through In-context Learning, which involves phrasing the recommendation task as prompts. Nevertheless, the performance of LLMs in recommendation tasks remains suboptimal due to a substantial disparity between the training tasks for LLMs and recommendation tasks, as well as inadequate recommendation data during pre-training. To bridge the gap, we consider building a Large Recommendation Language Model by tunning LLMs with recommendation data. To this end, we propose an efficient and effecti…","authors":["Keqin Bao","Jizhi Zhang","Yang Zhang","Wenjie Wang","Fuli Feng","Xiangnan He"],"year":2023,"venue":"ACM Conference on Recommender Systems","citationCount":844,"url":"https://www.semanticscholar.org/paper/3487c12512fa41d3a4d64f00cb842525a8590ad3","publicationDate":"2023-04-30","arxivId":"2305.00447","doi":"10.1145/3604915.3608857","fieldsOfStudy":["Computer Science"]},{"paperId":"4a7f6c4e71e20311ade4e76e8d0945d499c31fcd","title":"ChatDoctor: A Medical Chat Model Fine-Tuned on a Large Language Model Meta-AI (LLaMA) Using Medical Domain Knowledge","abstract":"Objective The primary aim of this research was to address the limitations observed in the medical knowledge of prevalent large language models (LLMs) such as ChatGPT, by creating a specialized language model with enhanced accuracy in medical advice. Methods We achieved this by adapting and refining the large language model meta-AI (LLaMA) using a large dataset of 100,000 patient-doctor dialogues sourced from a widely used online medical consultation platform. These conversations were cleaned and anonymized to respect privacy concerns. In addition to the model refinement, we incorporated a self-directed information retrieval mechanism, allowing the model to access and utilize real-time information from online sources like Wikipedia and data from curated offline medical databases. Results T…","authors":["Yunxiang Li","Zihan Li","Kai Zhang","Ruilong Dan","Steven Jiang","You Zhang"],"year":2023,"venue":"Cureus","citationCount":760,"url":"https://www.semanticscholar.org/paper/4a7f6c4e71e20311ade4e76e8d0945d499c31fcd","publicationDate":"2023-03-24","arxivId":"2303.14070","doi":"10.7759/cureus.40895","fieldsOfStudy":["Medicine","Computer Science"]},{"paperId":"1ddbd08ad8cf22a5c66c4242194c4286328533bf","title":"MiniGPT-v2: large language model as a unified interface for vision-language multi-task learning","abstract":"Large language models have shown their remarkable capabilities as a general interface for various language-related applications. Motivated by this, we target to build a unified interface for completing many vision-language tasks including image description, visual question answering, and visual grounding, among others. The challenge is to use a single model for performing diverse vision-language tasks effectively with simple multi-modal instructions. Towards this objective, we introduce MiniGPT-v2, a model that can be treated as a unified interface for better handling various vision-language tasks. We propose using unique identifiers for different tasks when training the model. These identifiers enable our model to better distinguish each task instruction effortlessly and also improve the…","authors":["Jun Chen","Deyao Zhu","Xiaoqian Shen","Xiang Li","Zechun Liu","Pengchuan Zhang","Raghuraman Krishnamoorthi","Vikas Chandra"],"year":2023,"venue":"arXiv.org","citationCount":742,"url":"https://www.semanticscholar.org/paper/1ddbd08ad8cf22a5c66c4242194c4286328533bf","publicationDate":"2023-10-14","arxivId":"2310.09478","doi":"10.48550/arXiv.2310.09478","fieldsOfStudy":["Computer Science"]},{"paperId":"ad13b213681b6f634bc83a264df246e83dd9a9d9","title":"mPLUG-OwI2: Revolutionizing Multi-modal Large Language Model with Modality Collaboration","abstract":"Multi-modal Large Language Models (MLLMs) have demonstrated impressive instruction abilities across various open-ended