Large language model (LLM) simulations of human behavior have the potential to revolutionize the social and behavioral sciences, if and only if they faithfully reflect real human behaviors. Current evaluations are fragmented, based on bespoke tasks and metrics, creating a patchwork of incomparable results. To address this, we introduce SimBench, the first large-scale, standardized benchmark for a robust, reproducible science of LLM simulation. By unifying 20 diverse datasets covering tasks from moral decision-making to economic choice across a large global participant pool, SimBench provides the necessary foundation to ask fundamental questions about when, how, and why LLM simulations succeed or fail. We show that, while even the best LLMs today have limited simulation ability (score: 40.80/100), performance scales log-linearly with model size. Simulation performance is not improved by increased inference-time compute. We demonstrate an alignment-simulation trade-off: instruction-tuning improves performance on low-entropy (consensus) questions but degrades it on high-entropy (diverse) ones. Models particularly struggle when simulating specific demographic groups. Finally, we demonstrate that simulation ability correlates most strongly with deep, knowledge-intensive reasoning (MMLU-Pro, r=0.939). By making progress measurable, we aim to accelerate the development of more faithful LLM simulators.

Users can create diverse environments by controlling environmental conditions, add and configure custom 3D objects and execute adversarial attacks that faithfully follow threat model. Deep learning models deployed in safety critical applications like autonomous driving use simulations to test their robustness against adversarial attacks in realistic conditions. However, these simulations are non-differentiable, forcing researchers to create attacks that do not integrate simulation environmental factors, reducing attack success. To address this limitation, we introduce UnDREAM, the first software framework that bridges the gap between photorealistic simulators and differentiable renderers to enable end-to-end optimization of adversarial perturbations on any 3D objects. UnDREAM enables manipulation of the environment by offering complete control over weather, lighting, backgrounds, camera angles, trajectories, and realistic human and object movements, thereby allowing the creation of diverse scenes. We showcase a wide array of distinct physically plausible adversarial objects that UnDREAM enables researchers to swiftly explore in different configurable environments. Ensuring the adversarial robustness of vision systems is important, as computer vision is applied in safety-critical domains like autonomous vehicles.

Evaluating the safety of autonomous vehicles (AVs) requires diverse, safety-critical scenarios, with collisions being especially important yet rare and unsafe to collect in the real world. Therefore, the community has been focusing on generating safety-critical scenarios in simulation. However, controlling attributes such as collision type and time-to-accident (TTA) remains challenging. We introduce a new task called controllable collision scenario generation, where the goal is to produce trajectories that realize a user-specified collision type and TTA, to investigate the feasibility of automatically generating desired collision scenarios. To support this task, we present COLLIDE, a large-scale collision scenario dataset constructed by transforming real-world driving logs into diverse collisions, balanced across five representative collision types and different TTA intervals. We propose a framework that predicts Collision Pattern, a compact and interpretable representation that captures the spatial configuration of the ego and the adversarial vehicles at impact, before rolling out full adversarial trajectories. Experiments show that our approach outperforms strong baselines in both collision rate and controllability. Furthermore, generated scenarios consistently induce higher planner failure rates, revealing limitations of existing planners. We demonstrate that these scenarios fine-tune planners for robustness improvements, contributing to safer AV deployment in different collision scenarios. Project page is available at https://submit-user.github.io/anon2025

While attention has been empirically shown to improve model performance, it lacks a rigorous mathematical justification. This short paper establishes a novel connection between attention mechanisms and multinomial regression. Specifically, we show that in a fixed multinomial regression setting, optimizing over latent features yields solutions that align with the dynamics induced on features by attention blocks. In other words, the evolution of representations through a transformer can be interpreted as a trajectory that recovers the optimal features for classification.

Reasoning methods that adaptively allocate test-time compute have advanced LLM performance on easy to verify domains such as math and code. In this work, we study how to utilize this approach to train models that exhibit a degree of robustness to safety vulnerabilities, and show that doing so can provide benefits. We build a recipe called $\textit{TARS}$ (Training Adaptive Reasoners for Safety), a reinforcement learning (RL) approach that trains models to reason about safety using chain-of-thought traces and a reward signal that balances safety with task completion. To build TARS, we identify three critical design choices: (1) a ``lightweight'' warmstart SFT stage, (2) a mix of harmful, harmless, and ambiguous prompts to prevent shortcut behaviors such as too many refusals, and (3) a reward function to prevent degeneration of reasoning capabilities during training. Models trained with TARS exhibit adaptive behaviors by spending more compute on ambiguous queries, leading to better safety-refusal trade-offs. They also internally learn to better distinguish between safe and unsafe prompts and attain greater robustness to both white-box (e.g., GCG) and black-box attacks (e.g., PAIR). Overall, our work provides an effective, open recipe for training LLMs against jailbreaks and harmful requests by reasoning per prompt.

