Achieving robust cognitive autonomy in robots navigating complex, unpredictable environments remains a fundamental challenge in robotics. This paper presents Underwater Robot Self-Organizing Autonomy (UROSA), a groundbreaking architecture leveraging distributed Large Language Model AI agents integrated within the Robot Operating System 2 (ROS 2) framework to enable advanced cognitive capabilities in Autonomous Underwater Vehicles. UROSA decentralises cognition into specialised AI agents responsible for multimodal perception, adaptive reasoning, dynamic mission planning, and real-time decision-making. Central innovations include flexible agents dynamically adapting their roles, retrieval-augmented generation utilising vector databases for efficient knowledge management, reinforcement learning-driven behavioural optimisation, and autonomous on-the-fly ROS 2 node generation for runtime functional extensibility. Extensive empirical validation demonstrates UROSA's promising adaptability and reliability through realistic underwater missions in simulation and real-world deployments, showing significant advantages over traditional rule-based architectures in handling unforeseen scenarios, environmental uncertainties, and novel mission objectives. This work not only advances underwater autonomy but also establishes a scalable, safe, and versatile cognitive robotics framework capable of generalising to a diverse array of real-world applications.

Large Language Models (LLMs) can achieve inflated scores on multiple-choice tasks by exploiting inherent biases in option positions or labels, rather than demonstrating genuine understanding. This study introduces SCOPE, an evaluation framework designed to measure and mitigate such selection bias in a dataset-independent manner. By repeatedly invoking a null prompt that lacks semantic content, SCOPE estimates each model's unique position-bias distribution. It then redistributes the answer slot according to the inverse-bias distribution, thereby equalizing the lucky-rate, the probability of selecting the correct answer by chance. Furthermore, it prevents semantically similar distractors from being placed adjacent to the answer, thereby blocking near-miss guesses based on superficial proximity cues. Across multiple benchmark experiments, SCOPE consistently outperformed existing debiasing methods in terms of stable performance improvements and showed clearer confidence distributions over correct options. This framework thus offers a new standard for enhancing the fairness and reliability of LLM evaluations.

Knowledge distillation (KD) is a key technique for compressing large language models into smaller ones while preserving performance. Despite the recent traction of KD research, its effectiveness for smaller language models (LMs) and the mechanisms driving knowledge transfer remain underexplored. In this work, we present the first large-scale empirical and statistical analysis of KD across models ranging from 0.5B to 7B parameters on 14 complex reasoning tasks in a zero-shot setting. Our findings reveal that KD can improve the average performance of smaller models by up to $10\%$, with a peak task specific gain of $22\%$, while providing only marginal benefits ($\sim 1.3\%$) for larger models. Surprisingly, teacher performance has a minimal impact on student outcomes, while teacher task expertise impacts KD effectiveness. A correlation study indicates that smaller LMs benefit more from KD, whereas larger LMs show diminished gains. Additionally, we uncover a misalignment between improvements in student performance and reasoning fidelity, suggesting that while KD enhances accuracy, it does not always maintain the structured decision-making processes of the teacher. Our ablation study further highlights the importance of teacher signals and logit smoothing in influencing students' performance after distillation. Overall, our study offers a comprehensive empirical and statistical assessment of KD, highlighting both its benefits and trade-offs when distilling knowledge from larger to smaller LMs.

Effective feedback is essential for refining instructional practices in mathematics education, and researchers often turn to advanced natural language processing (NLP) models to analyze classroom dialogues from multiple perspectives. However, utterance-level discourse analysis encounters two primary challenges: (1) multifunctionality, where a single utterance may serve multiple purposes that a single tag cannot capture, and (2) the exclusion of many utterances from domain-specific discourse move classifications, leading to their omission in feedback. To address these challenges, we proposed a multi-perspective discourse analysis that integrates domain-specific talk moves with dialogue act (using the flattened multi-functional SWBD-MASL schema with 43 tags) and discourse relation (applying Segmented Discourse Representation Theory with 16 relations). Our top-down analysis framework enables a comprehensive understanding of utterances that contain talk moves, as well as utterances that do not contain talk moves. This is applied to two mathematics education datasets: TalkMoves (teaching) and SAGA22 (tutoring). Through distributional unigram analysis, sequential talk move analysis, and multi-view deep dive, we discovered meaningful discourse patterns, and revealed the vital role of utterances without talk moves, demonstrating that these utterances, far from being mere fillers, serve crucial functions in guiding, acknowledging, and structuring classroom discourse. These insights underscore the importance of incorporating discourse relations and dialogue acts into AI-assisted education systems to enhance feedback and create more responsive learning environments. Our framework may prove helpful for providing human educator feedback, but also aiding in the development of AI agents that can effectively emulate the roles of both educators and students.

Training large language models (LLMs) typically involves pre-training on massive corpora, only to restart the process entirely when new data becomes available. A more efficient and resource-conserving approach would be continual pre-training, where models are updated with new data rather than retraining from scratch. However, the introduction of new data often causes distribution shifts, leading to performance degradation on previously learned tasks. In this paper, we take a deeper look at two popular proposals for addressing this distribution shift within the continual learning literature: experience replay and gradient alignment. We consider continual pre-training of models within the Llama family of architectures at a large scale across languages with 100 billion tokens of training data in each language, finding that both replay and gradient alignment lead to more stable learning without forgetting. This conclusion holds both as we vary the model scale and as we vary the number and diversity of tasks. Moreover, we are the first to demonstrate the effectiveness of gradient alignment techniques in the context of LLM pre-training and propose an efficient implementation of meta-experience replay (MER) (Riemer et al., 2019a) that imbues experience replay with the benefits of gradient alignment despite negligible compute and memory overhead. Our scaling analysis across model sizes and replay rates indicates that small rates of replaying old examples are definitely a more valuable use of compute than investing in model size, but that it is more compute efficient to scale the size of the model than invest in high rates of replaying old examples. Large Language Models (LLMs) need regular updates to be current with new information and domains, posing a problem for organizations looking to maintain LLMs without repeatedly performing expensive retraining from scratch. Performing updates to a model that has already received pre-training on a new distribution is the classic problem of continual learning (Ring, 1994) or lifelong learning (Thrun, 1994). We should draw a strong distinction between this setting and other settings such as fine-tuning or instruction tuning, which are generally characterized by training on much smaller datasets for a much smaller number of gradient steps.