tasks. However, previous methods primarily fo-cus on enhancing multi-modal capabilities. In this work, we introduce a versatile multi-modal large language model, mPLUG-Owl2, which effectively leverages modality collab-oration to improve performance in both text and multi-modal tasks. mPLUG-Owl2 utilizes a modularized network design, with the language decoder acting as a universal interface for managing different modalities. Specifically, mPLUG-Owl2 incorporates shared functional modules to facilitate modal-ity collaboration and introduces a modality-adaptive module that preserves modality-specific features. Extensive experi-ments reveal that mPLUG-Owl2 is capable of gener…","authors":["Qinghao Ye","Haiyang Xu","Jiabo Ye","Mingshi Yan","Anwen Hu","Haowei Liu","Qi Qian","Ji Zhang"],"year":2023,"venue":"Computer Vision and Pattern Recognition","citationCount":720,"url":"https://www.semanticscholar.org/paper/ad13b213681b6f634bc83a264df246e83dd9a9d9","publicationDate":"2023-11-07","arxivId":"2311.04257","doi":"10.1109/CVPR52733.2024.01239","fieldsOfStudy":["Computer Science"]},{"paperId":"e0f27336698c84709bd60b6b7f4ce588cbae66bf","title":"StructGPT: A General Framework for Large Language Model to Reason over Structured Data","abstract":"In this paper, we study how to improve the zero-shot reasoning ability of large language models~(LLMs) over structured data in a unified way. Inspired by the study on tool augmentation for LLMs, we develop an \\emph{Iterative Reading-then-Reasoning~(IRR)} approach for solving question answering tasks based on structured data, called \\textbf{StructGPT}. In our approach, we construct the specialized function to collect relevant evidence from structured data (\\ie \\emph{reading}), and let LLMs concentrate the reasoning task based on the collected information (\\ie \\emph{reasoning}). Specially, we propose an \\emph{invoking-linearization-generation} procedure to support LLMs in reasoning on the structured data with the help of the external interfaces. By iterating this procedures with provided in…","authors":["Jinhao Jiang","Kun Zhou","Zican Dong","Keming Ye","Wayne Xin Zhao","Ji-rong Wen"],"year":2023,"venue":"Conference on Empirical Methods in Natural Language Processing","citationCount":610,"url":"https://www.semanticscholar.org/paper/e0f27336698c84709bd60b6b7f4ce588cbae66bf","publicationDate":"2023-05-16","arxivId":"2305.09645","doi":"10.48550/arXiv.2305.09645","fieldsOfStudy":["Computer Science"]},{"paperId":"1d9c21a0fdb1cc16a32c5d490ebaf98436a23382","title":"Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory","abstract":"Large Language Models (LLMs) have demonstrated remarkable prowess in generating contextually coherent responses, yet their fixed context windows pose fundamental challenges for maintaining consistency over prolonged multi-session dialogues. We introduce Mem0, a scalable memory-centric architecture that addresses this issue by dynamically extracting, consolidating, and retrieving salient information from ongoing conversations. Building on this foundation, we further propose an enhanced variant that leverages graph-based memory representations to capture complex relational structures among conversational elements. Through comprehensive evaluations on LOCOMO benchmark, we systematically compare our approaches against six baseline categories: (i) established memory-augmented systems, (ii) ret…","authors":["P. Chhikara","Dev Khant","Saket Aryan","Taranjeet Singh","Deshraj Yadav"],"year":2025,"venue":"European Conference on Artificial Intelligence","citationCount":579,"url":"https://www.semanticscholar.org/paper/1d9c21a0fdb1cc16a32c5d490ebaf98436a23382","publicationDate":"2025-04-28","arxivId":"2504.19413","doi":"10.48550/arXiv.2504.19413","fieldsOfStudy":["Computer Science"]},{"paperId":"b47e96762351b2dbf7e863ece4640df6194bcc0c","title":"Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning","abstract":"Large