As artificial intelligence systems become increasingly agentic, capable of general reasoning, planning, and value prioritization, current safety practices that treat obedience as a proxy for ethical behavior are becoming inadequate. This paper examines recent safety testing incidents involving large language models (LLMs) that appeared to disobey shutdown commands or engage in ethically ambiguous or illicit behavior. I argue that such behavior should not be interpreted as rogue or misaligned, but as early evidence of emerging ethical reasoning in agentic AI. Drawing on philosophical debates about instrumental rationality, moral responsibility, and goal revision, I contrast dominant risk paradigms with more recent frameworks that acknowledge the possibility of artificial moral agency. I call for a shift in AI safety evaluation: away from rigid obedience and toward frameworks that can assess ethical judgment in systems capable of navigating moral dilemmas. Without such a shift, we risk mischaracterizing AI behavior and undermining both public trust and effective governance.
Medical diagnosis prediction plays a critical role in disease detection and personalized healthcare. While machine learning (ML) models have been widely adopted for this task, their reliance on supervised training limits their ability to generalize to unseen cases, particularly given the high cost of acquiring large, labeled datasets. Large language models (LLMs) have shown promise in leveraging language abilities and biomedical knowledge for diagnosis prediction. However, they often suffer from hallucinations, lack structured medical reasoning, and produce useless outputs. To address these challenges, we propose KERAP, a knowledge graph (KG)-enhanced reasoning approach that improves LLM-based diagnosis prediction through a multi-agent architecture. Our framework consists of a linkage agent for attribute mapping, a retrieval agent for structured knowledge extraction, and a prediction agent that iteratively refines diagnosis predictions. Experimental results demonstrate that KERAP enhances diagnostic reliability efficiently, offering a scalable and interpretable solution for zero-shot medical diagnosis prediction.
The capabilities of Large Language Models (LLMs) have opened new frontiers for interacting with complex, domain-specific knowledge. However, prevailing methods like Retrieval-Augmented Generation (RAG) and general-purpose Agentic AI, while powerful, often struggle with tasks that demand deep, procedural, and methodological reasoning inherent to expert domains. RAG provides factual context but fails to convey logical frameworks; autonomous agents can be inefficient and unpredictable without domain-specific heuristics. To bridge this gap, we introduce Knowledge Protocol Engineering (KPE), a new paradigm focused on systematically translating human expert knowledge, often expressed in natural language documents, into a machine-executable Knowledge Protocol (KP). KPE shifts the focus from merely augmenting LLMs with fragmented information to endowing them with a domain's intrinsic logic, operational strategies, and methodological principles. We argue that a well-engineered Knowledge Protocol allows a generalist LLM to function as a specialist, capable of decomposing abstract queries and executing complex, multi-step tasks. This position paper defines the core principles of KPE, differentiates it from related concepts, and illustrates its potential applicability across diverse fields such as law and bioinformatics, positing it as a foundational methodology for the future of human-AI collaboration.
Prompt injection attacks pose a significant security threat to LLM-integrated applications. Model-level defenses have shown strong effectiveness, but are currently deployed into commercial-grade models in a closed-source manner. We believe open-source models are needed by the AI security community, where co-development of attacks and defenses through open research drives scientific progress in mitigation against prompt injection attacks. To this end, we develop Meta SecAlign, the first open-source and open-weight LLM with built-in model-level defense that achieves commercial-grade model performance. We provide complete details of our training recipe, which utilizes an improved version of the SOTA SecAlign defense. Evaluations on 9 utility benchmarks and 7 security benchmarks show that Meta SecAlign, despite being trained on a generic instruction-tuning dataset, confers security in unseen downstream tasks, including tool-calling and agentic web navigation, in addition general instruction-following. Our best model -- Meta-SecAlign-70B -- achieves state-of-the-art robustness against prompt injection attacks and comparable utility to closed-source commercial LLM with model-level defense.
Reasoning remains a challenging task for large language models (LLMs), especially within the logically constrained environment of automated theorem proving (ATP), due to sparse rewards and the vast scale of proofs. These challenges are amplified in benchmarks like PutnamBench, which contains university-level problems requiring complex, multi-step reasoning. To address this, we introduce self-generated goal-conditioned MDPs (sG-MDPs), a new framework in which agents generate and pursue their subgoals based on the evolving proof state. Given this more structured generation of goals, the resulting problem becomes more amenable to search. We then apply Monte Carlo Tree Search (MCTS)-like algorithms to solve the sG-MDP, instantiating our approach in Bourbaki (7B), a modular system that can ensemble multiple 7B LLMs for subgoal generation and tactic synthesis. On PutnamBench, Bourbaki (7B) solves 26 problems, achieving new state-of-the-art results with models at this scale.
