Large Language Models (LLMs) demonstrate strong generalization and reasoning abilities, making them well-suited for complex decision-making tasks such as medical consultation (MC). However, existing LLM-based methods often fail to capture the dual nature of MC, which entails two distinct sub-tasks: symptom inquiry, a sequential decision-making process, and disease diagnosis, a classification problem. This mismatch often results in ineffective symptom inquiry and unreliable disease diagnosis. To address this, we propose \textbf{DDO}, a novel LLM-based framework that performs \textbf{D}ual-\textbf{D}ecision \textbf{O}ptimization by decoupling the two sub-tasks and optimizing them with distinct objectives through a collaborative multi-agent workflow. Experiments on three real-world MC datasets show that DDO consistently outperforms existing LLM-based approaches and achieves competitive performance with state-of-the-art generation-based methods, demonstrating its effectiveness in the MC task. The code is available at https://github.com/zh-jia/DDO.

Current evaluation frameworks for foundation models rely heavily on static, manually curated benchmarks, limiting their ability to capture the full breadth of model capabilities. This paper introduces Active learning for Capability Evaluation (ACE), a novel framework for scalable, automated, and fine-grained evaluation of foundation models. ACE leverages the knowledge embedded in powerful frontier models to decompose a domain into semantically meaningful capabilities and generates diverse evaluation tasks, significantly reducing human effort. In Mathematics, ACE generated 433 capabilities and 11,800 tasks, covering 94% of Wikipedia-defined skills in the domain while introducing novel, coherent ones. To maximize efficiency, ACE fits a capability model in latent semantic space, allowing reliable approximation of a subject model's performance by evaluating only a subset of capabilities via active learning. It reaches within 0.01 RMSE of exhaustive evaluation by evaluating less than half of capabilities. Compared to static datasets, ACE provides more balanced coverage and uncovers fine-grained differences that aggregate metrics fail to capture. Our results demonstrate that ACE provides a more complete and informative picture of model capabilities, which is essential for safe and well-informed deployment of foundation models.

Training effective Vision-Language Models (VLMs) for GUI agents typically depends on large-scale annotated datasets, whose collection is both labor-intensive and error-prone. We introduce K-step GUI Transition, a self-supervised inverse dynamics task in which VLMs learn GUI dynamics by predicting the initial action that causes a transition between two GUI states. This approach eliminates the need for natural language instructions and enables scalable dataset construction from existing GUI trajectories or automated exploration. Building on this task, we propose GUI-Shift, a reinforcement learning (RL) framework that combines rule-based optimization with data filtering to improve VLM performance. We conduct extensive experiments using multiple VLM backbones across four benchmarks, spanning GUI task automation (AndroidControl, GUI Odyssey) and GUI grounding (ScreenSpot-v2, ScreenSpot-Pro). Our results show that training on GUI-Shift generalizes well to both GUI automation and grounding tasks, yielding up to an 11.2% increase in GUI automation accuracy. This study underscores the potential of self-supervised RL to leverage unlabeled GUI trajectories and offers a scalable alternative to training with annotated samples.

Large language models have recently made great strides in reasoning task performance through chain-of-thought (CoT) strategies trained via reinforcement learning; however, these "reasoning large language models" (RLLMs) remain imperfect reasoners, and understanding the frequencies and causes of their failure modes is important for both users and developers. We test o1-mini, o3-mini, DeepSeek-R1, Claude 3.7 Sonnet, Gemini 2.5 Pro Preview, and Grok 3 Mini Beta on graph coloring as a variable-complexity constraint-satisfaction logic problem, and find evidence from both error rate comparisons and CoT/explanation text analysis that RLLMs are prone to hallucinate graph edges not specified in the prompt. This phenomenon persists across multiple problem complexity levels and semantic frames, and it appears to account for a significant fraction of the incorrect answers from every tested model, and the vast majority of them for some models. We also validate the generalizability of this input-conflicting hallucination phenomenon with smaller-scale experiments on a type of stable matching problem. Our results indicate that RLLMs may possess broader issues with misrepresentation of problem specifics, and we offer suggestions for design choices to mitigate this weakness.

Large Language Models (LLMs) have shown remarkable capabilities, with optimizing their input prompts playing a pivotal role in maximizing their performance. However, while LLM prompts consist of both the task-agnostic system prompts and task-specific user prompts, existing work on prompt optimization has focused on user prompts specific to individual queries or tasks, and largely overlooked the system prompt that is, once optimized, applicable across different tasks and domains. Motivated by this, we introduce the novel problem of bilevel system prompt optimization, whose objective is to design system prompts that are robust to diverse user prompts and transferable to unseen tasks. To tackle this problem, we then propose a meta-learning framework, which meta-learns the system prompt by optimizing it over various user prompts across multiple datasets, while simultaneously updating the user prompts in an iterative manner to ensure synergy between them. We conduct experiments on 14 unseen datasets spanning 5 different domains, on which we show that our approach produces system prompts that generalize effectively to diverse user prompts. Also, our findings reveal that the optimized system prompt enables rapid adaptation even to unseen tasks, requiring fewer optimization steps for test-time user prompts while achieving improved performance.

