arXiv:2609.11977v1 Announce Type: new Abstract: Co-work agents execute complex workflows that combine information gathering, tool use, coding, and file manipulation across many model invocations. Because cost and latency accumulate over the full episode, their practical value depends not only on peak capability but also on how efficiently that capability is delivered. Yet many steps in everyday work emphasize state tracking, coordination, recovery, and follow-through rather than frontier-scale reasoning. We present Occamy-1.0, a cost-efficient co-work model obtained by further training the post-trained Qwen3.6-35B-A3B checkpoint. We construct execution-grounded data and environments, capture replayable long-horizon trajectories across multiple harnesses, and use staged post-training to develop and consolidate complementary execution capabilities. Across a broad suite of co-work benchmarks, Occamy-1.0 is consistently among the strongest comparably sized models and remains competitive with substantially larger frontier systems on several tasks. Under our stated evaluation and pricing protocol, its aggregate performance across four representative benchmarks places it at the low-cost knee of the observed cost--performance Pareto frontier. Supporting evaluations in tool calling, coding, and instruction following further show that this specialization preserves broad agentic capability. We release the model weights and a subset of the training data to support research on practical co-work agents and agentic post-training.

arXiv:2609.12373v1 Announce Type: new Abstract: Persona drift remains a central challenge for personalized language models, as user profiles evolve over long interactions rather than remain permanently fixed. Models must therefore revise persistent persona states when preferences genuinely change, while avoiding updates driven by transient, ambiguous, or unresolved observations. We propose CORE, which separates turn-local evidence from persistent persona-state revision and selectively updates grounded user preferences through uncertainty-aware belief revision. We also introduce PERSIST, a held-out post-anchor benchmark for persona-state robustness under sequential interaction stress, covering ambiguity, conflict, and controlled social influence. Across ALOE, PersonaChat, and PERSIST, CORE improves personalized alignment and robustness, with complementary gains in normalized closed-slot state fidelity. Human evaluation and mechanistic controls further support explicit update control beyond stronger generation or persistent memory alone.

arXiv:2609.12394v1 Announce Type: new Abstract: Mobile GUI agents are shifting from multi-module frameworks to native models trained end-to-end, yet industrial deployment faces three persistent gaps. Sandbox training produces a distribution mismatch with production environments; expensive real-device failures remain underutilized; and fixed benchmarks saturate, losing the power to guide iteration. We present BlueLM-GUI, a 35B-A3B mobile GUI agent built as a real-device-centric flywheel that closes these gaps through three principles. Every Sample Matters: a dual-track pipeline with Heterogeneous Triple-System Consensus evaluation and an Error Correction \& Derivation Module salvages every trajectory into usable supervision. Every Rollout Is Real: a three-stage recipe---continual pre-training, supervised fine-tuning, and agentic reinforcement learning on hundreds of real phones---grounds every rollout in real production environments, so the capability the model learns transfers directly to deployment. Every Query Evolves: a quota-driven benchmark methodology with three orthogonal axes enables precise attribution and allows the benchmark to be systematically upgraded as the model improves. BlueLM-GUI achieves 87.4 on MobileGUI-VBench, surpassing the best closed-source model by 5.1 points, and 84.9 on AndroidWorld, the best result among open-source models and competitive with closed-source models. These results demonstrate that grounding model training and iterative improvement in both real devices and the three Every principles yields strong, robust, and transferable mobile GUI capability.

arXiv:2609.12437v1 Announce Type: cross Abstract: Adaptive RAG often uses retrieval-time signals to decide whether another retrieval, reranking, or multimodal step should run. We ask whether these signals add routing value once the query itself is already known. Across document, audio, and video RAG, we compare matched query-only and query+retrieval routers while holding the optional actions, router family, training procedure, and evaluation fixed. On the held-out final evaluation, adding the tested retrieval signals does not produce a reliable routing improvement over the query-only baseline. Some retrieval signals are associated with whether a later step will help, but that predictability does not consistently lead to bet- ter RUN/SKIP decisions. The main lesson is therefore methodological: retrieval-state features should not be credited with routing value unless they improve over a matched query-only control. Our results do not show that routing or retrieval state is generally useless; they show that the incremental value of retrieval signals must be demonstrated rather than assumed.