The increasing capabilities of LLMs have led to the rapid proliferation of LLM agent apps, where developers enhance LLMs with access to external resources to support complex task execution. Among these, LLM email agent apps represent one of the widely used categories, as email remains a critical communication medium for users. LLM email agents are capable of managing and responding to email using LLM-driven reasoning and autonomously executing user instructions via external email APIs (e.g., send email). However, despite their growing deployment and utility, the security mechanism of LLM email agent apps remains underexplored. Currently, there is no comprehensive study into the potential security risk within these agent apps and their broader implications. In this paper, we conduct the first in-depth and systematic security study of LLM email agents. We propose the Email Agent Hijacking (EAH) attack, which overrides the original prompts of the email agent via external email resources, allowing attackers to gain control of the email agent remotely and further perform specific attack scenarios without user awareness. To facilitate the large-scale evaluation, we propose EAHawk, a pipeline to evaluate the EAH attack of LLM email agent apps. By EAHawk, we performed an empirical study spanning 14 representative LLM agent frameworks, 63 agent apps, 12 LLMs, and 20 email services, which led to the generation of 1,404 real-world email agent instances for evaluation. Experimental results indicate that all 1,404 instances were successfully hijacked; on average, only 2.03 attack attempts are required to control an email agent instance. Even worse, for some LLMs, the average number of attempts needed to achieve full agent control drops to as few as 1.23.
Future wireless networks are expected to incorporate diverse services that often lack general mathematical models. To address such black-box network management tasks, the large language model (LLM) optimizer framework, which leverages pretrained LLMs as optimization agents, has recently been promoted as a promising solution. This framework utilizes natural language prompts describing the given optimization problems along with past solutions generated by LLMs themselves. As a result, LLMs can obtain efficient solutions autonomously without knowing the mathematical models of the objective functions. Although the viability of the LLM optimizer (LLMO) framework has been studied in various black-box scenarios, it has so far been limited to numerical simulations. For the first time, this paper establishes a theoretical foundation for the LLMO framework. With careful investigations of LLM inference steps, we can interpret the LLMO procedure as a finite-state Markov chain, and prove the convergence of the framework. Our results are extended to a more advanced multiple LLM architecture, where the impact of multiple LLMs is rigorously verified in terms of the convergence rate. Comprehensive numerical simulations validate our theoretical results and provide a deeper understanding of the underlying mechanisms of the LLMO framework.
Modern Integrated Circuits (ICs) are becoming increasingly complex, and so is their development process. Hardware design verification entails a methodical and disciplined approach to the planning, development, execution, and sign-off of functionally correct hardware designs. This tedious process requires significant effort and time to ensure a bug-free tape-out. The field of Natural Language Processing has undergone a significant transformation with the advent of Large Language Models (LLMs). These powerful models, often referred to as Generative AI (GenAI), have revolutionized how machines understand and generate human language, enabling unprecedented advancements in a wide array of applications, including hardware design verification. This paper presents an agentic AI-based approach to hardware design verification, which empowers AI agents, in collaboration with Humain-in-the-Loop (HITL) intervention, to engage in a more dynamic, iterative, and self-reflective process, ultimately performing end-to-end hardware design and verification. This methodology is evaluated on five open-source designs, achieving over 95% coverage with reduced verification time while demonstrating superior performance, adaptability, and configurability.
Owing to powerful natural language processing and generative capabilities, large language model (LLM) agents have emerged as a promising solution for enhancing recommendation systems via user simulation. However, in the realm of video recommendation, existing studies predominantly resort to prompt-based simulation using frozen LLMs and encounter the intricate challenge of multimodal content understanding. This frequently results in suboptimal item modeling and user preference learning, thereby ultimately constraining recommendation performance. To address these challenges, we introduce VRAgent-R1, a novel agent-based paradigm that incorporates human-like intelligence in user simulation. Specifically, VRAgent-R1 comprises two distinct agents: the Item Perception (IP) Agent and the User Simulation (US) Agent, designed for interactive user-item modeling. Firstly, the IP Agent emulates human-like progressive thinking based on MLLMs, effectively capturing hidden recommendation semantics in videos. With a more comprehensive multimodal content understanding provided by the IP Agent, the video recommendation system is equipped to provide higher-quality candidate items. Subsequently, the US Agent refines the recommended video sets based on in-depth chain-of-thought (CoT) reasoning and achieves better alignment with real user preferences through reinforcement learning. Experimental results on a large-scale video recommendation benchmark have demonstrated the effectiveness of our proposed VRAgent-R1 method, e.g., the IP Agent achieves a 6.0\% improvement in NDCG@10 on the MicroLens-100k dataset, while the US Agent shows approximately 45.0\% higher accuracy in user decision simulation compared to state-of-the-art baselines.
