University of Illinois
Abstract
Leveraging multimodal large language models (MLLMs) to develop embodied agents offers significant promise for addressing complex real-world tasks. However, current evaluation benchmarks remain predominantly language-centric or heavily reliant on simulated environments, rarely probing the nuanced, knowledge-intensive reasoning essential for practical, real-world scenarios. To bridge this critical gap, we introduce the task of Sparsely Grounded Visual Navigation, explicitly designed to evaluate the sequential decision-making abilities of MLLMs in challenging, knowledge-intensive real-world environments. We operationalize this task with CityNav, a comprehensive benchmark encompassing four diverse global cities, specifically constructed to assess raw MLLM-driven agents in city navigation. Agents are required to rely solely on visual inputs and internal multimodal reasoning to sequentially navigate 50+ decision points without additional environmental annotations or specialized architectural modifications. Crucially, agents must autonomously achieve localization through interpreting city-specific cues and recognizing landmarks, perform spatial reasoning, and strategically plan and execute routes to their destinations. Through extensive evaluations, we demonstrate that current state-of-the-art MLLMs and standard reasoning techniques (e.g., Chain-of-Thought, Reflection) significantly underperform in this challenging setting. To address this, we propose Verbalization of Path (VoP), which explicitly grounds the agent's internal reasoning by probing an explicit cognitive map (key landmarks and directions toward the destination) from the MLLMs, substantially enhancing navigation success. Project Webpage: https://dwipddalal.github.io/AgentNav/
Leveraging multimodal large language models (MLLMs) to develop embodied agents offers significant promise for addressing complex real-world tasks. However, current evaluation benchmarks remain predominantly language-centric or heavily reliant on simulated environments, rarely probing the nuanced, knowledge-intensive reasoning essential for practical, real-world scenarios. To bridge this critical gap, we introduce the task of Sparsely Grounded Visual Navigation, explicitly designed to evaluate the sequential decision-making abilities of MLLMs in challenging, knowledge-intensive real-world environments. We operationalize this task with CityNav, a comprehensive benchmark encompassing four diverse global cities, specifically constructed to assess raw MLLM-driven agents in city navigation. Agents are required to rely solely on visual inputs and internal multimodal reasoning to sequentially navigate 50+ decision points without additional environmental annotations or specialized architectural modifications. Crucially, agents must autonomously achieve localization through interpreting city-specific cues and recognizing landmarks, perform spatial reasoning, and strategically plan and execute routes to their destinations. Through extensive evaluations, we demonstrate that current state-of-the-art MLLMs and standard reasoning techniques (e.g., Chain-of-Thought, Reflection) significantly underperform in this challenging setting. To address this, we propose Verbalization of Path (VoP), which explicitly grounds the agent's internal reasoning by probing an explicit cognitive map (key landmarks and directions toward the destination) from the MLLMs, substantially enhancing navigation success. Project Webpage: https://dwipddalal.github.io/AgentNav/
Why we are recommending this paper?
Due to your Interest in: Travel Search
This paper explores the development of intelligent navigation systems, directly aligning with your interest in travel planning and recommendations. The focus on multimodal large language models for embodied agents offers a novel approach to understanding and facilitating travel experiences.
Due to your Interest in: Travel Search
This paper explores the development of intelligent navigation systems, directly aligning with your interest in travel planning and recommendations. The focus on multimodal large language models for embodied agents offers a novel approach to understanding and facilitating travel experiences.
