The rapid evolution of large language models (LLMs) has pushed their boundaries to many applications in various domains. Recently, the research community has started to evaluate their potential adoption in autonomous vehicles and especially as complementary modules in the perception and planning software stacks. However, their evaluation is limited in synthetic datasets or manually driving datasets without the ground truth knowledge and more precisely, how the current perception and planning algorithms would perform in the cases under evaluation. F or this reason, this work evaluates LLMs on real-world edge cases where current autonomous vehicles have been proven to fail. The proposed architecture consists of an open vocabulary object detector coupled with prompt engineering and large language model contextual reasoning. W e evaluate several state-of-the-art models against real edge cases and provide qualitative comparison results along with a discussion on the findings for the potential application of LLMs as anomaly detectors in autonomous vehicles.

Reinforcement Learning (RL) has shown great potential for autonomous decision-making in the cybersecurity domain, enabling agents to learn through direct environment interaction. However, RL agents in Autonomous Cyber Operations (ACO) typically learn from scratch, requiring them to execute undesirable actions to learn their consequences. In this study, we integrate external knowledge in the form of a Large Language Model (LLM) pretrained on cybersecurity data that our RL agent can directly leverage to make informed decisions. By guiding initial training with an LLM, we improve baseline performance and reduce the need for exploratory actions with obviously negative outcomes. We evaluate our LLM-integrated approach in a simulated cybersecurity environment, and demonstrate that our guided agent achieves over 2x higher rewards during early training and converges to a favorable policy approximately 4,500 episodes faster than the baseline.

Infrastructure as Code (IaC) is fundamental to modern cloud computing, enabling teams to define and manage infrastructure through machine-readable configuration files. However, different cloud service providers utilize diverse IaC formats. The lack of a standardized format requires cloud architects to be proficient in multiple IaC languages, adding complexity to cloud deployment. While Large Language Models (LLMs) show promise in automating IaC creation and maintenance, progress has been limited by the lack of comprehensive benchmarks across multiple IaC formats. We present Multi-IaC-Bench, a novel benchmark dataset for evaluating LLM-based IaC generation and mutation across AWS CloudFormation, Terraform, and Cloud Development Kit (CDK) formats. The dataset consists of triplets containing initial IaC templates, natural language modification requests, and corresponding updated templates, created through a synthetic data generation pipeline with rigorous validation. We evaluate several state-of-the-art LLMs on Multi-IaC-Bench, demonstrating that while modern LLMs can achieve high success rates (>95%) in generating syntactically valid IaC across formats, significant challenges remain in semantic alignment and handling complex infrastructure patterns. Our ablation studies highlight the importance of prompt engineering and retry mechanisms in successful IaC generation. We release Multi-IaC-Bench to facilitate further research in AI-assisted infrastructure management and establish standardized evaluation metrics for this crucial domain.

Recent research has been increasingly focusing on developing 3D world models that simulate complex real-world scenarios. World models have found broad applications across various domains, including embodied AI, autonomous driving, entertainment, etc. A more realistic simulation with accurate physics will effectively narrow the sim-to-real gap and allow us to gather rich information about the real world conveniently. While traditional manual modeling has enabled the creation of virtual 3D scenes, modern approaches have leveraged advanced machine learning algorithms for 3D world generation, with most recent advances focusing on generative methods that can create virtual worlds based on user instructions. This work explores such a research direction by proposing LatticeWorld, a simple yet effective 3D world generation framework that streamlines the industrial production pipeline of 3D environments. LatticeWorld leverages lightweight LLMs (LLaMA-2-7B) alongside the industry-grade rendering engine (e.g., Unreal Engine 5) to generate a dynamic environment. Our proposed framework accepts textual descriptions and visual instructions as multimodal inputs and creates large-scale 3D interactive worlds with dynamic agents, featuring competitive multi-agent interaction, high-fidelity physics simulation, and real-time rendering. We conduct comprehensive experiments to evaluate LatticeWorld, showing that it achieves superior accuracy in scene layout generation and visual fidelity. Moreover, LatticeWorld achieves over a $90\times$ increase in industrial production efficiency while maintaining high creative quality compared with traditional manual production methods. Our demo video is available at https://youtu.be/8VWZXpERR18

Positional encoding mechanisms enable Transformers to model sequential structure and long-range dependencies in text. While absolute positional encodings struggle with extrapolation to longer sequences due to fixed positional representations, and relative approaches like Alibi exhibit performance degradation on extremely long contexts, the widely-used Rotary Positional Encoding (RoPE) introduces oscillatory attention patterns that hinder stable long-distance dependency modelling. We address these limitations through a geometric reformulation of positional encoding. Drawing inspiration from Lorentz transformations in hyperbolic geometry, we propose Hyperbolic Rotary Positional Encoding (HoPE), which leverages hyperbolic functions to implement Lorentz rotations on token representations. Theoretical analysis demonstrates that RoPE is a special case of our generalized formulation. HoPE fundamentally resolves RoPE's slation issues by enforcing monotonic decay of attention weights with increasing token distances. Extensive experimental results, including perplexity evaluations under several extended sequence benchmarks, show that HoPE consistently exceeds existing positional encoding methods. These findings underscore HoPE's enhanced capacity for representing and generalizing long-range dependencies. Data and code will be available.