Future self-adaptive robots are expected to operate in highly dynamic environments while effectively managing uncertainties. However, identifying the sources and impacts of uncertainties in such robotic systems and defining appropriate mitigation strategies is challenging due to the inherent complexity of self-adaptive robots and the lack of comprehensive knowledge about the various factors influencing uncertainty. Hence, practitioners often rely on intuition and past experiences from similar systems to address uncertainties. In this article, we evaluate the potential of large language models (LLMs) in enabling a systematic and automated approach to identify uncertainties in self-adaptive robotics throughout the software engineering lifecycle. For this evaluation, we analyzed 10 advanced LLMs with varying capabilities across four industrial-sized robotics case studies, gathering the practitioners' perspectives on the LLM-generated responses related to uncertainties. Results showed that practitioners agreed with 63-88% of the LLM responses and expressed strong interest in the practicality of LLMs for this purpose.

Despite the rapid growth of machine learning research, corresponding code implementations are often unavailable, making it slow and labor-intensive for researchers to reproduce results and build upon prior work. In the meantime, recent Large Language Models (LLMs) excel at understanding scientific documents and generating high-quality code. Inspired by this, we introduce PaperCoder, a multi-agent LLM framework that transforms machine learning papers into functional code repositories. PaperCoder operates in three stages: planning, where it constructs a high-level roadmap, designs the system architecture with diagrams, identifies file dependencies, and generates configuration files; analysis, which focuses on interpreting implementation-specific details; and generation, where modular, dependency-aware code is produced. Moreover, each phase is instantiated through a set of specialized agents designed to collaborate effectively across the pipeline. We then evaluate PaperCoder on generating code implementations from machine learning papers based on both model-based and human evaluations, particularly from the authors of those papers, with author-released repositories as ground truth if available. Our results demonstrate the effectiveness of PaperCoder in creating high-quality, faithful implementations. Furthermore, it consistently shows strengths in the recently released PaperBench benchmark, surpassing strong baselines by substantial margins. Code is available at: https://github.com/going-doer/Paper2Code.

There has been extensive prior work exploring how psychological factors such as anthropomorphism affect the adoption of Shared Autonomous Vehicles (SAVs). However, limited research has been conducted on how prompt strategies in large language models (LLM)-powered conversational SAV agents affect users' perceptions, experiences, and intentions to adopt such technology. In this work, we investigate how conversational SAV agents powered by LLMs drive these psychological factors, such as psychological ownership, the sense of possession a user may come to feel towards an entity or object they may not legally own. We designed four SAV agents with varying levels of anthropomorphic characteristics and psychological ownership triggers. Quantitative measures of psychological ownership, anthropomorphism, quality of service, disclosure tendency, sentiment of SAV responses, and overall acceptance were collected after participants interacted with each SAV. Qualitative feedback was also gathered regarding the experience of psychological ownership during the interactions. The results indicate that an SAV designed to be more anthropomorphic and to induce psychological ownership improved users' perceptions of the SAV's human-like qualities, and its responses were perceived as more positive but also more subjective compared to the control conditions. Qualitative findings support established routes to psychological ownership in the SAV context and suggest that the conversational agent's perceived performance may also influence psychological ownership. Both quantitative and qualitative outcomes highlight the importance of personalization in designing effective SAV interactions. These findings provide practical guidance for designing conversational SAV agents that enhance user experience and adoption.

Car-following models (CFMs) are fundamental to traffic flow analysis and autonomous driving. Although calibrated physics-based and trained data-driven CFMs can replicate human driving behavior, their reliance on specific datasets limits generalization across diverse scenarios and reduces reliability in real-world deployment. Moreover, these models typically focus on behavioral fidelity and do not support the explicit optimization of local and string stability, which are increasingly important for the safe and efficient operation of autonomous vehicles (AVs). To address these limitations, we propose a Knowledge-Informed Deep Learning (KIDL) paradigm that distills the generalization capabilities of pre-trained Large Language Models (LLMs) into a lightweight and stability-aware neural architecture. LLMs are used to extract fundamental car-following knowledge beyond dataset-specific patterns, and this knowledge is transferred to a reliable, tractable, and computationally efficient model through knowledge distillation. KIDL also incorporates stability constraints directly into its training objective, ensuring that the resulting model not only emulates human-like behavior but also satisfies the local and string stability requirements essential for real-world AV deployment. We evaluate KIDL on the real-world NGSIM and HighD datasets, comparing its performance with representative physics-based, data-driven, and hybrid CFMs. Both empirical and theoretical results consistently demonstrate KIDL's superior behavioral generalization and traffic flow stability, offering a robust and scalable solution for next-generation traffic systems.

In the instruction fine-tuning of large language models (LLMs), it is widely recognized that a few high-quality instructions are superior to a large number of low-quality instructions. At present, many instruction selection methods have been proposed, but most of these methods select instruction based on heuristic quality metrics, and only consider data selection before training. These designs lead to insufficient optimization of instruction fine-tuning, and fixed heuristic indicators are often difficult to optimize for specific tasks. Therefore, we design a dynamic, task-objective-driven instruction selection framework RAISE(Reinforced Adaptive Instruction SElection), which incorporates the entire instruction fine-tuning process into optimization, selecting instructions at each step based on the expected impact of each instruction on model performance improvement. Our approach is well interpretable and has strong task-specific optimization capabilities. By modeling dynamic instruction selection as a sequential decision-making process, we use RL to train our selection strategy. Extensive experiments and result analysis prove the superiority of our method compared with other instruction selection methods. Notably, RAISE achieves superior performance by updating only 1% of the training steps compared to full-data training, demonstrating its efficiency and effectiveness.