UC Santa Barbara
AI Insights
- The study examines the performance of various language models on a dataset with diverse user profiles, highlighting the importance of personality traits and conversation style in determining likability. [3]
- The results show that top-performing models (GPT-5 and Claude Sonnet 4) are stable across different user types, while others struggle to adapt. [3]
- The introduction of Dynamic User Profile (DUP) improves performance for top models without additional training, indicating the potential of lightweight preference tracking. [3]
- Likability: The degree to which a language model is perceived as likable or relatable by users. [3]
- Conversation style preferences: The way individuals prefer to communicate, including aspects such as directness, formality, and conversation length. [3]
- Lightweight preference tracking (DUP) shows promise in improving performance without additional training. [3]
- The study's taxonomy of personality traits and conversation style preferences provides a comprehensive framework for understanding user behavior. [2]
Abstract
A personalized LLM should remember user facts, apply them correctly, and adapt over time to provide responses that the user prefers. Existing LLM personalization benchmarks are largely centered on two axes: accurately recalling user information and accurately applying remembered information in downstream tasks. We argue that a third axis, likability, is both subjective and central to user experience, yet under-measured by current benchmarks. To measure likability holistically, we introduce LikeBench, a multi-session, dynamic evaluation framework that measures likability across multiple dimensions by how much an LLM can adapt over time to a user's preferences to provide more likable responses. In LikeBench, the LLMs engage in conversation with a simulated user and learn preferences only from the ongoing dialogue. As the interaction unfolds, models try to adapt to responses, and after each turn, they are evaluated for likability across seven dimensions by the same simulated user. To the best of our knowledge, we are the first to decompose likability into multiple diagnostic metrics: emotional adaptation, formality matching, knowledge adaptation, reference understanding, conversation length fit, humor fit, and callback, which makes it easier to pinpoint where a model falls short. To make the simulated user more realistic and discriminative, LikeBench uses fine-grained, psychologically grounded descriptive personas rather than the coarse high/low trait rating based personas used in prior work. Our benchmark shows that strong memory performance does not guarantee high likability: DeepSeek R1, with lower memory accuracy (86%, 17 facts/profile), outperformed Qwen3 by 28% on likability score despite Qwen3's higher memory accuracy (93%, 43 facts/profile). Even SOTA models like GPT-5 adapt well in short exchanges but show only limited robustness in longer, noisier interactions.
A personalized LLM should remember user facts, apply them correctly, and adapt over time to provide responses that the user prefers. Existing LLM personalization benchmarks are largely centered on two axes: accurately recalling user information and accurately applying remembered information in downstream tasks. We argue that a third axis, likability, is both subjective and central to user experience, yet under-measured by current benchmarks. To measure likability holistically, we introduce LikeBench, a multi-session, dynamic evaluation framework that measures likability across multiple dimensions by how much an LLM can adapt over time to a user's preferences to provide more likable responses. In LikeBench, the LLMs engage in conversation with a simulated user and learn preferences only from the ongoing dialogue. As the interaction unfolds, models try to adapt to responses, and after each turn, they are evaluated for likability across seven dimensions by the same simulated user. To the best of our knowledge, we are the first to decompose likability into multiple diagnostic metrics: emotional adaptation, formality matching, knowledge adaptation, reference understanding, conversation length fit, humor fit, and callback, which makes it easier to pinpoint where a model falls short. To make the simulated user more realistic and discriminative, LikeBench uses fine-grained, psychologically grounded descriptive personas rather than the coarse high/low trait rating based personas used in prior work. Our benchmark shows that strong memory performance does not guarantee high likability: DeepSeek R1, with lower memory accuracy (86%, 17 facts/profile), outperformed Qwen3 by 28% on likability score despite Qwen3's higher memory accuracy (93%, 43 facts/profile). Even SOTA models like GPT-5 adapt well in short exchanges but show only limited robustness in longer, noisier interactions.
Why we are recommending this paper?
Due to your Interest in: Travel Personalization
Given your interest in travel personalization, this research directly tackles the challenge of tailoring LLM responses to individual user preferences. Understanding how LLMs can learn and adapt to subjective likes and dislikes is crucial for effective travel recommendations.
Due to your Interest in: Travel Personalization
Given your interest in travel personalization, this research directly tackles the challenge of tailoring LLM responses to individual user preferences. Understanding how LLMs can learn and adapt to subjective likes and dislikes is crucial for effective travel recommendations.