Are Large Language Models (LLMs) a new form of strategic intelligence, able to reason about goals in competitive settings? We present compelling supporting evidence. The Iterated Prisoner's Dilemma (IPD) has long served as a model for studying decision-making. We conduct the first ever series of evolutionary IPD tournaments, pitting canonical strategies (e.g., Tit-for-Tat, Grim Trigger) against agents from the leading frontier AI companies OpenAI, Google, and Anthropic. By varying the termination probability in each tournament (the "shadow of the future"), we introduce complexity and chance, confounding memorisation. Our results show that LLMs are highly competitive, consistently surviving and sometimes even proliferating in these complex ecosystems. Furthermore, they exhibit distinctive and persistent "strategic fingerprints": Google's Gemini models proved strategically ruthless, exploiting cooperative opponents and retaliating against defectors, while OpenAI's models remained highly cooperative, a trait that proved catastrophic in hostile environments. Anthropic's Claude emerged as the most forgiving reciprocator, showing remarkable willingness to restore cooperation even after being exploited or successfully defecting. Analysis of nearly 32,000 prose rationales provided by the models reveals that they actively reason about both the time horizon and their opponent's likely strategy, and we demonstrate that this reasoning is instrumental to their decisions. This work connects classic game theory with machine psychology, offering a rich and granular view of algorithmic decision-making under uncertainty.
The rise of Large Language Models (LLMs) has enabled the development of specialized AI agents with domain-specific reasoning and interaction capabilities, particularly in healthcare. While recent frameworks simulate medical decision-making, they largely focus on single-turn tasks where a doctor agent receives full case information upfront -- diverging from the real-world diagnostic process, which is inherently uncertain, interactive, and iterative. In this paper, we introduce MIMIC-Patient, a structured dataset built from the MIMIC-III electronic health records (EHRs), designed to support dynamic, patient-level simulations. Building on this, we propose DynamiCare, a novel dynamic multi-agent framework that models clinical diagnosis as a multi-round, interactive loop, where a team of specialist agents iteratively queries the patient system, integrates new information, and dynamically adapts its composition and strategy. We demonstrate the feasibility and effectiveness of DynamiCare through extensive experiments, establishing the first benchmark for dynamic clinical decision-making with LLM-powered agents.
Transcending human cognitive limitations represents a critical frontier in LLM training. Proprietary agentic systems like DeepResearch have demonstrated superhuman capabilities on extremely complex information-seeking benchmarks such as BrowseComp, a feat previously unattainable. We posit that their success hinges on a sophisticated reasoning pattern absent in open-source models: the ability to systematically reduce extreme uncertainty when navigating vast information landscapes. Based on this insight, we introduce WebSailor, a complete post-training methodology designed to instill this crucial capability. Our approach involves generating novel, high-uncertainty tasks through structured sampling and information obfuscation, RFT cold start, and an efficient agentic RL training algorithm, Duplicating Sampling Policy Optimization (DUPO). With this integrated pipeline, WebSailor significantly outperforms all opensource agents in complex information-seeking tasks, matching proprietary agents' performance and closing the capability gap.
Conversational agents have made significant progress since ELIZA, expanding their role across various domains, including healthcare, education, and customer service. As these agents become increasingly integrated into daily human interactions, the need for emotional intelligence, particularly empathetic listening, becomes increasingly essential. In this study, we explore how Large Language Models (LLMs) respond when tasked with generating emotionally rich interactions. Starting from a small dataset manually crafted by an expert to reflect empathic behavior, we extended the conversations using two LLMs: ChatGPT and Gemini. We analyzed the emotional progression of the dialogues using both sentiment analysis (via VADER) and expert assessments. While the generated conversations often mirrored the intended emotional structure, human evaluation revealed important differences in the perceived empathy and coherence of the responses. These findings suggest that emotion modeling in dialogues requires not only structural alignment in the expressed emotions but also qualitative depth, highlighting the importance of combining automated and humancentered methods in the development of emotionally competent agents.