arXiv:2508.16129v4 Announce Type: replace Abstract: Multimodal large language models (MLLMs) have recently demonstrated remarkable reasoning abilities under reinforcement learning (RL) paradigm. However, most existing multimodal medical reasoning models focus on basic reasoning, which refers to shallow inference based on visual feature matching. In contrast, real-world clinical diagnosis extends beyond basic reasoning, demanding complex reasoning that integrates heterogeneous clinical information (such as chief complaints and medical history) with multimodal medical imaging data. To bridge this gap, we introduce MM-Retinal-Reason, an ophthalmic multimodal dataset covering the full spectrum of perception and reasoning. Specifically, it is the first dataset in ophthalmology to encompass both basic and complex reasoning tasks with Chain-of-Thought (CoT) trajectories, aiming to enhance visual-centric reasoning and emulate realistic clinical decision-making. Building upon MM-Retinal-Reason, we propose OphthaReason, the first RL-enhanced ophthalmic multimodal reasoning model with step-by-step reasoning traces. To enable flexible adaptation to both basic and complex reasoning tasks, we further introduce Uncertainty-Aware Dynamic Thinking (UADT), which estimates sample-level uncertainty via entropy and dynamically modulates exploration depth through a shaped advantage mechanism. Comprehensive experiments demonstrate the effectiveness of our model on both basic and complex reasoning tasks, outperforming general-purpose MLLMs, medical MLLMs, RL-based medical MLLMs, and ophthalmic MLLMs by at least 15.47\%. Project Page: \href{https://github.com/lxirich/OphthaReason}{link}.

arXiv:2510.20486v2 Announce Type: replace-cross Abstract: Imbalanced labels can cause frequent samples to dominate AI-based quantitative remote sensing, degrading rare-event retrieval. In rain-rate retrieval based on satellite infrared brightness temperatures, this imbalance leads to systematic underestimation of rare high-intensity rainfall. In this study, Hurdle-Retrieval Model Imbalanced Learning (RMIL) is proposed. Following a divide-and-conquer strategy, Hurdle-RMIL separates zero inflation from the long-tailed distribution of positive rain. A hurdle model handles zero inflation, whereas RMIL exploits invariance under fixed observation conditions of the rainfall-to-satellite forward process to derive a Bayes-based transformation linking conditional distributions under naturally long-tailed and hypothetical balanced rainfall. This transformation enables the balanced-distribution model to be learned from natural samples without constructing a balanced dataset. Comparisons with conventional learning, classification-regression modeling, cost-sensitive learning, and generative learning using test data from multiple regions in China show that Hurdle-RMIL mitigates systematic underestimation and improves detection of rare high-intensity and extreme rainfall without markedly degrading lower-threshold accuracy. At 0.1-10 mm per hour, its root mean square error remains close to those of the best baselines, and it yields the highest equitable threat score (ETS) at most evaluated thresholds, with its advantage becoming more pronounced at high thresholds. At 30 mm per hour, its ETS is 0.051 versus 0.015 for the best baseline, and its mean error is -25.41 mm per hour versus -28.98 mm per hour. Case studies further show improved representations of rainfall intensity and spatial extent, demonstrating that Hurdle-RMIL effectively addresses rainfall-distribution imbalance and improves the retrieval of rare high-intensity rainfall.

arXiv:2608.26105v2 Announce Type: replace-cross Abstract: Native visual reasoning treats visual generation as the medium of reasoning itself: visual states (i.e. images and videos) are not merely inputs to be understood or outputs to be rendered, but first-class substrates for problem solving beyond language. Yet progress remains bottlenecked by the lack of scalable training tasks, reliable feedback, and controlled comparisons across generative substrates. In this work, we introduce VBVR-Pro, a closed-loop testbed that makes native visual reasoning through generation trainable, verifiable, optimizable, and experimentally controllable. 1) Task scaling. VBVR-Pro turns visual reasoning into a controlled task space of 300 procedurally generated tasks. Models trained on VBVR-Pro show strong transfer beyond the proposed suite across seven external visual reasoning benchmarks such as RISE-Video, MME-CoF-Pro, and BabyVision. 2) Verifiable rewards. VBVR-Pro provides verifiable reward scorers for task-grounded evaluation. Through a systematic study of leading MLLMs as judges, we identify recurring failure modes of the prevalent VLM-as-a-judge paradigm. In contrast, the proposed scorers are grounded in deterministic, task-specific rules, achieve fine-grained alignment with human judgments. Importantly, they serve as reliable reward signals for large-scale multi-task reinforcement learning and demonstrate stronger post-RL performance across visual reasoning tasks. 3) Mechanism study. VBVR-Pro enables controlled modality studies across more than 30 image, video, and interleaved generators. Our analysis shows that video generation remains strongest for tasks requiring persistent spatiotemporal state tracking, while interleaved generation provides a compute-efficient alternative. Critically, ablations and probing suggest the presence of vision-native trajectories that are crucial to visual reasoning. We release all data, models, scorers, and code.