University of Cambridge
AI Insights
- The paper discusses various methods for analyzing paired comparison data, including the Bradley-Terry model, generalized linear mixed models, and Bayesian inference. [3]
- Paired comparison data: Data where each pair of items is compared to determine which one is preferred or superior. [3]
- Bradley-Terry model: A statistical model used to analyze paired comparison data, assuming that the probability of a item being preferred over another is proportional to its ability parameter. [3]
- The paper concludes by highlighting the importance of considering multiple methods and models when analyzing paired comparison data, as each method has its own strengths and limitations. [3]
- Some sections of the paper may be difficult to follow for readers without prior knowledge of paired comparison data analysis. [3]
- The authors discuss various applications of paired comparison data analysis, including sports, social networks, and animal behavior. [3]
- The paper discusses various statistical methods for analyzing paired comparison data, including the Bradley-Terry model, generalized linear mixed models, and Bayesian inference. [3]
- The authors also emphasize the need for further research in developing more robust and efficient methods for analyzing paired comparison data. [2]
Abstract
We consider the problem of ranking objects from noisy pairwise comparisons, for example, ranking tennis players from the outcomes of matches. We follow a standard approach to this problem and assume that each object has an unobserved strength and that the outcome of each comparison depends probabilistically on the strengths of the comparands. However, we do not assume to know a priori how skills affect outcomes. Instead, we present an efficient algorithm for simultaneously inferring both the unobserved strengths and the function that maps strengths to probabilities. Despite this problem being under-constrained, we present experimental evidence that the conclusions of our Bayesian approach are robust to different model specifications. We include several case studies to exemplify the method on real-world data sets.
We consider the problem of ranking objects from noisy pairwise comparisons, for example, ranking tennis players from the outcomes of matches. We follow a standard approach to this problem and assume that each object has an unobserved strength and that the outcome of each comparison depends probabilistically on the strengths of the comparands. However, we do not assume to know a priori how skills affect outcomes. Instead, we present an efficient algorithm for simultaneously inferring both the unobserved strengths and the function that maps strengths to probabilities. Despite this problem being under-constrained, we present experimental evidence that the conclusions of our Bayesian approach are robust to different model specifications. We include several case studies to exemplify the method on real-world data sets.
Why we are recommending this paper?
Due to your Interest in: Travel Ranking
This paper addresses the core problem of ranking, a fundamental component of travel search and recommendation systems. The focus on noisy pairwise comparisons provides valuable insights into how to build more effective ranking models.
Due to your Interest in: Travel Ranking
This paper addresses the core problem of ranking, a fundamental component of travel search and recommendation systems. The focus on noisy pairwise comparisons provides valuable insights into how to build more effective ranking models.
Sichuan University
Abstract
The orienteering problem (OP) is a combinatorial optimization problem that seeks a path visiting a subset of locations to maximize collected rewards under a limited resource budget. This article presents a systematic PRISMA-based review of OP research published between 2017 and 2025, with a focus on models and methods that have shaped subsequent developments in the field. We introduce a component-based taxonomy that decomposes OP variants into time-, path-, node-, structure-, and information-based extensions. This framework unifies classical and emerging variants -- including stochastic, time-dependent, Dubins, Set, and multi-period OPs -- within a single structural perspective. We further categorize solution approaches into exact algorithms, heuristics and metaheuristics, and learning-based methods, with particular emphasis on matheuristics and recent advances in artificial intelligence, especially reinforcement learning and neural networks, which enhance scalability in large-scale and information-rich settings. Building on this unified view, we discuss how different components affect computational complexity and polyhedral properties and identify open challenges related to robustness, sustainability, and AI integration. The survey thus provides both a consolidated reference for existing OP research and a structured agenda for future theoretical and applied work.
The orienteering problem (OP) is a combinatorial optimization problem that seeks a path visiting a subset of locations to maximize collected rewards under a limited resource budget. This article presents a systematic PRISMA-based review of OP research published between 2017 and 2025, with a focus on models and methods that have shaped subsequent developments in the field. We introduce a component-based taxonomy that decomposes OP variants into time-, path-, node-, structure-, and information-based extensions. This framework unifies classical and emerging variants -- including stochastic, time-dependent, Dubins, Set, and multi-period OPs -- within a single structural perspective. We further categorize solution approaches into exact algorithms, heuristics and metaheuristics, and learning-based methods, with particular emphasis on matheuristics and recent advances in artificial intelligence, especially reinforcement learning and neural networks, which enhance scalability in large-scale and information-rich settings. Building on this unified view, we discuss how different components affect computational complexity and polyhedral properties and identify open challenges related to robustness, sustainability, and AI integration. The survey thus provides both a consolidated reference for existing OP research and a structured agenda for future theoretical and applied work.