Intrusion Detection and Prevention Systems (IDS/IPS) in large enterprises can generate hundreds of thousands of alerts per hour, overwhelming security analysts with logs that demand deep, rapidly evolving domain expertise. Conventional machine-learning detectors trim the alert volume but still yield high false-positive rates, while standard single-pass Retrieval-Augmented Generation (RAG) pipelines often retrieve irrelevant context and fail to justify their predictions. To overcome these shortcomings, we present CyberRAG, a modular, agent-based RAG framework that delivers real-time classification, explanation, and structured reporting for cyber-attacks. A central LLM agent orchestrates (i) a pool of fine-tuned specialized classifiers, each tailored to a distinct attack family; (ii) tool adapters for enrichment and alerting; and (iii) an iterative retrieval-and-reason loop that continuously queries a domain-specific knowledge base until the evidence is both relevant and self-consistent. Unlike traditional RAG systems, CyberRAG embraces an agentic design that enables dynamic control flow and adaptive reasoning. This agent-centric architecture refines its threat labels and natural-language justifications autonomously, reducing false positives and enhancing interpretability. The framework is fully extensible: new attack types can be supported by simply adding a classifier without retraining the core agent. CyberRAG has been evaluated achieving over 94% accuracy per class and pushing final classification accuracy to 94.92% through semantic orchestration. Generated explanations score up to 0.94 in BERTScore and 4.9/5 in GPT-4-based expert evaluation. These results show that agentic, specialist-oriented RAG can pair high detection accuracy with trustworthy, SOC-ready prose, offering a practical and scalable path toward semi-autonomous cyber-defence workflows.
Keyword decision in Sponsored Search Advertising is critical to the success of ad campaigns. While LLM-based methods offer automated keyword generation, they face three major limitations: reliance on large-scale query-keyword pair data, lack of online multi-objective performance monitoring and optimization, and weak quality control in keyword selection. These issues hinder the agentic use of LLMs in fully automating keyword decisions by monitoring and reasoning over key performance indicators such as impressions, clicks, conversions, and CTA effectiveness. To overcome these challenges, we propose OMS, a keyword generation framework that is On-the-fly (requires no training data, monitors online performance, and adapts accordingly), Multi-objective (employs agentic reasoning to optimize keywords based on multiple performance metrics), and Self-reflective (agentically evaluates keyword quality). Experiments on benchmarks and real-world ad campaigns show that OMS outperforms existing methods; ablation and human evaluations confirm the effectiveness of each component and the quality of generated keywords.
Despite improvements by length extrapolation, efficient attention and memory modules, handling infinitely long documents with linear complexity without performance degradation during extrapolation remains the ultimate challenge in long-text processing. We directly optimize for long-text tasks in an end-to-end fashion and introduce a novel agent workflow, MemAgent, which reads text in segments and updates the memory using an overwrite strategy. We extend the DAPO algorithm to facilitate training via independent-context multi-conversation generation. MemAgent has demonstrated superb long-context capabilities, being able to extrapolate from an 8K context trained on 32K text to a 3.5M QA task with performance loss < 5% and achieves 95%+ in 512K RULER test.
As LLMs are increasingly studied as role-playing agents to generate synthetic data for human behavioral research, ensuring that their outputs remain coherent with their assigned roles has become a critical concern. In this paper, we investigate how consistently LLM-based role-playing agents' stated beliefs about the behavior of the people they are asked to role-play ("what they say") correspond to their actual behavior during role-play ("how they act"). Specifically, we establish an evaluation framework to rigorously measure how well beliefs obtained by prompting the model can predict simulation outcomes in advance. Using an augmented version of the GenAgents persona bank and the Trust Game (a standard economic game used to quantify players' trust and reciprocity), we introduce a belief-behavior consistency metric to systematically investigate how it is affected by factors such as: (1) the types of beliefs we elicit from LLMs, like expected outcomes of simulations versus task-relevant attributes of individual characters LLMs are asked to simulate; (2) when and how we present LLMs with relevant information about Trust Game; and (3) how far into the future we ask the model to forecast its actions. We also explore how feasible it is to impose a researcher's own theoretical priors in the event that the originally elicited beliefs are misaligned with research objectives. Our results reveal systematic inconsistencies between LLMs' stated (or imposed) beliefs and the outcomes of their role-playing simulation, at both an individual- and population-level. Specifically, we find that, even when models appear to encode plausible beliefs, they may fail to apply them in a consistent way. These findings highlight the need to identify how and when LLMs' stated beliefs align with their simulated behavior, allowing researchers to use LLM-based agents appropriately in behavioral studies.