arXiv:2609.09242v1 Announce Type: cross Abstract: Large Language Models (LLMs) often generate natural-language comments while writing code, and these comments become part of the context used to generate the code that follows. However, it remains unclear which properties of comments affect code-generation performance. We study this question through observational analyses and controlled interventions. On LiveCodeBench, neither comment frequency nor broad comment intent reliably predicts pass@1. We then prefill weaker recipient models with comment blocks written by stronger source models, allowing us to separate comment surface form from the solution content they convey. Comments from source solutions that pass the tests raise recipient pass@1 by 17.2% on average. In contrast, comments describing failed solutions provide no reliable gain, while comments written for a different problem reduce pass@1 by 20.8%. Finally, across a wide range of models and prompt variants, most recipient models show no significant recovery of the external-comment gain, and the best case recovers only 24%. These results show that comments help code generation not merely because they are comments, but because they can provide correct solution content that prompting cannot reliably elicit.

arXiv:2609.09752v1 Announce Type: cross Abstract: Automatic Guided Vehicles (AGVs) substantially boost warehouse throughput, but routing large-scale AGV fleets remains challenging. Classical Multi-Agent Pathfinding solvers suffer from exploding combinatorial complexity and super-quadratic runtime, while relying on idealized grid or piecewise-linear motion models that mismatch real-world kinematics. Recent Reinforcement Learning (RL) solutions improve flexibility via decentralized agent policies but depend on discretized spatiotemporal representations, require millions of episodes to converge, and incur full-map observation at every step, which leads to large models, slow convergence, and high inference latency that violates real-time industrial control constraints. To address these bottlenecks, we propose HiRAD, a hierarchical RL framework for continuous-space AGV routing with real-time guarantees: (1) a step-level spatiotemporal representation that translates continuous motion into a differentiable RL problem, (2) a hierarchical strategy that splits heading choice from velocity control to reduce the action space, and (3) an asynchronous event-driven decision pipeline that lowers inference complexity from O(n^2) to O(n) and cuts per-step latency by as much as 71 percent. Across random graphs and two warehouse maps, HiRAD reduces makespan by 45 percent to 63 percent and shortens end-to-end runtime.

arXiv:2608.15877v3 Announce Type: replace Abstract: Search and recommendation serve a shared discovery objective but encode intent differently. We study this boundary through Dear Algo on Threads, a deployed product where open-ended requests such as \emph{more NBA news} or \emph{less politics} steer subsequent feed recommendations rather than return a one-shot result list. Its agentic intent layer compiles explicit, inferred, negative, and compound intent into a grounded executable plan, then invokes conventional retrieval and optional semantic or multimodal reranking. The layer shares an intent-to-retrieval contract without requiring one model or serving path across search-like and recommendation-like modes. We evaluate Dear Algo under a precision-first objective. In a blinded audit of 300 public request-item pairs (296 evaluable), a strict categorical LLM-as-a-judge gate achieved 94.4\% exact-Relevant precision [88.8\%, 98.9\%]. Across 72 normalized request clusters, the full configuration produced 7.73 judge-qualified candidates per 20 slots versus 6.61 for an LLM-derived-query baseline, a gain of 1.11 [0.12, 2.12]. In a candidate-randomized serving-path study restricted to the reranker path's first 72 eligible hours, the user-weighted judge-Irrelevant share among judged admissions was 2.80\% versus 4.78\% off (-1.97 points [-3.02, -0.94]), while Exact-Relevant share was 2.24 points higher [0.08, 4.41]. Together, these studies show how explicit natural-language intent can be carried into feed recommendation under a precision-first evaluation framework

