Evaluating open-ended responses from large audio language models (LALMs) is challenging because human annotators often genuinely disagree on answer correctness due to multiple valid interpretations, partial correctness, and subjective judgment. Traditional metrics reporting only mean scores fail to capture this uncertainty. We present ORCA (Open-ended Response Correctness Assessment), a framework that models the variability in human judgments using Beta distributions to predict both expected correctness and uncertainty. Our three-stage annotation framework combines human judgment with structured feedback and iterative refinement to simultaneously curate training data and improve benchmark quality. We collected 11,721 annotations across 3,580 question-answer pairs from 15 LALMs on two audio QA benchmarks, achieving inter-annotator agreement of 0.82 (Krippendorff's alpha). ORCA achieves 0.91 Spearman correlation with mean human judgments, matching or outperforming LLM-judge baselines while providing uncertainty estimates and requiring significantly less compute. We release our models, code, and curated dataset.

Perch 2.0 is a supervised bioacoustics foundation model pretrained on 14,597 species, including birds, mammals, amphibians, and insects, and has state-of-the-art performance on multiple benchmarks. Given that Perch 2.0 includes almost no marine mammal audio or classes in the training data, we evaluate Perch 2.0 performance on marine mammal and underwater audio tasks through few-shot transfer learning. We perform linear probing with the embeddings generated from this foundation model and compare performance to other pretrained bioacoustics models. In particular, we compare Perch 2.0 with previous multispecies whale, Perch 1.0, SurfPerch, AVES-bio, BirdAVES, and Birdnet V2.3 models, which have open-source tools for transfer-learning and agile modeling. We show that the embeddings from the Perch 2.0 model have consistently high performance for few-shot transfer learning, generally outperforming alternative embedding models on the majority of tasks, and thus is recommended when developing new linear classifiers for marine mammal classification with few labeled examples.

Moctezuma's pod continues their dominance in the Gulf of California. Breakthroughs, discoveries, and DIY tips sent every weekday. Orca whales are skilled pack hunters with an ever-growing list of prey . Recently, ocean researchers discovered that the apex predators aren't afraid of taking on equally formidable foes-- great white sharks . Now, a study published on November 3 in the journal documented even more remarkable hunting behavior.

Large language models have shown great success on natural language tasks in recent years, but they have also shown great promise when adapted to new modalities, e.g., for scientific machine learning tasks. Even though decoder-only models are more popular within NLP and scale exceedingly well at generating natural language, most proposed approaches for cross-modal adaptation focus on encoder-only models, raising the question of how model architecture affects these approaches. In this paper, we therefore perform a series of ablation studies to answer this question, systematically comparing encoder-only and decoder-only models on cross-modal adaptation for time-dependent simulation tasks based on partial differential equations (PDEs). We find that decoder-only models are far worse than encoder-only models, when existing approaches are applied unmodified. In contrast to several other domains, scaling decoder-only models also does not help. To harness the potential of decoder-only models in this context, we introduce two novel approaches, Parallel Flipping and Sequence Doubling, attempting to mimic bidirectionality in autoregressive models. Both our methods improve overall performance using decoder-only models for all tasks and all cross-model adaptation methods, closing the gap to encoder-only model performance. We hope that our findings broaden the spectrum of models used on cross-modal adaptation tasks to further scientific ML.

Large Vision-Language Models (LVLMs) exhibit strong multimodal capabilities but remain vulnerable to hallucinations from intrinsic errors and adversarial attacks from external exploitations, limiting their reliability in real-world applications. We present ORCA, an agentic reasoning framework that improves the factual accuracy and adversarial robustness of pretrained LVLMs through test-time structured inference reasoning with a suite of small vision models (less than 3B parameters). ORCA operates via an Observe--Reason--Critique--Act loop, querying multiple visual tools with evidential questions, validating cross-model inconsistencies, and refining predictions iteratively without access to model internals or retraining. ORCA also stores intermediate reasoning traces, which supports auditable decision-making. Though designed primarily to mitigate object-level hallucinations, ORCA also exhibits emergent adversarial robustness without requiring adversarial training or defense mechanisms. We evaluate ORCA across three settings: (1) clean images on hallucination benchmarks, (2) adversarially perturbed images without defense, and (3) adversarially perturbed images with defense applied. On the POPE hallucination benchmark, ORCA improves standalone LVLM performance by +3.64\% to +40.67\% across different subsets. Under adversarial perturbations on POPE, ORCA achieves an average accuracy gain of +20.11\% across LVLMs. When combined with defense techniques on adversarially perturbed AMBER images, ORCA further improves standalone LVLM performance, with gains ranging from +1.20\% to +48.00\% across evaluation metrics. These results demonstrate that ORCA offers a promising path toward building more reliable and robust multimodal systems.