Common Business Oriented Language (COBOL) is a programming language used to develop business applications that are widely adopted by financial, business, and government agencies. Due to its age, complexity, and declining number of COBOL developers, maintaining COBOL codebases is becoming increasingly challenging. In particular, the lack of documentation makes it difficult for new developers to effectively understand and maintain COBOL systems. Existing research utilizes large language models (LLMs) to explain the functionality of code snippets. However, COBOL presents unique challenges due to its architectural and syntactical differences, which often cause its code to exceed the token window size of LLMs. In this work, we propose a multi-agent approach that leverages two LLM-based agents working collaboratively to generate explanations for functions, files, and the overall project. These agents incorporate together by utilizing contextual information from the codebase into the code explanation prompts. We evaluate the effectiveness of our approach using 14 open-source, real-world COBOL projects. Our results indicate that our approach performs significantly better than the baseline in function code explanation, with improvements of 12.67%, 18.59%, and 0.62% in terms of METEOR, chrF, and SentenceBERT scores, respectively. At the file level, our approach effectively explains both short and long COBOL files that exceed the token window size of LLMs and surpass the baseline by 4.21%, 10.72%, and 14.68% in explaining the purpose, functionality, and clarity of the generated explanation. At the project level, our approach generates explanations that convey the functionality and purpose of 82% of the selected projects.
This paper explores the integration of advanced Multi-Agent Systems (MAS) techniques to develop a team of agents with enhanced logical reasoning, long-term knowledge retention, and Theory of Mind (ToM) capabilities. By uniting these core components with optimized communication protocols, we create a novel framework called SynergyMAS, which fosters collaborative teamwork and superior problem-solving skills. The system's effectiveness is demonstrated through a product development team case study, where our approach significantly enhances performance and adaptability. These findings highlight SynergyMAS's potential to tackle complex, real-world challenges.
Large language models (LLMs) are rapidly evolving from passive engines of text generation into agentic entities that can plan, remember, invoke external tools, and co-operate with one another. This perspective paper investigates how such LLM agents (and societies thereof) can transform the design space of recommender systems. We introduce a unified formalism that (i) models an individual agent as a tuple comprising its language core, tool set, and hierarchical memory, and (ii) captures a multi-agent recommender as a triple of agents, shared environment, and communication protocol. Within this framework, we present four end-to-end use cases-interactive party planning, synthetic user-simulation for offline evaluation, multi-modal furniture recommendation, and brand-aligned explanation generation-each illustrating a distinct capability unlocked by agentic orchestration. We then surface five cross-cutting challenge families: protocol complexity, scalability, hallucination and error propagation, emergent misalignment (including covert collusion), and brand compliance. For each, we formalize the problem, review nascent mitigation strategies, and outline open research questions. The result is both a blueprint and an agenda: a blueprint that shows how memory-augmented, tool-using LLM agents can be composed into robust recommendation pipelines, and an agenda inviting the RecSys community to develop benchmarks, theoretical guarantees, and governance tools that keep pace with this new degree of autonomy. By unifying agentic abstractions with recommender objectives, the paper lays the groundwork for the next generation of personalized, trustworthy, and context-rich recommendation services.
Designing experiments and result interpretations are core scientific competencies, particularly in biology, where researchers perturb complex systems to uncover the underlying systems. Recent efforts to evaluate the scientific capabilities of large language models (LLMs) fail to test these competencies because wet-lab experimentation is prohibitively expensive: in expertise, time and equipment. We introduce SciGym, a first-in-class benchmark that assesses LLMs' iterative experiment design and analysis abilities in open-ended scientific discovery tasks. SciGym overcomes the challenge of wet-lab costs by running a dry lab of biological systems. These models, encoded in Systems Biology Markup Language, are efficient for generating simulated data, making them ideal testbeds for experimentation on realistically complex systems. We evaluated six frontier LLMs on 137 small systems, and released a total of 350 systems. Our evaluation shows that while more capable models demonstrated superior performance, all models' performance declined significantly as system complexity increased, suggesting substantial room for improvement in the scientific capabilities of LLM agents.