arXiv:2509.22404v2 Announce Type: replace-cross Abstract: Anatomical understanding, which is the ability to identify, localize, or segment anatomical structures, is critical in medical image analysis; however, its progress is constrained by the scarcity of expert-labeled data. A promising remedy is to leverage an annotated reference image to guide the interpretation of an unlabeled target. Although recent vision-language models (VLMs) exhibit non-trivial visual reasoning, their reference-based understanding and fine-grained localization remain limited. We introduce RAU, a framework for reference-based anatomical understanding with VLMs. We first show that a VLM learns to identify anatomical regions through relative spatial reasoning between reference and target images, trained on a moderately sized dataset. We validate this capability through visual question answering (VQA) and bounding box prediction. Next, we demonstrate that the VLM-derived spatial cues can be seamlessly integrated with the fine-grained segmentation capability of SAM2, enabling localization and pixel-level segmentation of small anatomical regions, such as vessel segments. Across two in-distribution and two out-of-distribution datasets, RAU consistently outperforms a SAM2 fine-tuning baseline using the same memory setup, yielding more accurate segmentations and more reliable localization. More importantly, its generalization ability to unseen modalities makes it scalable to unseen datasets, a property crucial for medical image applications. To the best of our knowledge, RAU is the first to explore the capability of VLMs for reference-based identification, localization, and segmentation of anatomical structures in medical images. Its promising performance highlights the potential of VLM-driven approaches for anatomical understanding in automated clinical workflows.

arXiv:2512.19705v2 Announce Type: replace-cross Abstract: We study how generative artificial intelligence (GenAI) reshapes financial analysts' information production. Using the 2023 integration of GenAI into FACTSET as a plausibly exogenous change in AI access, we find that FACTSET-associated reports become markedly richer--featuring 26% more distinct information sources, 24% broader topical coverage, and 21% more analytical methods--while also improving timeliness. However, these gains do not uniformly improve decision quality: relative forecast accuracy declines when analysts face greater information-processing demands. Yet, a machine-learning benchmark processing the same observable inputs shows no analogous deterioration, pointing to a human processing constraint rather than poorer underlying information. Placebo tests using other data vendors make a common platform-wide technology trend unlikely. Overall, GenAI relaxes information-acquisition constraints while making human attention a more important bottleneck.

arXiv:2605.24117v1 Announce Type: new Abstract: Large language model (LLM) agents accumulate rich episodic trajectories while solving real-world tasks, but it remains unclear whether such experience can be distilled into reusable procedural skills. We introduce SkillEvolBench, a diagnostic benchmark for evaluating this step from experience reuse to skill formation. It contains 180 tasks across six real-world agent environments, organized into role-conditioned task families with shared latent procedures. Agents learn from acquisition tasks, update an external skill library using compacted trajectories and verifier feedback, and then face frozen deployment tasks testing context shift, adversarial shortcuts, and composition. By comparing self-generated and curated-start skill evolution against no-skill and raw-trajectory controls, SkillEvolBench separates procedural abstraction from base capability, curated prior knowledge, and direct reuse of episodic traces. Across ten model configurations and three agent harnesses, we find that current agents often adapt locally but rarely form robust reusable skills. Skill-based conditions can improve acquisition or replay, and individual models sometimes gain on specific deployment axes, but these gains are unstable under frozen deployment. Raw-trajectory reuse frequently outperforms distilled skills, suggesting that current abstraction procedures discard contextual and procedural cues that remain useful for future tasks. Capacity and cost analyses further show that writing more skills or larger Tier-3 resource libraries is not sufficient: additional updates can improve coverage while introducing episode-specific drift and procedural clutter. These findings position SkillEvolBench as a testbed for measuring when one-off experience becomes durable procedural knowledge rather than task-local memory.