Why we are recommending this paper?
Due to your Interest in: Travel Planning
The orienteering problem is a classic optimization challenge, often used in route planning scenarios – a key interest for you. This survey provides a comprehensive overview of the field, potentially uncovering advanced techniques relevant to travel itinerary creation.
Due to your Interest in: Travel Planning
The orienteering problem is a classic optimization challenge, often used in route planning scenarios – a key interest for you. This survey provides a comprehensive overview of the field, potentially uncovering advanced techniques relevant to travel itinerary creation.
TU Darmstadt
Abstract
The existence of a polynomial pivot rule for the simplex method for linear programming, policy iteration for Markov decision processes, and strategy improvement for parity games each are prominent open problems in their respective fields. While numerous natural candidates for efficient rules have been eliminated, all existing lower bound constructions are tailored to individual or small sets of pivot rules. We introduce a unified framework for formalizing classes of rules according to the information about the input that they rely on. Within this framework, we show lower bounds for \emph{ranking-based} classes of rules that base their decisions on orderings of the improving pivot steps induced by the underlying data. Our first result is a superpolynomial lower bound for strategy improvement, obtained via a family of sink parity games, which applies to memory-based generalizations of Bland's rule that only access the input by comparing the ranks of improving edges in some global order. Our second result is a subexponential lower bound for policy iteration, obtained via a family of Markov decision processes, which applies to memoryless rules that only access the input by comparing improving actions according to their ranks in a global order, their reduced costs, and the associated improvements in objective value. Both results carry over to the simplex method for linear programming.
The existence of a polynomial pivot rule for the simplex method for linear programming, policy iteration for Markov decision processes, and strategy improvement for parity games each are prominent open problems in their respective fields. While numerous natural candidates for efficient rules have been eliminated, all existing lower bound constructions are tailored to individual or small sets of pivot rules. We introduce a unified framework for formalizing classes of rules according to the information about the input that they rely on. Within this framework, we show lower bounds for \emph{ranking-based} classes of rules that base their decisions on orderings of the improving pivot steps induced by the underlying data. Our first result is a superpolynomial lower bound for strategy improvement, obtained via a family of sink parity games, which applies to memory-based generalizations of Bland's rule that only access the input by comparing the ranks of improving edges in some global order. Our second result is a subexponential lower bound for policy iteration, obtained via a family of Markov decision processes, which applies to memoryless rules that only access the input by comparing improving actions according to their ranks in a global order, their reduced costs, and the associated improvements in objective value. Both results carry over to the simplex method for linear programming.
Why we are recommending this paper?
Due to your Interest in: Travel Ranking
This paper delves into the theoretical foundations of ranking algorithms, which is essential for understanding how to build robust and efficient travel recommendation systems. The exploration of pivot rules offers a deeper understanding of ranking methodologies.
Due to your Interest in: Travel Ranking
This paper delves into the theoretical foundations of ranking algorithms, which is essential for understanding how to build robust and efficient travel recommendation systems. The exploration of pivot rules offers a deeper understanding of ranking methodologies.
Universit de Toulouse
Abstract
The lack of GPS data limits the ability to reconstruct the actual routes taken by cyclists in urban areas. This article introduces an inference method based solely on trip durations and origin-destination pairs from bike-sharing system (BSS) users. Travel time distributions are modeled using log-normal mixture models, allowing us to identify the presence of distinct behaviors. The approach is applied to 3.8 million trips recorded in 2022 in the Toulouse metropolitan area, with observed durations compared against travel times estimated by OpenStreetMap (OSM). Results show that, for many station pairs, trip durations align closely with the fastest route suggested by OSM, reflecting a dominant and routine practice. In other cases, mixture models reveal more heterogeneous behaviors, including longer trips, detours, or intermediate stops. This approach highlights both the stability and diversity of cycling practices, providing a robust tool for usage analysis in data-limited contexts, and offering new insights into urban mobility dynamics without relying on spatially explicit data.