The 2025 Ocean Photographer of the Year announced its winners this week. This photo was taken on April 1, 2024, off Point No Point, WA. In Puget Sound, there's a community of people who prefer watching orcas from the land rather than from boats. Land-based whale watchers in Puget Sound can sometimes get lucky, as these wild apex predators occasionally approach the shore, seemingly curious about their human spectators. My friend is one of those land-based whale enthusiasts, and April 1, 2024, was no ordinary day for her.

Positional bias (PB), manifesting as non-uniform sensitivity across different contextual locations, significantly impairs long-context comprehension and processing capabilities. Previous studies have addressed PB either by modifying the underlying architectures or by employing extensive contextual awareness training. However, the former approach fails to effectively eliminate the substantial performance disparities, while the latter imposes significant data and computational overhead. To address PB effectively, we introduce \textbf{Pos2Distill}, a position to position knowledge distillation framework. Pos2Distill transfers the superior capabilities from advantageous positions to less favorable ones, thereby reducing the huge performance gaps. The conceptual principle is to leverage the inherent, position-induced disparity to counteract the PB itself. We identify distinct manifestations of PB under \textbf{\textsc{r}}etrieval and \textbf{\textsc{r}}easoning paradigms, thereby designing two specialized instantiations: \emph{Pos2Distill-R\textsuperscript{1}} and \emph{Pos2Distill-R\textsuperscript{2}} respectively, both grounded in this core principle. By employing the Pos2Distill approach, we achieve enhanced uniformity and significant performance gains across all contextual positions in long-context retrieval and reasoning tasks. Crucially, both specialized systems exhibit strong cross-task generalization mutually, while achieving superior performance on their respective tasks.

Causal inference is essential for decision-making science while the complexity of the data analysis workflow, ranging from data wrangling to causal analysis, increases substantially as the scale of data grows in complicated business environments. Especially, the execution of the workflow in relational databases by non-experts can result in repetitive bottlenecks which impede timely and responsible business insights. To address this challenge, we propose ORCA (Orchestrating Causal Agent), an LLM agentic system that can automate routine workflows in RDBMS while preserving expert oversight via human-AI interactions. ORCA orchestrates the full data analysis pipeline: interpreting natural language queries, navigating tables from DB servers, generating proper SQL codes, preprocessing data, and configuring modeling processes using causal inference libraries. Domain experts still can control the automation through iterative interactions with ORCA, enabling robust data-driven decision making with less technical expertise in statistical computing. Empirical evaluations on benchmark and synthetic e-commerce datasets demonstrate competitive performance of ORCA in table understanding, query generation, and cause-effect estimation -- achieving over $7\times$ improvement in estimating average treatment compared to GPT-4o mini.

Breakthroughs, discoveries, and DIY tips sent every weekday. Orcas (Orcinus orca) don't appear to be big fans of yachts, but some of them may be curious about humans themselves. According to a study published on June 30 in the Journal of Comparative Psychology, cetology researchers have confirmed dozens of instances of killer whales intentionally approaching people and offering them food--a behavior they typically reserve for building bonds between orca pods. "Orcas often share food with each other--it's a prosocial activity and a way that they build relationships with each other," Jared Towers, a study lead author and executive director of the British Columbia-based research organization Bay Cetology said in a statement. "That they also share with humans may show their interest in relating to us as well."

Learning-based environmental sound recognition has emerged as a crucial method for ultra-low-power environmental monitoring in biological research and city-scale sensing systems. These systems usually operate under limited resources and are often powered by harvested energy in remote areas. Recent efforts in on-device sound recognition suffer from low accuracy due to resource constraints, whereas cloud offloading strategies are hindered by high communication costs. In this work, we introduce ORCA, a novel resource-efficient cloud-assisted environmental sound recognition system on batteryless devices operating over the Low-Power Wide-Area Networks (LPWANs), targeting wide-area audio sensing applications. We propose a cloud assistance strategy that remedies the low accuracy of on-device inference while minimizing the communication costs for cloud offloading. By leveraging a self-attention-based cloud sub-spectral feature selection method to facilitate efficient on-device inference, ORCA resolves three key challenges for resource-constrained cloud offloading over LPWANs: 1) high communication costs and low data rates, 2) dynamic wireless channel conditions, and 3) unreliable offloading. We implement ORCA on an energy-harvesting batteryless microcontroller and evaluate it in a real world urban sound testbed. Our results show that ORCA outperforms state-of-the-art methods by up to $80 \times$ in energy savings and $220 \times$ in latency reduction while maintaining comparable accuracy.