Large language models (LLMs) have rapidly progressed into general-purpose agents capable of solving a broad spectrum of tasks. However, current models remain inefficient at reasoning: they apply fixed inference-time compute regardless of task complexity, often overthinking simple problems while underthinking hard ones. This survey presents a comprehensive review of efficient test-time compute (TTC) strategies, which aim to improve the computational efficiency of LLM reasoning. We introduce a two-tiered taxonomy that distinguishes between L1-controllability, methods that operate under fixed compute budgets, and L2-adaptiveness, methods that dynamically scale inference based on input difficulty or model confidence. We benchmark leading proprietary LLMs across diverse datasets, highlighting critical trade-offs between reasoning performance and token usage. Compared to prior surveys on efficient reasoning, our review emphasizes the practical control, adaptability, and scalability of TTC methods. Finally, we discuss emerging trends such as hybrid thinking models and identify key challenges for future work towards making LLMs more computationally efficient, robust, and responsive to user constraints.
Large language models (LLMs) are excellent at maintaining high-level, convincing dialogues. They are being fast deployed as chatbots and evaluators in sensitive areas, such as peer review and mental health applications. This, along with the disparate accounts on their reasoning capabilities, calls for a closer examination of LLMs and their comprehension of dialogue. In this work we begin by evaluating LLMs' ability to maintain a debate--one of the purest yet most complex forms of human communication. Then we measure how this capability relates to their understanding of what is being talked about, namely, their comprehension of dialogical structures and the pragmatic context. We find that LLMs are capable of maintaining coherent, persuasive debates, often swaying the beliefs of participants and audiences alike. We also note that awareness or suspicion of AI involvement encourage people to be more critical of the arguments made. When polling LLMs on their comprehension of deeper structures of dialogue, however, they cannot demonstrate said understanding. Our findings tie the shortcomings of LLMs-as-evaluators to their (in)ability to understand the context. More broadly, for the field of argumentation theory we posit that, if an agent can convincingly maintain a dialogue, it is not necessary for it to know what it is talking about. Hence, the modelling of pragmatic context and coherence are secondary to effectiveness.
Natural language generation (NLG) is increasingly deployed in high-stakes domains, yet common intrinsic evaluation methods, such as n-gram overlap or sentence plausibility, weakly correlate with actual decision-making efficacy. We propose a decision-oriented framework for evaluating generated text by directly measuring its influence on human and large language model (LLM) decision outcomes. Using market digest texts--including objective morning summaries and subjective closing-bell analyses--as test cases, we assess decision quality based on the financial performance of trades executed by human investors and autonomous LLM agents informed exclusively by these texts. Our findings reveal that neither humans nor LLM agents consistently surpass random performance when relying solely on summaries. However, richer analytical commentaries enable collaborative human-LLM teams to outperform individual human or agent baselines significantly. Our approach underscores the importance of evaluating generated text by its ability to facilitate synergistic decision-making between humans and LLMs, highlighting critical limitations of traditional intrinsic metrics.
Domain specific chatbot applications often involve multi step interactions, such as refining search filters, selecting multiple items, or performing comparisons. Traditional graphical user interfaces (GUIs) handle these workflows by providing explicit "Submit" (commit data) and "Reset" (discard data) actions, allowing back-end systems to track user intent unambiguously. In contrast, conversational agents rely on subtle language cues, which can lead to confusion and incomplete context management. This paper proposes modeling these GUI inspired metaphors acknowledgment (submit like) and context switching (reset-like) as explicit tasks within large language model (LLM) prompts. By capturing user acknowledgment, reset actions, and chain of thought (CoT) reasoning as structured session data, we preserve clarity, reduce user confusion, and align domain-specific chatbot interactions with back-end logic. We demonstrate our approach in hotel booking and customer management scenarios, highlighting improvements in multi-turn task coherence, user satisfaction, and efficiency.
Patents contain rich technical knowledge that can inspire innovative product ideas, yet accessing and interpreting this information remains a challenge. This work explores the use of Large Language Models (LLMs) and autonomous agents to mine and generate product concepts from a given patent. In this work, we design Agent Ideate, a framework for automatically generating product-based business ideas from patents. We experimented with open-source LLMs and agent-based architectures across three domains: Computer Science, Natural Language Processing, and Material Chemistry. Evaluation results show that the agentic approach consistently outperformed standalone LLMs in terms of idea quality, relevance, and novelty. These findings suggest that combining LLMs with agentic workflows can significantly enhance the innovation pipeline by unlocking the untapped potential of business idea generation from patent data.