Large reasoning models (LRMs) have exhibited strong performance on complex reasoning tasks, with further gains achievable through increased computational budgets at inference. However, current test-time scaling methods predominantly rely on redundant sampling, ignoring the historical experience utilization, thereby limiting computational efficiency. To overcome this limitation, we propose Sticker-TTS, a novel test-time scaling framework that coordinates three collaborative LRMs to iteratively explore and refine solutions guided by historical attempts. At the core of our framework are distilled key conditions-termed stickers-which drive the extraction, refinement, and reuse of critical information across multiple rounds of reasoning. To further enhance the efficiency and performance of our framework, we introduce a two-stage optimization strategy that combines imitation learning with self-improvement, enabling progressive refinement. Extensive evaluations on three challenging mathematical reasoning benchmarks, including AIME-24, AIME-25, and OlymMATH, demonstrate that Sticker-TTS consistently surpasses strong baselines, including self-consistency and advanced reinforcement learning approaches, under comparable inference budgets. These results highlight the effectiveness of sticker-guided historical experience utilization. Our code and data are available at https://github.com/RUCAIBox/Sticker-TTS.

Twitter and other social media platforms have become vital sources of real time information during disasters and public safety emergencies. Automatically classifying disaster related tweets can help emergency services respond faster and more effectively. Traditional Machine Learning (ML) models such as Logistic Regression, Naive Bayes, and Support Vector Machines have been widely used for this task, but they often fail to understand the context or deeper meaning of words, especially when the language is informal, metaphorical, or ambiguous. We posit that, in this context, transformer based models can perform better than traditional ML models. In this paper, we evaluate the effectiveness of transformer based models, including BERT, DistilBERT, RoBERTa, and DeBERTa, for classifying disaster related tweets. These models are compared with traditional ML approaches to highlight the performance gap. Experimental results show that BERT achieved the highest accuracy (91%), significantly outperforming traditional models like Logistic Regression and Naive Bayes (both at 82%). The use of contextual embeddings and attention mechanisms allows transformer models to better understand subtle language in tweets, where traditional ML models fall short. This research demonstrates that transformer architectures are far more suitable for public safety applications, offering improved accuracy, deeper language understanding, and better generalization across real world social media text.

Reliable abstention is critical for retrieval-augmented generation (RAG) systems, particularly in safety-critical domains such as women's health, where incorrect answers can lead to harm. We present an energy-based model (EBM) that learns a smooth energy landscape over a dense semantic corpus of 2.6M guideline-derived questions, enabling the system to decide when to generate or abstain. We benchmark the EBM against a calibrated softmax baseline and a k-nearest neighbour (kNN) density heuristic across both easy and hard abstention splits, where hard cases are semantically challenging near-distribution queries. The EBM achieves superior abstention performance abstention on semantically hard cases, reaching AUROC 0.961 versus 0.950 for softmax, while also reducing FPR@95 (0.235 vs 0.331). On easy negatives, performance is comparable across methods, but the EBM's advantage becomes most pronounced in safety-critical hard distributions. A comprehensive ablation with controlled negative sampling and fair data exposure shows that robustness stems primarily from the energy scoring head, while the inclusion or exclusion of specific negative types (hard, easy, mixed) sharpens decision boundaries but is not essential for generalisation to hard cases. These results demonstrate that energy-based abstention scoring offers a more reliable confidence signal than probability-based softmax confidence, providing a scalable and interpretable foundation for safe RAG systems.

Large Language Models (LLMs) have exhibited significant proficiency in code debugging, especially in automatic program repair, which may substantially reduce the time consumption of developers and enhance their efficiency. Significant advancements in debugging datasets have been made to promote the development of code debugging. However, these datasets primarily focus on assessing the LLM's function-level code repair capabilities, neglecting the more complex and realistic repository-level scenarios, which leads to an incomplete understanding of the LLM's challenges in repository-level debugging. While several repository-level datasets have been proposed, they often suffer from limitations such as limited diversity of tasks, languages, and error types. To mitigate this challenge, this paper introduces RepoDebug, a multi-task and multi-language repository-level code debugging dataset with 22 subtypes of errors that supports 8 commonly used programming languages and 3 debugging tasks. Furthermore, we conduct evaluation experiments on 10 LLMs, where Claude 3.5 Sonnect, the best-performing model, still cannot perform well in repository-level debugging.

Large language model (LLM) agents achieve impressive single-task performance but commonly exhibit repeated failures, inefficient exploration, and limited cross-task adaptability. Existing reflective strategies (e.g., Reflexion, ReAct) improve per-episode behavior but typically produce ephemeral, task-specific traces that are not reused across tasks. Reinforcement-learning based alternatives can produce transferable policies but require substantial parameter updates and compute. In this work we introduce Meta-Policy Reflexion (MPR): a hybrid framework that consolidates LLM-generated reflections into a structured, predicate-like Meta-Policy Memory (MPM) and applies that memory at inference time through two complementary mechanisms soft memory-guided decoding and hard rule admissibility checks(HAC). MPR (i) externalizes reusable corrective knowledge without model weight updates, (ii) enforces domain constraints to reduce unsafe or invalid actions, and (iii) retains the adaptability of language-based reflection. We formalize the MPM representation, present algorithms for update and decoding, and validate the approach in a text-based agent environment following the experimental protocol described in the provided implementation (AlfWorld-based). Empirical results reported in the supplied material indicate consistent gains in execution accuracy and robustness when compared to Reflexion baselines; rule admissibility further improves stability. We analyze mechanisms that explain these gains, discuss scalability and failure modes, and outline future directions for multimodal and multi-agent extensions.