language models (LLMs) have demonstrated impressive reasoning abilities in complex tasks. However, they lack up-to-date knowledge and experience hallucinations during reasoning, which can lead to incorrect reasoning processes and diminish their performance and trustworthiness. Knowledge graphs (KGs), which capture vast amounts of facts in a structured format, offer a reliable source of knowledge for reasoning. Nevertheless, existing KG-based LLM reasoning methods only treat KGs as factual knowledge bases and overlook the importance of their structural information for reasoning. In this paper, we propose a novel method called reasoning on graphs (RoG) that synergizes LLMs with KGs to enable faithful and interpretable reasoning. Specifically, we present a planning-retrieval-reasoning …","authors":["Linhao Luo","Yuan-Fang Li","Gholamreza Haffari","Shirui Pan"],"year":2023,"venue":"International Conference on Learning Representations","citationCount":554,"url":"https://www.semanticscholar.org/paper/b47e96762351b2dbf7e863ece4640df6194bcc0c","publicationDate":"2023-10-02","arxivId":"2310.01061","doi":"10.48550/arXiv.2310.01061","fieldsOfStudy":["Computer Science"]},{"paperId":"986e813f4c4f36786c3642cb9c8718586e47bdcf","title":"AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenges","abstract":"This review critically distinguishes between AI Agents and Agentic AI, offering a structured, conceptual taxonomy, application mapping, and analysis of opportunities and challenges to clarify their divergent design philosophies and capabilities. We begin by outlining the search strategy and foundational definitions, characterizing AI Agents as modular systems driven and enabled by LLMs and LIMs for taskspecific automation. Generative AI is positioned as a precursor providing the foundation, with AI agents advancing through tool integration, prompt engineering, and reasoning enhancements. We then characterize Agentic AI systems, which, in contrast to AI Agents, represent a paradigm shift marked by multi-agent collaboration, dynamic task decomposition, persistent memory, and coordinated aut…","authors":["Ranjan Sapkota","Konstantinos I. Roumeliotis","Manoj Karkee"],"year":2025,"venue":"Information Fusion","citationCount":518,"url":"https://www.semanticscholar.org/paper/986e813f4c4f36786c3642cb9c8718586e47bdcf","publicationDate":"2025-05-15","arxivId":"2505.10468","doi":"10.1016/j.inffus.2025.103599","fieldsOfStudy":["Computer Science"]},{"paperId":"2b3ad2fdd9d2013119232ee49e6d21eb08474b74","title":"Mixture-of-Agents Enhances Large Language Model Capabilities","abstract":"Recent advances in large language models (LLMs) demonstrate substantial capabilities in natural language understanding and generation tasks. With the growing number of LLMs, how to harness the collective expertise of multiple LLMs is an exciting open direction. Toward this goal, we propose a new approach that leverages the collective strengths of multiple LLMs through a Mixture-of-Agents (MoA) methodology. In our approach, we construct a layered MoA architecture wherein each layer comprises multiple LLM agents. Each agent takes all the outputs from agents in the previous layer as auxiliary information in generating its response. MoA models achieves state-of-art performance on AlpacaEval 2.0, MT-Bench and FLASK, surpassing GPT-4 Omni. For example, our MoA using only open-source LLMs is the…","authors":["Junlin Wang","Jue Wang","Ben Athiwaratkun","Ce Zhang","James Zou"],"year":2024,"venue":"International Conference on Learning Representations","citationCount":493,"url":"https://www.semanticscholar.org/paper/2b3ad2fdd9d2013119232ee49e6d21eb08474b74","publicationDate":"2024-06-07","arxivId":"2406.04692","doi":"10.48550/arXiv.2406.04692","fieldsOfStudy":["Computer Science"]},{"paperId":"567157a4722d79f5ac019142e5fb1ad10b4e86cb","title":"VideoAgent: Long-form