The lack of GPS data limits the ability to reconstruct the actual routes taken by cyclists in urban areas. This article introduces an inference method based solely on trip durations and origin-destination pairs from bike-sharing system (BSS) users. Travel time distributions are modeled using log-normal mixture models, allowing us to identify the presence of distinct behaviors. The approach is applied to 3.8 million trips recorded in 2022 in the Toulouse metropolitan area, with observed durations compared against travel times estimated by OpenStreetMap (OSM). Results show that, for many station pairs, trip durations align closely with the fastest route suggested by OSM, reflecting a dominant and routine practice. In other cases, mixture models reveal more heterogeneous behaviors, including longer trips, detours, or intermediate stops. This approach highlights both the stability and diversity of cycling practices, providing a robust tool for usage analysis in data-limited contexts, and offering new insights into urban mobility dynamics without relying on spatially explicit data.
AI Insights
- The detailed interpretation of the components can be challenging. [3]
- The study uses a statistical approach to analyze bike-sharing system (BSS) data in Toulouse, France, and reveals that a large proportion of users converge towards a dominant practice, reflecting routine or habitual trips. [2]
Why we are recommending this paper?
Due to your Interest in: Travel Planning
Due to your Interest in: Travel Planning
Zhejiang University
Abstract
The deployment of Large Language Models (LLMs) in interactive systems necessitates a deep alignment with the nuanced and dynamic preferences of individual users. Current alignment techniques predominantly address universal human values or static, single-turn preferences, thereby failing to address the critical needs of long-term personalization and the initial user cold-start problem. To bridge this gap, we propose PersonalAgent, a novel user-centric lifelong agent designed to continuously infer and adapt to user preferences. PersonalAgent constructs and dynamically refines a unified user profile by decomposing dialogues into single-turn interactions, framing preference inference as a sequential decision-making task. Experiments show that PersonalAgent achieves superior performance over strong prompt-based and policy optimization baselines, not only in idealized but also in noisy conversational contexts, while preserving cross-session preference consistency. Furthermore, human evaluation confirms that PersonalAgent excels at capturing user preferences naturally and coherently. Our findings underscore the importance of lifelong personalization for developing more inclusive and adaptive conversational agents. Our code is available here.
The deployment of Large Language Models (LLMs) in interactive systems necessitates a deep alignment with the nuanced and dynamic preferences of individual users. Current alignment techniques predominantly address universal human values or static, single-turn preferences, thereby failing to address the critical needs of long-term personalization and the initial user cold-start problem. To bridge this gap, we propose PersonalAgent, a novel user-centric lifelong agent designed to continuously infer and adapt to user preferences. PersonalAgent constructs and dynamically refines a unified user profile by decomposing dialogues into single-turn interactions, framing preference inference as a sequential decision-making task. Experiments show that PersonalAgent achieves superior performance over strong prompt-based and policy optimization baselines, not only in idealized but also in noisy conversational contexts, while preserving cross-session preference consistency. Furthermore, human evaluation confirms that PersonalAgent excels at capturing user preferences naturally and coherently. Our findings underscore the importance of lifelong personalization for developing more inclusive and adaptive conversational agents. Our code is available here.
AI Insights
- The proposed approach demonstrates the importance of rich contextual and behavioral signals for personalization, highlighting the need for more nuanced understanding of user characteristics and behaviors. [3]
- Previous work in user profile and recommender systems emphasizes the importance of rich contextual and behavioral signals for personalization. [3]
- Personalization in Conversational AI The paper presents a new approach to making conversational AI more personalized. [3]
- It focuses on understanding users' preferences and adapting to their needs, rather than relying solely on static information. [3]
- The paper presents a comprehensive approach to personalization in conversational AI, focusing on aligning large language models (LLMs) with user preferences. [2]
Why we are recommending this paper?