arXiv:2605.24636v2 Announce Type: new Abstract: While large language models (LLMs) hold transformative potential for medicine, their reasoning robustness and safety in real-world clinical scenarios remain critically underexplored, particularly in dentistry. Here we introduce GlobalDentBench, the first multinational dental benchmark, featuring a taxonomy that encompasses 14 dental specialties across 88 countries and regions spanning six continents. The benchmark comprises 8,978 expert-validated questions across three formats (multiple-choice, short-answer, and case-based questions) and assesses three progressive reasoning levels: knowledge recall (L1), routine reasoning (L2), and individualized reasoning (L3). To ensure data quality, the automated construction framework was calibrated by six senior dentists, achieving expert agreement rates of 99.98% for multiple-choice and short-answer questions and 96.78% for the more complex case-based questions. Evaluation of 12 frontier LLMs on GlobalDentBench revealed a sharp, stepwise performance degradation with increasing reasoning complexity. Specifically, accuracy plummeted from 81.34% on multiple-choice to 64.53% on short-answer and 22.34% on case-based questions, while declining markedly from 74.01% at L1 to 55.64% at L2 and 35.71% at L3. More critically, risk analysis of real-world dental cases demonstrated an alarming overall unsafe rate of 31.01% in LLM-generated clinical recommendations, with 4.51% posing risks of irreversible patient harm and risks particularly pronounced in specialties such as orthodontics. These findings expose fundamental limitations in the medical reasoning and safety of current LLMs. Consequently, GlobalDentBench provides a scalable foundation for trustworthy clinical AI evaluation, underscoring the urgent need for rigorous validation before the safe deployment of these models in healthcare.

arXiv:2605.25250v1 Announce Type: new Abstract: Lipid nanoparticles (LNPs) are among the most clinically mature platforms for nucleic acid delivery, yet designing lipids that are both effective and biologically safe remains a major bottleneck. In practical screening, toxicity is a decision-level constraint: if a lipid is toxic, its efficiency prediction is clinically irrelevant. We propose LipoAgent, a safety-aware multi-agent LLM framework for lipid discovery. LipoAgent combines domain-specific finetuning with a conditional prediction objective that enforces toxicity as a prerequisite for efficiency prediction, and further improves reliability via multi-agent verification with lightweight human oversight when disagreement persists. Across multiple foundation models, LipoAgent achieves an average 32% relative improvement in mRNA transfection efficiency prediction compared with other reported models for lipid design. Wet-lab validation confirms that virtual screening rankings reliably translate to biological transfection outcomes. The code is publicly available at https://github.com/SAI-Lab-NYU/LipoAgent.git.

arXiv:2605.25354v1 Announce Type: new Abstract: While LLMs excel at reasoning over prompts using static pretrained knowledge, they struggle significantly with context learning-the ability to dynamically extract, internalize, and apply new knowledge from complex, task-specific contexts. Recent evaluations on the CL-Bench reveal a critical capability gap: frontier models solve only 17.2% of context-dependent tasks on average.

arXiv:2605.25430v1 Announce Type: new Abstract: Coding agents produce rich trajectories while solving software-engineering tasks. To enable agent self-evolution, these trajectories can be distilled into reusable procedural skills that compactly encode experience to guide future behavior. However, existing skill construction and maintenance methods often rely on fixed prompts and heuristic update rules, leaving it unclear how knowledge should be selected, abstracted, and maintained to best serve downstream agents. We propose CODESKILL, an LLM-based framework that reformulates skill extraction and skill-bank maintenance as a learnable management policy. CODESKILL extracts multi-granularity procedural skills from coding-agent trajectories, evolves skills with new experience, and maintains a compact skill bank for future task solving. We train CODESKILL with reinforcement learning, using a hybrid reward that combines dense rubric-based skill-quality feedback with sparse verifiable execution feedback from the frozen downstream agent. Experiments on EnvBench, SWE-Bench Verified, and Terminal-Bench 2 show that CODESKILL improves average pass rate by 9.69 over the no-skill baseline and by 4.01 over the strongest prompt-based or memory baseline, while maintaining the skill bank at a stable size during iterative construction.

