MMR-Bench: A Comprehensive Benchmark for Multimodal LLM Routing

Multimodal large language models (MLLMs) have advanced rapidly, yet heterogeneity in architecture, alignment strategies, and efficiency means that no single model is uniformly superior across tasks. In practical deployments, workloads span lightweight OCR to complex multimodal reasoning; using one MLLM for all queries either over-provisions compute on easy instances or sacrifices accuracy on hard ones. Query-level model selection (routing) addresses this tension, but extending routing from text-only LLMs to MLLMs is nontrivial due to modality fusion, wide variation in computational cost across models, and the absence of a standardized, budget-aware evaluation. We present MMR-Bench, a unified benchmark that isolates the multimodal routing problem and enables comparison under fixed candidate sets and cost models. MMR-Bench provides (i) a controlled environment with modality-aware inputs and variable compute budgets, (ii) a broad suite of vision-language tasks covering OCR, general VQA, and multimodal math reasoning, and (iii) strong single-model reference, oracle upper bounds, and representative routing policies. Using MMR-Bench, we show that incorporating multimodal signals improves routing quality. Empirically, these cues improve the cost-accuracy frontier and enable the routed system to exceed the strongest single model's accuracy at roughly 33% of its cost. Furthermore, policies trained on a subset of models and tasks generalize zero-shot to new datasets and text-only benchmarks without retuning, establishing MMR-Bench as a foundation for studying adaptive multimodal model selection and efficient MLLM deployment. The code will be available at: https://github.com/Hunter-Wrynn/MMR-Bench.

EcoAlign: An Economically Rational Framework for Efficient LVLM Alignment

Large Vision-Language Models (LVLMs) exhibit powerful reasoning capabilities but suffer sophisticated jailbreak vulnerabilities. Fundamentally, aligning LVLMs is not just a safety challenge but a problem of economic efficiency. Current alignment methods struggle with the trade-off between safety, utility, and operational costs. Critically, a focus solely on final outputs (process-blindness) wastes significant computational budget on unsafe deliberation. This flaw allows harmful reasoning to be disguised with benign justifications, thereby circumventing simple additive safety scores. To address this, we propose EcoAlign, an inference-time framework that reframes alignment as an economically rational search by treating the LVLM as a boundedly rational agent. EcoAlign incrementally expands a thought graph and scores actions using a forward-looking function (analogous to net present value) that dynamically weighs expected safety, utility, and cost against the remaining budget. To prevent deception, path safety is enforced via the weakest-link principle. Extensive experiments across 3 closed-source and 2 open-source models on 6 datasets show that EcoAlign matches or surpasses state-of-the-art safety and utility at a lower computational cost, thereby offering a principled, economical pathway to robust LVLM alignment.

PixCLIP: Achieving Fine-grained Visual Language Understanding via Any-granularity Pixel-Text Alignment Learning

While the Contrastive Language-Image Pretraining(CLIP) model has achieved remarkable success in a variety of downstream vison language understanding tasks, enhancing its capability for fine-grained image-text alignment remains an active research focus. To this end, most existing works adopt the strategy of explicitly increasing the granularity of visual information processing, e.g., incorporating visual prompts to guide the model focus on specific local regions within the image. Meanwhile, researches on Multimodal Large Language Models(MLLMs) have demonstrated that training with long and detailed textual descriptions can effectively improve the model's fine-grained vision-language alignment. However, the inherent token length limitation of CLIP's text encoder fundamentally limits CLIP to process more granular textual information embedded in long text sequences. To synergistically leverage the advantages of enhancing both visual and textual content processing granularity, we propose PixCLIP, a novel framework designed to concurrently accommodate visual prompt inputs and process lengthy textual descriptions. Specifically, we first establish an automated annotation pipeline capable of generating pixel-level localized, long-form textual descriptions for images. Utilizing this pipeline, we construct LongGRIT, a high-quality dataset comprising nearly 1.5 million samples. Secondly, we replace CLIP's original text encoder with the LLM and propose a three-branch pixel-text alignment learning framework, facilitating fine-grained alignment between image regions and corresponding textual descriptions at arbitrary granularity. Experiments demonstrate that PixCLIP showcases breakthroughs in pixel-level interaction and handling long-form texts, achieving state-of-the-art performance.