Due to your Interest in: Travel Personalization
Due to your Interest in: Travel Personalization
Complexity Science Hub
Abstract
The automotive industry is undergoing transformation, driven by the electrification of powertrains, the rise of software-defined vehicles, and the adoption of circular economy concepts. These trends blur the boundaries between the automotive sector and other industries. Unlike internal combustion engine (ICE) production, where mechanical capabilities dominated, competitiveness in electric vehicle (EV) production increasingly depends on expertise in electronics, batteries, and software. This study investigates whether and how firms' ability to leverage cross-industry diversification contributes to competitive advantage. We develop a country-level product space covering all industries and an industry-specific product space covering over 900 automotive components. This allows us to identify clusters of parts that are exported together, revealing shared manufacturing capabilities. Closeness centrality in the country-level product space, rather than simple proximity, is a strong predictor of where new comparative advantages are likely to emerge. We examine this relationship across industrial sectors to establish patterns of path dependency, diversification and capability formation, and then focus on the EV transition. New strengths in vehicles and aluminium products in the EU are expected to generate 5 and 4.6 times more EV-specific strengths, respectively, than other EV-relevant sectors over the next decade, compared to only 1.6 and 4.5 new strengths in already diversified China. Countries such as South Korea, China, the US and Canada show strong potential for diversification into EV-related products, while established producers in the EU are likely to come under pressure. These findings suggest that the success of the automotive transformation depends on regions' ability to mobilize existing industrial capabilities, particularly in sectors such as machinery and electronic equipment.
The automotive industry is undergoing transformation, driven by the electrification of powertrains, the rise of software-defined vehicles, and the adoption of circular economy concepts. These trends blur the boundaries between the automotive sector and other industries. Unlike internal combustion engine (ICE) production, where mechanical capabilities dominated, competitiveness in electric vehicle (EV) production increasingly depends on expertise in electronics, batteries, and software. This study investigates whether and how firms' ability to leverage cross-industry diversification contributes to competitive advantage. We develop a country-level product space covering all industries and an industry-specific product space covering over 900 automotive components. This allows us to identify clusters of parts that are exported together, revealing shared manufacturing capabilities. Closeness centrality in the country-level product space, rather than simple proximity, is a strong predictor of where new comparative advantages are likely to emerge. We examine this relationship across industrial sectors to establish patterns of path dependency, diversification and capability formation, and then focus on the EV transition. New strengths in vehicles and aluminium products in the EU are expected to generate 5 and 4.6 times more EV-specific strengths, respectively, than other EV-relevant sectors over the next decade, compared to only 1.6 and 4.5 new strengths in already diversified China. Countries such as South Korea, China, the US and Canada show strong potential for diversification into EV-related products, while established producers in the EU are likely to come under pressure. These findings suggest that the success of the automotive transformation depends on regions' ability to mobilize existing industrial capabilities, particularly in sectors such as machinery and electronic equipment.
AI Insights
- EV-specific closeness: The average pairwise closeness between products in each chapter pair (i.e., sum over Cij normalized by Nj). [3]
- Closeness gain to EV specific products: Normalized closeness between the 30 HS chapters with the highest EU export value in 2022 to EV specific products, measured as the average pairwise closeness between products in each chapter pair (i.e., sum over Cij normalized by Nj). [3]
- RCA: Revealed Comparative Advantage. [3]
- A measure of a country's comparative advantage in producing a particular product or industry. [3]
- The study uses a component-based approach combining firm-level and trade data to examine the structural transformation of the automotive industry. [2]
Why we are recommending this paper?
Due to your Interest in: Travel Industry
Due to your Interest in: Travel Industry
Interests not found
We did not find any papers that match the below interests.
Try other terms also consider if the content exists in arxiv.org.
- Travel Recommendations
- Travel
- Travel Itinerary Creation
You can edit or add more interests any time.
Help us improve your experience!
This project is on its early stages your feedback can be pivotal on the future of the project.
Let us know what you think about this week's papers and suggestions!
Give Feedback