Jim Acosta and James Carville speculated whether President Trump will try to rig the 2026 midterms in his favor on "The Jim Acosta Show." Liberal journalist Jim Acosta "interviewed" the artificially animated avatar of deceased teenager Joaquin Oliver to promote a gun control message on Monday. Working with the gun control group Change the Ref, founded by Oliver's parents, Acosta had conversation on his Substack with an avatar created by the father of the son, who was killed in the Parkland high school shooting in 2018. He would have turned 25 on Monday. "I would like to know what your solution would be for gun violence," Acosta asked.

A searchable database of public records concerning use of force and misconduct by California law enforcement officers -- some 1.5 million pages from nearly 700 law enforcement agencies -- is now available to the public. The Police Records Access Project, a database built by UC Berkeley and Stanford University, is being published by the Los Angeles Times, San Francisco Chronicle, KQED and CalMatters. It will vastly expand public access to internal affairs records that show how law enforcement agencies throughout the state handle misconduct allegations and uses of police force that result in death or serious injury. The database currently includes records from nearly 12,000 cases. The database is the product of years of work by a multidisciplinary team of journalists, data scientists, lawyers and civil liberties advocates, led by the Berkeley Institute for Data Science (BIDS), UC Berkeley Journalism's Investigative Reporting Program (IRP) and Stanford University's Big Local News.

Recent advances in natural language processing (NLP) have opened up greater opportunities to enable fine-tuned large language models (LLMs) to behave as more powerful interactive agents through improved instruction-following ability. However, understanding how this impacts confidence calibration for reliable model output has not been researched in full. In this work, we examine various open-sourced LLMs, identifying significant calibration degradation after instruction tuning in each. Seeking a practical solution, we look towards label smoothing, which has been shown as an effective method to regularize for overconfident predictions but has yet to be widely adopted in the supervised fine-tuning (SFT) of LLMs. We first provide insight as to why label smoothing is sufficient to maintain calibration throughout the SFT process. However, settings remain where the effectiveness of smoothing is severely diminished, in particular the case of large vocabulary LLMs (LV-LLMs). We posit the cause to stem from the ability to become over-confident, which has a direct relationship with the hidden size and vocabulary size, and justify this theoretically and experimentally. Finally, we address an outstanding issue regarding the memory footprint of the cross-entropy loss computation in the label smoothed loss setting, designing a customized kernel to dramatically reduce memory consumption without sacrificing speed or performance in comparison to existing solutions for non-smoothed losses.

Stochasticity in language model fine-tuning, often caused by the small batch sizes typically used in this regime, can destabilize training by introducing large oscillations in generation quality. A popular approach to mitigating this instability is to take an Exponential moving average (EMA) of weights throughout training. While EMA reduces stochasticity, thereby smoothing training, the introduction of bias from old iterates often creates a lag in optimization relative to vanilla training. In this work, we propose the Bias-Corrected Exponential Moving Average (BEMA), a simple and practical augmentation of EMA that retains variance-reduction benefits while eliminating bias. BEMA is motivated by a simple theoretical model wherein we demonstrate provable acceleration of BEMA over both a standard EMA and vanilla training. Through an extensive suite of experiments on Language Models, we show that BEMA leads to significantly improved convergence rates and final performance over both EMA and vanilla training in a variety of standard LM benchmarks, making BEMA a practical and theoretically motivated intervention for more stable and efficient fine-tuning.

Existing AIG (AI-generated) text detectors struggle in real-world settings despite succeeding in internal testing, suggesting that they may not be robust enough. We rigorously examine the machine-learning procedure to build these detectors to address this. Most current AIG text detection datasets focus on zero-shot generations, but little work has been done on few-shot or one-shot generations, where LLMs are given human texts as an example. In response, we introduce the Diverse Adversarial Corpus of Texts Yielded from Language models (DACTYL), a challenging AIG text detection dataset focusing on one-shot/few-shot generations. We also include texts from domain-specific continued-pre-trained (CPT) language models, where we fully train all parameters using a memory-efficient optimization approach. Many existing AIG text detectors struggle significantly on our dataset, indicating a potential vulnerability to one-shot/few-shot and CPT-generated texts. We also train our own classifiers using two approaches: standard binary cross-entropy (BCE) optimization and a more recent approach, deep X-risk optimization (DXO). While BCE-trained classifiers marginally outperform DXO classifiers on the DACTYL test set, the latter excels on out-of-distribution (OOD) texts. In our mock deployment scenario in student essay detection with an OOD student essay dataset, the best DXO classifier outscored the best BCE-trained classifier by 50.56 macro-F1 score points at the lowest false positive rates for both. Our results indicate that DXO classifiers generalize better without overfitting to the test set. Our experiments highlight several areas of improvement for AIG text detectors.