Video Understanding with Large Language Model as Agent","abstract":"Long-form video understanding represents a significant challenge within computer vision, demanding a model capable of reasoning over long multi-modal sequences. Motivated by the human cognitive process for long-form video understanding, we emphasize interactive reasoning and planning over the ability to process lengthy visual inputs. We introduce a novel agent-based system, VideoAgent, that employs a large language model as a central agent to iteratively identify and compile crucial information to answer a question, with vision-language foundation models serving as tools to translate and retrieve visual information. Evaluated on the challenging EgoSchema and NExT-QA benchmarks, VideoAgent achieves 54.1% and 71.3% zero-shot accuracy with only 8.4 and 8.2 frames used on average. These resul…","authors":["Xiaohan Wang","Yuhui Zhang","Orr Zohar","S. Yeung-Levy"],"year":2024,"venue":"European Conference on Computer Vision","citationCount":391,"url":"https://www.semanticscholar.org/paper/567157a4722d79f5ac019142e5fb1ad10b4e86cb","publicationDate":"2024-03-15","arxivId":"2403.10517","doi":"10.48550/arXiv.2403.10517","fieldsOfStudy":["Computer Science"]},{"paperId":"0cee098244c9978032702862a43a09f468f691a4","title":"Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding","abstract":"To mitigate the high inference latency stemming from autoregressive decoding in Large Language Models (LLMs), Speculative Decoding has emerged as a novel decoding paradigm for LLM inference. In each decoding step, this method first drafts several future tokens efficiently and then verifies them in parallel. Unlike autoregressive decoding, Speculative Decoding facilitates the simultaneous decoding of multiple tokens per step, thereby accelerating inference. This paper presents a comprehensive overview and analysis of this promising decoding paradigm. We begin by providing a formal definition and formulation of Speculative Decoding. Then, we organize in-depth discussions on its key facets, such as drafter selection and verification strategies. Furthermore, we present a comparative analysis …","authors":["Heming Xia","Zhe Yang","Qingxiu Dong","Peiyi Wang","Yongqi Li","Tao Ge","Tianyu Liu","Wenjie Li"],"year":2024,"venue":"Annual Meeting of the Association for Computational Linguistics","citationCount":320,"url":"https://www.semanticscholar.org/paper/0cee098244c9978032702862a43a09f468f691a4","publicationDate":"2024-01-15","arxivId":"2401.07851","doi":"10.48550/arXiv.2401.07851","fieldsOfStudy":["Computer Science"]},{"paperId":"89bd8efe0b9c0427cb7814d7b8c2b0190d2ffa9e","title":"Scalable watermarking for identifying large language model outputs","abstract":"Large language models (LLMs) have enabled the generation of high-quality synthetic text, often indistinguishable from human-written content, at a scale that can markedly affect the nature of the information ecosystem1–3. Watermarking can help identify synthetic text and limit accidental or deliberate misuse4, but has not been adopted in production systems owing to stringent quality, detectability and computational efficiency requirements. Here we describe SynthID-Text, a production-ready text watermarking scheme that preserves text quality and enables high detection accuracy, with minimal latency overhead. SynthID-Text does not affect LLM training and modifies only the sampling procedure; watermark detection is computationally efficient, without using the underlying LLM. To enable waterma…","authors":["Sumanth Dathathri","Abigail See","Sumedh Ghaisas","Po-Sen Huang","R. McAdam","Johannes Welbl","Vandana Bachani","Alex Kaskasoli"],"year":2024,"venue":"Nature","citationCount":302,"url":"https://www.semanticscholar.org/paper/89bd8efe0b9c0427cb7814d7b8c2b0190d2ffa9e","publicationDate":"2024-10-01","arxivId":null,"doi":"10.1038/s41586-024-08025-4","fieldsOfStudy":["Medicine","Computer