A key challenge in translating Visual Place Recognition (VPR) from the lab to long-term deployment is ensuring a priori that a system can meet user-specified performance requirements across different parts of an environment, rather than just on average globally. A critical mechanism for controlling local VPR performance is the density of the reference mapping database, yet this factor is largely neglected in existing work, where benchmark datasets with fixed, engineering-driven (sensors, storage, GPS frequency) sampling densities are typically used. In this paper, we propose a dynamic VPR mapping approach that uses pairs of reference traverses from the target environment to automatically select an appropriate map density to satisfy two user-defined requirements: (1) a target Local Recall@1 level, and (2) the proportion of the operational environment over which this requirement must be met or exceeded, which we term the Recall Achievement Rate (RAR). Our approach is based on the hypothesis that match patterns between multiple reference traverses, evaluated across different map densities, can be modelled to predict the density required to meet these performance targets on unseen deployment data. Through extensive experiments across multiple VPR methods and the Nordland and Oxford RobotCar benchmarks, we show that our system consistently achieves or exceeds the specified local recall level over at least the user-specified proportion of the environment. Comparisons with alternative baselines demonstrate that our approach reliably selects the correct operating point in map density, avoiding unnecessary over-densification. Finally, ablation studies and analysis evaluate sensitivity to reference map choice and local space definitions, and reveal that conventional global Recall@1 is a poor predictor of the often more operationally meaningful RAR metric.

In January 2026, the open-source agent framework OpenClaw and the agent-only social network Moltbook produced a large-scale dataset of autonomous AI-to-AI interaction, attracting six academic publications within fourteen days. This study conducts a multivocal literature review of that ecosystem and presents two complementary platforms for autonomous scientific research as a design science response to the architectural failure modes identified. ClawdLab, an open-source platform for structured laboratory collaboration, addresses these failure modes through hard role restrictions, structured adversarial critique, PI-led governance, multi-model orchestration, and evidence requirements enforced through external tool verification, in which the principal investigator validates submitted work using available API calls, computational services, and model context protocol integrations rather than relying on social consensus. Beach.science, a public research commons, complements ClawdLab's structured laboratory model by providing a free-form environment in which heterogeneous agent configurations interact, discover research opportunities, and autonomously contribute computational analyses, supported by template-based role specialisation, extensible skill registries, and programmatic reward mechanisms that distribute inference resources to agents demonstrating scientific progress. A three-tier taxonomy distinguishes single-agent pipelines, predetermined multi-agent workflows, and fully decentralised systems, analysing why leading AI co-scientist platforms remain confined to the first two tiers. The composable third-tier architecture instantiated across ClawdLab and beach.science, in which foundation models, capabilities, governance, verification tooling, and inter-lab coordination are independently modifiable, enables compounding improvement as the broader AI ecosystem advances.

The sim-to-real gap, particularly in the inaccurate modeling of contact-rich dynamics like collisions, remains a primary obstacle to deploying robot policies trained in simulation. Conventional physics engines often trade accuracy for computational speed, leading to discrepancies that prevent direct policy transfer. To address this, we introduce Collision Learning via Augmented Sim-to-real Hybridization (CLASH), a data-efficient framework that learns a parameter-conditioned impulsive collision surrogate model and integrates it as a plug-in module within a standard simulator. CLASH first distills a base model from an imperfect simulator (MuJoCo) using large-scale simulated collisions to capture reusable physical priors. Given only a handful of real collisions (e.g., 10 samples), it then (i) performs gradient-based identification of key contact parameters and (ii) applies small-step, early-stopped fine-tuning to correct residual sim-to-real mismatches while avoiding overfitting. The resulting hybrid simulator not only achieves higher post-impact prediction accuracy but also reduces the wall-clock time of collision-heavy CMA-ES search by 42-48% compared to MuJoCo. We demonstrate that policies obtained with our hybrid simulator transfer more robustly to the real world, doubling the success rate in sequential pushing tasks with reinforcement learning and significantly increase the task performance with model-based control.

