This research presents a novel approach to computational framing analysis, called Semantic Relations-based Unsupervised Framing Analysis (SUFA). SUFA leverages semantic relations and dependency parsing algorithms to identify and assess entity-centric emphasis frames in news media reports. This innovative method is derived from two studies -- qualitative and computational -- using a dataset related to gun violence, demonstrating its potential for analyzing entity-centric emphasis frames. This article discusses SUFA's strengths, limitations, and application procedures. Overall, the SUFA approach offers a significant methodological advancement in computational framing analysis, with its broad applicability across both the social sciences and computational domains.

Moonbeam is a transformer-based foundation model for symbolic music, pretrained on a large and diverse collection of MIDI data totaling 81.6K hours of music and 18 billion tokens. Moonbeam incorporates music-domain inductive biases by capturing both absolute and relative musical attributes through the introduction of a novel domain-knowledge-inspired tokenization method and Multidimensional Relative Attention (MRA), which captures relative music information without additional trainable parameters. Leveraging the pretrained Moonbeam, we propose 2 finetuning architectures with full anticipatory capabilities, targeting 2 categories of downstream tasks: symbolic music understanding and conditional music generation (including music infilling). Our model outperforms other large-scale pretrained music models in most cases in terms of accuracy and F1 score across 3 downstream music classification tasks on 4 datasets. Moreover, our finetuned conditional music generation model outperforms a strong transformer baseline with a REMI-like tokenizer. We open-source the code, pretrained model, and generated samples on Github.

The rise of Large Audio Language Models (LAMs) brings both potential and risks, as their audio outputs may contain harmful or unethical content. However, current research lacks a systematic, quantitative evaluation of LAM safety especially against jailbreak attacks, which are challenging due to the temporal and semantic nature of speech. To bridge this gap, we introduce AJailBench, the first benchmark specifically designed to evaluate jailbreak vulnerabilities in LAMs. We begin by constructing AJailBench-Base, a dataset of 1,495 adversarial audio prompts spanning 10 policy-violating categories, converted from textual jailbreak attacks using realistic text to speech synthesis. Using this dataset, we evaluate several state-of-the-art LAMs and reveal that none exhibit consistent robustness across attacks. To further strengthen jailbreak testing and simulate more realistic attack conditions, we propose a method to generate dynamic adversarial variants. Our Audio Perturbation Toolkit (APT) applies targeted distortions across time, frequency, and amplitude domains. To preserve the original jailbreak intent, we enforce a semantic consistency constraint and employ Bayesian optimization to efficiently search for perturbations that are both subtle and highly effective. This results in AJailBench-APT, an extended dataset of optimized adversarial audio samples. Our findings demonstrate that even small, semantically preserved perturbations can significantly reduce the safety performance of leading LAMs, underscoring the need for more robust and semantically aware defense mechanisms.

Large Reasoning Models (LRMs) have achieved remarkable success on reasoning-intensive tasks such as mathematics and programming. However, their enhanced reasoning capabilities do not necessarily translate to improved safety performance-and in some cases, may even degrade it. This raises an important research question: how can we enhance the safety of LRMs? In this paper, we present a comprehensive empirical study on how to enhance the safety of LRMs through Supervised Fine-Tuning (SFT). Our investigation begins with an unexpected observation: directly distilling safe responses from DeepSeek-R1 fails to significantly enhance safety. We analyze this phenomenon and identify three key failure patterns that contribute to it. We then demonstrate that explicitly addressing these issues during the data distillation process can lead to substantial safety improvements. Next, we explore whether a long and complex reasoning process is necessary for achieving safety. Interestingly, we find that simply using short or template-based reasoning process can attain comparable safety performance-and are significantly easier for models to learn than more intricate reasoning chains. These findings prompt a deeper reflection on the role of reasoning in ensuring safety. Finally, we find that mixing math reasoning data during safety fine-tuning is helpful to balance safety and over-refusal. Overall, we hope our empirical study could provide a more holistic picture on enhancing the safety of LRMs. The code and data used in our experiments are released in https://github.com/thu-coai/LRM-Safety-Study.

Reinforcement learning (RL) has demonstrated strong potential in training large language models (LLMs) capable of complex reasoning for real-world problem solving. More recently, RL has been leveraged to create sophisticated LLM-based search agents that adeptly combine reasoning with search engine use. While the use of RL for training search agents is promising, the optimal design of such agents remains not fully understood. In particular, key factors -- such as (1) reward formulation, (2) the choice and characteristics of the underlying LLM, and (3) the role of the search engine in the RL process -- require further investigation. In this work, we conduct comprehensive empirical studies to systematically investigate these and offer actionable insights. We highlight several key findings: format rewards are effective in improving final performance, whereas intermediate retrieval rewards have limited impact; the scale and initialization of the LLM (general-purpose vs. reasoning-specialized) significantly influence RL outcomes; and the choice of search engine plays a critical role in shaping RL training dynamics and the robustness of the trained agent during inference. These establish important guidelines for successfully building and deploying LLM-based search agents in real-world applications. Code is available at https://github.com/PeterGriffinJin/Search-R1.