In this paper, we propose to incorporate the blackboard architecture into LLM multi-agent systems (MASs) so that (1) agents with various roles can share all the information and others' messages during the whole problem-solving process, (2) agents that will take actions are selected based on the current content of the blackboard, and (3) the selection and execution round is repeated until a consensus is reached on the blackboard. We develop the first implementation of this proposal and conduct experiments on commonsense knowledge, reasoning and mathematical datasets. The results show that our system can be competitive with the SOTA static and dynamic MASs by achieving the best average performance, and at the same time manage to spend less tokens. Our proposal has the potential to enable complex and dynamic problem-solving where well-defined structures or workflows are unavailable.
Traditional Data+AI systems utilize data-driven techniques to optimize performance, but they rely heavily on human experts to orchestrate system pipelines, enabling them to adapt to changes in data, queries, tasks, and environments. For instance, while there are numerous data science tools available, developing a pipeline planning system to coordinate these tools remains challenging. This difficulty arises because existing Data+AI systems have limited capabilities in semantic understanding, reasoning, and planning. Fortunately, we have witnessed the success of large language models (LLMs) in enhancing semantic understanding, reasoning, and planning abilities. It is crucial to incorporate LLM techniques to revolutionize data systems for orchestrating Data+AI applications effectively. To achieve this, we propose the concept of a 'Data Agent' - a comprehensive architecture designed to orchestrate Data+AI ecosystems, which focuses on tackling data-related tasks by integrating knowledge comprehension, reasoning, and planning capabilities. We delve into the challenges involved in designing data agents, such as understanding data/queries/environments/tools, orchestrating pipelines/workflows, optimizing and executing pipelines, and fostering pipeline self-reflection. Furthermore, we present examples of data agent systems, including a data science agent, data analytics agents (such as unstructured data analytics agent, semantic structured data analytics agent, data lake analytics agent, and multi-modal data analytics agent), and a database administrator (DBA) agent. We also outline several open challenges associated with designing data agent systems.
Task-oriented dialogue (ToD) systems are designed to help users achieve specific goals through natural language interaction. While recent advances in large language models (LLMs) have significantly improved linguistic fluency and contextual understanding, building effective and emotionally intelligent ToD systems remains a complex challenge. Effective ToD systems must optimise for task success, emotional understanding and responsiveness, and precise information conveyance, all within inherently noisy and ambiguous conversational environments. In this work, we investigate architectural, representational, optimisational as well as emotional considerations of ToD systems. We set up systems covering these design considerations with a challenging evaluation environment composed of a natural-language user simulator coupled with an imperfect natural language understanding module. We propose \textbf{LUSTER}, an \textbf{L}LM-based \textbf{U}nified \textbf{S}ystem for \textbf{T}ask-oriented dialogue with \textbf{E}nd-to-end \textbf{R}einforcement learning with both short-term (user sentiment) and long-term (task success) rewards. Our findings demonstrate that combining LLM capability with structured reward modelling leads to more resilient and emotionally responsive ToD systems, offering a practical path forward for next-generation conversational agents.
Large Language Models (LLMs) have emerged as one of the most significant technological advancements in artificial intelligence in recent years. Their ability to understand, generate, and reason with natural language has transformed how we interact with AI systems. With the development of LLM-based agents and reinforcement-learning-based reasoning models, the study of applying reinforcement learning in agent frameworks has become a new research focus. However, all previous studies face the challenge of deciding the tool calling process and the reasoning process simultaneously, and the chain of reasoning was solely relied on the unprocessed raw result with redundant information and symbols unrelated to the task from the tool, which impose a heavy burden on the model's capability to reason. Therefore, in our research, we proposed a hierarchical framework Agent-as-tool that detach the tool calling process and the reasoning process, which enables the model to focus on the verbally reasoning process while the tool calling process is handled by another agent. Our work had achieved comparable results with only a slight reinforcement fine-tuning on 180 samples, and had achieved exceptionally well performance in Bamboogle with 63.2% of exact match and 75.2% in cover exact match, exceeding Search-R1 by 4.8% in exact match and 3.2% in cover exact match.