arXiv:2605.25446v1 Announce Type: new Abstract: Electrocardiography (ECG) is central to cardiovascular care, but conventional AI models are often restricted to common arrhythmias and may generalize poorly across populations or clinically subtle diseases. We developed ECG Contrastive Language-Image Pre-training (ECGCLIP), a signal-language contrastive learning framework that aligns ECG waveforms with expert diagnostic reports. ECGCLIP was pre-trained on 2,837,962 ECG studies from 1,324,856 patients and evaluated on a held-out internal test set plus nine independent external cohorts comprising about 1.5 million ECGs. Evaluation covered 89 downstream tasks, including 45 ECG diagnoses, 39 echocardiographic targets, and 5 rare cardiac diseases, using PRAUC as the primary metric. ECGCLIP consistently improved performance over random initialization and Merl-R18 baselines. On the internal test set, ECGCLIP-R34 achieved strong performance for atrial fibrillation (PRAUC 0.900) and ST-segment elevation myocardial infarction (PRAUC 0.383), with robust generalization across all external cohorts. It also improved low-prevalence and diagnostically elusive diseases, including Ebstein anomaly, constrictive pericarditis, dextrocardia, and cardiac amyloidosis, with internal PRAUC values of 0.253, 0.175, 0.121, and 0.201, respectively. ECGCLIP was data efficient, matching or exceeding full-dataset baseline performance with only 10% of training data. Feature visualization and saliency analysis suggested clinically meaningful representations aligned with established electrocardiographic criteria. These findings indicate that large-scale ECG-report contrastive pre-training can expand routine ECG interpretation beyond common arrhythmias toward broad cardiovascular assessment and opportunistic screening of echocardiographic and rare conditions.

arXiv:2605.26029v1 Announce Type: new Abstract: We introduce CausaLab, a scalable environment for evaluating interactive causal discovery by LLM agents. Unlike prior evaluations, CausaLab evaluates both whether an agent can solve a problem using causal evidence and whether its answer is supported by a correct hypothesis about the underlying causal mechanism. Each episode places an agent in a synthetic laboratory: it receives prior measurement records, intervenes on a manipulator crystal, and predicts the resonance frequency of a held-out reactor crystal governed by the same mechanism. The hidden data-generating process is a randomly sampled structural causal model (SCM), so success requires recovering both a causal graph and structural equations rather than recalling prior knowledge. CausaLab also includes a domain-specific language that records the agent's evolving SCM hypothesis, making trajectories inspectable and comparable with ground truth. Experiments show a persistent gap between prediction and mechanism recovery: in the purely observational 6-node setting, GPT-5.2-high reaches 92% task accuracy but only 0.471 all-edge $F_1$. This observation further motivates our exploration of different interaction strategies: Mixed observation--intervention strategies improve structural fidelity: in the mixed 6-node setting, GPT-5.2-high achieves 80% on both task accuracy and all-edge $F_1$. Yet even strong agents struggle to design informative interventions, as pure intervention strategies perform poorly on both task accuracy and all-edge $F_1$. We identify premature stopping as a major weakness of agents, and show that asking the model to verify the consistency between its hypothesis and past data can help mitigate this issue. CausaLab therefore separates predictive success from causal understanding and exposes current LLM agents' limits as experimental causal reasoners.

arXiv:2605.24037v1 Announce Type: cross Abstract: Multimodal motion forecasting is inherently under-supervised: each training scene provides only one realized future, yet multiple plausible futures exist. This sparse supervision often leads to mode collapse (redundant hypotheses and insufficient mode coverage) and unreliable confidence ranking when predicting a small set of trajectories. We propose Mode-as-Sequence, a unified decoding framework that translates an unordered mode set into an ordered mode sequence and explicitly models mode-to-mode dependency. Under this framework, we develop two complementary instantiations. ModeSeq performs recurrent mode decoding, where each mode is generated conditioned on the previously generated modes, encouraging diverse, non-redundant hypotheses with calibrated confidence ordering. To remove the mode-by-mode autoregressive bottleneck, we further propose Parallel ModeSeq, which preserves the same causal dependency using masked mode-to-mode self-attention while decoding all modes in a single forward pass, enabling efficient large-$K$ inference and scalable joint-scene prediction. To learn representative modes and calibrated confidence under sparse labels, we introduce Early-Match-Take-All (EMTA) and its joint-scene extension MA-EMTA, together with a lightweight ranking regularizer that reduces confidence inversions. Extensive experiments on large-scale benchmarks demonstrate consistent improvements in both ranking-oriented metrics and best-of-K accuracy across datasets, horizons, and object types. In the Waymo Open Dataset challenges, ModeSeq achieves 1st place in the 2024 LiDAR-free motion prediction track, and Parallel ModeSeq achieves 1st place in the 2025 Interaction Prediction Challenge, validating the effectiveness of Mode-as-Sequence for both accuracy and efficiency.