MobiFuse: Learning Universal Human Mobility Patterns through Cross-domain Data Fusion

Human mobility modeling is critical for urban planning and transportation management, yet existing datasets often lack the resolution and semantic richness required for comprehensive analysis. To address this, we proposed a cross-domain data fusion framework that integrates multi-modal data of distinct nature and spatio-temporal resolution, including geographical, mobility, socio-demographic, and traffic information, to construct a privacy-preserving and semantically enriched human travel trajectory dataset. This framework is demonstrated through two case studies in Los Angeles (LA) and Egypt, where a domain adaptation algorithm ensures its transferability across diverse urban contexts. Quantitative evaluation shows that the generated synthetic dataset accurately reproduces mobility patterns observed in empirical data. Moreover, large-scale traffic simulations for LA County based on the generated synthetic demand align well with observed traffic. On California's I-405 corridor, the simulation yields a Mean Absolute Percentage Error of 5.85% for traffic volume and 4.36% for speed compared to Caltrans PeMS observations.

Driving with Regulation: Interpretable Decision-Making for Autonomous Vehicles with Retrieval-Augmented Reasoning via LLM

This work presents an interpretable decision-making framework for autonomous vehicles that integrates traffic regulations, norms, and safety guidelines comprehensively and enables seamless adaptation to different regions. While traditional rule-based methods struggle to incorporate the full scope of traffic rules, we develop a Traffic Regulation Retrieval (TRR) Agent based on Retrieval-Augmented Generation (RAG) to automatically retrieve relevant traffic rules and guidelines from extensive regulation documents and relevant records based on the ego vehicle's situation. Given the semantic complexity of the retrieved rules, we also design a reasoning module powered by a Large Language Model (LLM) to interpret these rules, differentiate between mandatory rules and safety guidelines, and assess actions on legal compliance and safety. Additionally, the reasoning is designed to be interpretable, enhancing both transparency and reliability. The framework demonstrates robust performance on both hypothesized and real-world cases across diverse scenarios, along with the ability to adapt to different regions with ease.

Reconstructing Human Mobility Pattern: A Semi-Supervised Approach for Cross-Dataset Transfer Learning

Understanding human mobility patterns is crucial for urban planning, transportation management, and public health. This study tackles two primary challenges in the field: the reliance on trajectory data, which often fails to capture the semantic interdependencies of activities, and the inherent incompleteness of real-world trajectory data. We have developed a model that reconstructs and learns human mobility patterns by focusing on semantic activity chains. We introduce a semi-supervised iterative transfer learning algorithm to adapt models to diverse geographical contexts and address data scarcity. Our model is validated using comprehensive datasets from the United States, where it effectively reconstructs activity chains and generates high-quality synthetic mobility data, achieving a low Jensen-Shannon Divergence (JSD) value of 0.001, indicating a close similarity between synthetic and real data. Additionally, sparse GPS data from Egypt is used to evaluate the transfer learning algorithm, demonstrating successful adaptation of US mobility patterns to Egyptian contexts, achieving a 64\% of increase in similarity, i.e., a JSD reduction from 0.09 to 0.03. This mobility reconstruction model and the associated transfer learning algorithm show significant potential for global human mobility modeling studies, enabling policymakers and researchers to design more effective and culturally tailored transportation solutions.