Motivated by teleoperation applications in cluttered environments with limited computational power, we propose a local planner that does not require the knowledge or construction of a global map of the obstacles. The proposed solution consists of a real-time trajectory planning algorithm that relies on the histogram of obstacle distribution and a planner manager that triggers different planning modes depending on obstacles location around the MA V . The proposed solution is validated, for a teleoperation application, with both simulations and indoor experiments. Benchmark comparisons based on a designed simulation platform are also provided. I. INTRODUCTION Micro aerial vehicles (MA Vs) are used in many applications, such as rescue search, forestry monitoring, infrastructure maintenance, aerial photography, etc. When the MA V operates in cluttered environments, obstacle avoidance is a major problem. Solutions to this problem are highly dependent on the type of environment, the available onboard sensors, the availability of a global map of the environment, and the available computational power. While solutions to this problem rely on both perception and planning/navigation aspects (the classical sense and avoid scenario), the present paper focuses on the navigation aspect. Many traditional navigation methods are summarized in detail in [1].

Consider this prompt "Draw a unicorn with two horns". Should large language models (LLMs) recognize that a unicorn has only one horn by definition and ask users for clarifications, or proceed to generate something anyway? We introduce concept incongruence to capture such phenomena where concept boundaries clash with each other, either in user prompts or in model representations, often leading to under-specified or mis-specified behaviors. In this work, we take the first step towards defining and analyzing model behavior under concept incongruence. Focusing on temporal boundaries in the Role-Play setting, we propose three behavioral metrics--abstention rate, conditional accuracy, and answer rate--to quantify model behavior under incongruence due to the role's death. We show that models fail to abstain after death and suffer from an accuracy drop compared to the Non-Role-Play setting. Through probing experiments, we identify two main causes: (i) unreliable encoding of the "death" state across different years, leading to unsatisfactory abstention behavior, and (ii) role playing causes shifts in the model's temporal representations, resulting in accuracy drops. We leverage these insights to improve consistency in the model's abstention and answer behaviors. Our findings suggest that concept incongruence leads to unexpected model behaviors and point to future directions on improving model behavior under concept incongruence.

Robustly evaluating the long-form storytelling capabilities of Large Language Models (LLMs) remains a significant challenge, as existing benchmarks often lack the necessary scale, diversity, or objective measures. To address this, we introduce WebNovelBench, a novel benchmark specifically designed for evaluating long-form novel generation. WebNovelBench leverages a large-scale dataset of over 4,000 Chinese web novels, framing evaluation as a synopsis-to-story generation task. We propose a multi-faceted framework encompassing eight narrative quality dimensions, assessed automatically via an LLM-as-Judge approach. Scores are aggregated using Principal Component Analysis and mapped to a percentile rank against human-authored works. Our experiments demonstrate that WebNovelBench effectively differentiates between human-written masterpieces, popular web novels, and LLM-generated content. We provide a comprehensive analysis of 24 state-of-the-art LLMs, ranking their storytelling abilities and offering insights for future development. This benchmark provides a scalable, replicable, and data-driven methodology for assessing and advancing LLM-driven narrative generation.

CNN host Jake Tapper told Joe Scarborough during a Wednesday conversation on "Morning Joe" that former President Biden made an effort to convince the MSNBC host that he was fit to run for re-election. The Washington Post editorial board called for more oversight of the Oval Office on Wednesday to ensure a cover-up of the president's health doesn't happen again following revelations in a bombshell book alleging the White House hid former President Joe Biden's decline from the public. "It now seems that, for a considerable time, Biden might have lacked the stamina and cognitive capacity the job demands -- and that his family and closest aides concealed this from the public," the paper's editorial board wrote. "Their apparent decision to put personal loyalties ahead of their duty to the country must be reckoned with. A legal mechanism should be considered to ensure that this doesn't happen again," the board proposed.

Although the Cannes Film Festival is the world's most prestigious movie showcase, its spotlight rarely falls on nonfiction film. Years go by without a single documentary competing for its biggest honor, the Palme d'Or, and there is no separate documentary prize. Juliette Binoche, the president of this year's jury, devoted part of her opening-night remarks to Fatma Hassona, the Palestinian photojournalist who was killed in an Israeli airstrike the day after it was announced that her documentary Put Your Soul on Your Hand and Walk would be premiering at Cannes. But the film itself was slotted into a low-profile sidebar devoted to independent productions. The festival did, however, roll out the red carpet for The Six Billion Dollar Man, Eugene Jarecki's portrait of WikiLeaks founder Julian Assange, which premiered out of competition on Wednesday evening.