Ensuring responsible use of artificial intelligence (AI) has become imperative as autonomous systems increasingly influence critical societal domains. However, the concept of trustworthy AI remains broad and multi-faceted. This thesis advances knowledge in the safety, fairness, transparency, and accountability of AI systems. In safety, we extend classical deterministic shielding techniques to become resilient against delayed observations, enabling practical deployment in real-world conditions. We also implement both deterministic and probabilistic safety shields into simulated autonomous vehicles to prevent collisions with road users, validating the use of these techniques in realistic driving simulators. We introduce fairness shields, a novel post-processing approach to enforce group fairness in sequential decision-making settings over finite and periodic time horizons. By optimizing intervention costs while strictly ensuring fairness constraints, this method efficiently balances fairness with minimal interference. For transparency and accountability, we propose a formal framework for assessing intentional behaviour in probabilistic decision-making agents, introducing quantitative metrics of agency and intention quotient. We use these metrics to propose a retrospective analysis of intention, useful for determining responsibility when autonomous systems cause unintended harm. Finally, we unify these contributions through the ``reactive decision-making'' framework, providing a general formalization that consolidates previous approaches. Collectively, the advancements presented contribute practically to the realization of safer, fairer, and more accountable AI systems, laying the foundations for future research in trustworthy AI.

Quantum computing presents a promising path toward precise quantum chemical simulations, particularly for systems that challenge classical methods. This work investigates the performance of the Variational Quantum Eigensolver (VQE) in estimating the ground-state energy of the silicon atom, a relatively heavy element that poses significant computational complexity. Within a hybrid quantum-classical optimization framework, we implement VQE using a range of ansatz, including Double Excitation Gates, ParticleConservingU2, UCCSD, and k-UpCCGSD, combined with various optimizers such as gradient descent, SPSA, and ADAM. The main contribution of this work lies in a systematic methodological exploration of how these configuration choices interact to influence VQE performance, establishing a structured benchmark for selecting optimal settings in quantum chemical simulations. Key findings show that parameter initialization plays a decisive role in the algorithm's stability, and that the combination of a chemically inspired ansatz with adaptive optimization yields superior convergence and precision compared to conventional approaches.

Soil salinization poses a significant threat to both ecosystems and agriculture because it limits plants' ability to absorb water and, in doing so, reduces crop productivity. This phenomenon alters the soil's spectral properties, creating a measurable relationship between salinity and light reflectance that enables remote monitoring. While laboratory spectroscopy provides precise measurements, its reliance on in-situ sampling limits scalability to regional or global levels. Conversely, hyperspectral satellite imagery enables wide-area observation but lacks the fine-grained interpretability of laboratory instruments. To bridge this gap, we introduce DeepSalt, a deep-learning-based spectral transfer framework that leverages knowledge distillation and a novel Spectral Adaptation Unit to transfer high-resolution spectral insights from laboratory-based spectroscopy to satellite-based hyperspectral sensing. Our approach eliminates the need for extensive ground sampling while enabling accurate, large-scale salinity estimation, as demonstrated through comprehensive empirical benchmarks. DeepSalt achieves significant performance gains over methods without explicit domain adaptation, underscoring the impact of the proposed Spectral Adaptation Unit and the knowledge distillation strategy. The model also effectively generalized to unseen geographic regions, explaining a substantial portion of the salinity variance.

Multilingual Retrieval-Augmented Generation (mRAG) often retrieves English documents and translates them into the query language for low-resource settings. However, poor translation quality degrades response generation performance. Existing approaches either assume sufficient translation quality or utilize the rewriting method, which introduces factual distortion and hallucinations. To mitigate these problems, we propose Quality-Aware Translation Tagging in mRAG (QTT-RAG), which explicitly evaluates translation quality along three dimensions-semantic equivalence, grammatical accuracy, and naturalness&fluency-and attach these scores as metadata without altering the original content. We evaluate QTT-RAG against CrossRAG and DKM-RAG as baselines in two open-domain QA benchmarks (XORQA, MKQA) using six instruction-tuned LLMs ranging from 2.4B to 14B parameters, covering two low-resource languages (Korean and Finnish) and one high-resource language (Chinese). QTT-RAG outperforms the baselines by preserving factual integrity while enabling generator models to make informed decisions based on translation reliability. This approach allows for effective usage of cross-lingual documents in low-resource settings with limited native language documents, offering a practical and robust solution across multilingual domains.

Recent advances in generative modeling have shown significant promise in designing novel periodic crystal structures. Existing approaches typically rely on either large language models (LLMs) or equivariant denoising models, each with complementary strengths: LLMs excel at handling discrete atomic types but often struggle with continuous features such as atomic positions and lattice parameters, while denoising models are effective at modeling continuous variables but encounter difficulties in generating accurate atomic compositions. To bridge this gap, we propose CrysLLMGen, a hybrid framework that integrates an LLM with a diffusion model to leverage their complementary strengths for crystal material generation. During sampling, CrysLLMGen first employs a fine-tuned LLM to produce an intermediate representation of atom types, atomic coordinates, and lattice structure. While retaining the predicted atom types, it passes the atomic coordinates and lattice structure to a pre-trained equivariant diffusion model for refinement. Our framework outperforms state-of-the-art generative models across several benchmark tasks and datasets. Specifically, CrysLLMGen not only achieves a balanced performance in terms of structural and compositional validity but also generates more stable and novel materials compared to LLM-based and denoisingbased models Furthermore, CrysLLMGen exhibits strong conditional generation capabilities, effectively producing materials that satisfy user-defined constraints. Code is available at https://github.com/kdmsit/crysllmgen