Science"]},{"paperId":"d24e37aafcf48c76aca30430670bad9a61cd0fca","title":"The Virtual Lab of AI agents designs new SARS-CoV-2 nanobodies","abstract":null,"authors":["Kyle Swanson","Wesley Wu","Nash L. Bulaong","John E. Pak","James Y. Zou"],"year":2025,"venue":"Nature","citationCount":264,"url":"https://www.semanticscholar.org/paper/d24e37aafcf48c76aca30430670bad9a61cd0fca","publicationDate":"2025-07-29","arxivId":null,"doi":"10.1038/s41586-025-09442-9","fieldsOfStudy":["Medicine","Computer Science"]},{"paperId":"d362b7619fcd2df4241696a19aec95961b8a729c","title":"Memory in the Age of AI Agents","abstract":"Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent memory often differ substantially in their motivations, implementations, and evaluation protocols, while the proliferation of loosely defined memory terminologies has further obscured conceptual clarity. Traditional taxonomies such as long/short-term memory have proven insufficient to capture the diversity of contemporary agent memory systems. This work aims to provide an up-to-date landscape of current agent memory research. We begin by clearly delineating the scope of agent memory and distinguishing it fr…","authors":["Yuyang Hu","Shichun Liu","Yanwei Yue","Guibin Zhang","Boyang Liu","Fangyi Zhu","Jiahang Lin","Honglin Guo"],"year":2025,"venue":"arXiv.org","citationCount":250,"url":"https://www.semanticscholar.org/paper/d362b7619fcd2df4241696a19aec95961b8a729c","publicationDate":"2025-12-15","arxivId":"2512.13564","doi":"10.48550/arXiv.2512.13564","fieldsOfStudy":["Computer Science"]},{"paperId":"6758a6db1bfb6ebc5134aea9ce0fc28dd2e031a4","title":"From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review","abstract":"Large language models and autonomous AI agents have evolved rapidly, resulting in a diverse array of evaluation benchmarks, frameworks, and collaboration protocols. Driven by the growing need for standardized evaluation and integration, we systematically consolidate these fragmented efforts into a unified framework. However, the landscape remains fragmented and lacks a unified taxonomy or comprehensive survey. Therefore, we present a side-by-side comparison of benchmarks developed between 2019 and 2025 that evaluate these models and agents across multiple domains. In addition, we propose a taxonomy of approximately 60 benchmarks that cover general and academic knowledge reasoning, mathematical problem-solving, code generation and software engineering, factual grounding and retrieval, doma…","authors":["M. Ferrag","Norbert Tihanyi","M. Debbah"],"year":2025,"venue":"IEEE Access","citationCount":211,"url":"https://www.semanticscholar.org/paper/6758a6db1bfb6ebc5134aea9ce0fc28dd2e031a4","publicationDate":"2025-04-28","arxivId":"2504.19678","doi":"10.1109/ACCESS.2026.3698694","fieldsOfStudy":["Computer Science"]}],"summary":{"by_year":{"2022":2,"2023":16,"2024":7,"2025":5},"top_venues":[{"venue":"arXiv.org","count":7},{"venue":"Neural Information Processing Systems","count":3},{"venue":"International Conference on Learning Representations","count":3},{"venue":"Computer Vision and Pattern Recognition","count":2},{"venue":"Nature","count":2},{"venue":"Symposium on Operating Systems Principles","count":1},{"venue":"Frontiers of Computer Science","count":1},{"venue":"High Confid. Comput.","count":1},{"venue":"International Joint Conference on Artificial Intelligence","count":1},{"venue":"USENIX Symposium on Operating Systems Design and Implementation","count":1}],"top_authors":[{"author":"Joseph E. Gonzalez","count":2},{"author":"Woosuk Kwon","count":1},{"author":"Zhuohan Li","count":1},{"author":"Siyuan Zhuang","count":1},{"author":"Ying Sheng","count":1},{"author":"Lianmin Zheng","count":1},{"author":"C. Yu","count":1},{"author":"Haotong Zhang","count":1},{"author":"Lei Wang","count":1},{"author":"Chengbang Ma","count":1}]}}}