Human Mobility Modeling with Limited Information via Large Language Models

Understanding human mobility patterns has traditionally been a complex challenge in transportation modeling. Due to the difficulties in obtaining high-quality training datasets across diverse locations, conventional activity-based models and learning-based human mobility modeling algorithms are particularly limited by the availability and quality of datasets. Furthermore, current research mainly focuses on the spatial-temporal travel pattern but lacks an understanding of the semantic information between activities, which is crucial for modeling the interdependence between activities. In this paper, we propose an innovative Large Language Model (LLM) empowered human mobility modeling framework. Our proposed approach significantly reduces the reliance on detailed human mobility statistical data, utilizing basic socio-demographic information of individuals to generate their daily mobility patterns. We have validated our results using the NHTS and SCAG-ABM datasets, demonstrating the effective modeling of mobility patterns and the strong adaptability of our framework across various geographic locations.

Semantic Trajectory Data Mining with LLM-Informed POI Classification

Human travel trajectory mining is crucial for transportation systems, enhancing route optimization, traffic management, and the study of human travel patterns. Previous rule-based approaches without the integration of semantic information show a limitation in both efficiency and accuracy. Semantic information, such as activity types inferred from Points of Interest (POI) data, can significantly enhance the quality of trajectory mining. However, integrating these insights is challenging, as many POIs have incomplete feature information, and current learning-based POI algorithms require the integrity of datasets to do the classification. In this paper, we introduce a novel pipeline for human travel trajectory mining. Our approach first leverages the strong inferential and comprehension capabilities of large language models (LLMs) to annotate POI with activity types and then uses a Bayesian-based algorithm to infer activity for each stay point in a trajectory. In our evaluation using the OpenStreetMap (OSM) POI dataset, our approach achieves a 93.4% accuracy and a 96.1% F-1 score in POI classification, and a 91.7% accuracy with a 92.3% F-1 score in activity inference.

RLRF:Reinforcement Learning from Reflection through Debates as Feedback for Bias Mitigation in LLMs

Biases and stereotypes in Large Language Models (LLMs) can have negative implications for user experience and societal outcomes. Current approaches to bias mitigation like Reinforcement Learning from Human Feedback (RLHF) rely on costly manual feedback. While LLMs have the capability to understand logic and identify biases in text, they often struggle to effectively acknowledge and address their own biases due to factors such as prompt influences, internal mechanisms, and policies. We found that informing LLMs that the content they generate is not their own and questioning them about potential biases in the text can significantly enhance their recognition and improvement capabilities regarding biases. Based on this finding, we propose RLRF (Reinforcement Learning from Reflection through Debates as Feedback), replacing human feedback with AI for bias mitigation. RLRF engages LLMs in multi-role debates to expose biases and gradually reduce biases in each iteration using a ranking scoring mechanism. The dialogue are then used to create a dataset with high-bias and low-bias instances to train the reward model in reinforcement learning. This dataset can be generated by the same LLMs for self-reflection or a superior LLMs guiding the former in a student-teacher mode to enhance its logical reasoning abilities. Experimental results demonstrate the significant effectiveness of our approach in bias reduction.

Deceiving to Enlighten: Coaxing LLMs to Self-Reflection for Enhanced Bias Detection and Mitigation

Large Language Models (LLMs) embed complex biases and stereotypes that can lead to detrimental user experiences and societal consequences, often without conscious awareness from the models themselves. This paper emphasizes the importance of equipping LLMs with mechanisms for better self-reflection and bias recognition. Our experiments demonstrate that by informing LLMs that their generated content does not represent their own views and questioning them about bias, their capability to identify and address biases improves. This enhancement is attributed to the internal attention mechanisms and potential internal sensitivity policies of LLMs. Building upon these findings, we propose a novel method to diminish bias in LLM outputs. This involves engaging LLMs in multi-role scenarios acting as different roles where they are tasked for bias exposure, with a role of an impartial referee in the end of each loop of debate. A ranking scoring mechanism is employed to quantify bias levels, enabling more refined reflections and superior output quality. Comparative experimental results confirm that our method outperforms existing approaches in reducing bias, making it a valuable contribution to efforts towards more ethical AI systems.