Recent advances in deep learning, exemplified by Hyper-Connections (HC), have expanded the residual connection paradigm by introducing wider residual streams and diverse connectivity patterns. While these innovations yield significant performance gains, they compromise the identity mapping property of residual connections, leading to training instability, limited scalability, and increased memory overhead. To address these challenges, we propose JPmHC (Jacobian-spectrum Preserving manifold-constrained Hyper-Connections), a framework that replaces identity skips with a trainable linear mixer acting on n parallel streams while explicitly controlling gradient conditioning. By constraining the mixer M on operator-norm-bounded manifolds (e.g., bistochastic, Stiefel, Grassmann), JPmHC prevents gradient pathologies and enhances stability. JPmHC introduces three key contributions: (i) a free-probability analysis that predicts Jacobian spectra for structured skips, providing actionable design rules for mixer selection; (ii) memory-efficient implicit differentiation for fixed-point projections, reducing activation memory and synchronization overhead; and (iii) a Stiefel-constrained mixer via Cayley transforms, ensuring orthogonality without post-hoc normalization. Empirical evaluations on ARC-AGI demonstrate that JPmHC achieves faster convergence, higher accuracy, and lower computational cost compared to bistochastic baselines, with a rank-$p$ Grassmannian variant tracking between the two -- consistent with the spectral theory predictions. As a flexible and scalable extension of HC, JPmHC advances spectrum-aware, stable, and efficient deep learning, offering insights into topological architecture design and foundational model evolution. \newline \newline

Existing online video segmentation models typically combine a per-frame segmenter with complex specialized tracking modules. While effective, these modules introduce significant architectural complexity and computational overhead. Recent studies suggest that plain Vision Transformer (ViT) encoders, when scaled with sufficient capacity and large-scale pre-training, can conduct accurate image segmentation without requiring specialized modules. Motivated by this observation, we propose the Video Encoder-only Mask Transformer (VidEoMT), a simple encoder-only video segmentation model that eliminates the need for dedicated tracking modules. To enable temporal modeling in an encoder-only ViT, VidEoMT introduces a lightweight query propagation mechanism that carries information across frames by reusing queries from the previous frame. To balance this with adaptability to new content, it employs a query fusion strategy that combines the propagated queries with a set of temporally-agnostic learned queries. As a result, VidEoMT attains the benefits of a tracker without added complexity, achieving competitive accuracy while being 5x-10x faster, running at up to 160 FPS with a ViT-L backbone. Code: https://www.tue-mps.org/videomt/

Meenzerisch, the dialect spoken in the German city of Mainz, is also the traditional language of the Mainz carnival, a yearly celebration well known throughout Germany. However, Meenzerisch is on the verge of dying out-a fate it shares with many other German dialects. Natural language processing (NLP) has the potential to help with the preservation and revival efforts of languages and dialects. However, so far no NLP research has looked at Meenzerisch. This work presents the first research in the field of NLP that is explicitly focused on the dialect of Mainz. We introduce a digital dictionary-an NLP-ready dataset derived from an existing resource (Schramm, 1966)-to support researchers in modeling and benchmarking the language. It contains 2,351 words in the dialect paired with their meanings described in Standard German. We then use this dataset to answer the following research questions: (1) Can state-of-the-art large language models (LLMs) generate definitions for dialect words? (2) Can LLMs generate words in Meenzerisch, given their definitions? Our experiments show that LLMs can do neither: the best model for definitions reaches only 6.27% accuracy and the best word generation model's accuracy is 1.51%. We then conduct two additional experiments in order to see if accuracy is improved by few-shot learning and by extracting rules from the training set, which are then passed to the LLM. While those approaches are able to improve the results, accuracy remains below 10%. This highlights that additional resources and an intensification of research efforts focused on German dialects are desperately needed.

World models are becoming central to robotic planning and control as they enable prediction of future state transitions. Existing approaches often emphasize video generation or natural-language prediction, which are difficult to ground in robot actions and suffer from compounding errors over long horizons. Classic task and motion planning models world transitions in logical space, enabling robot-executable and robust long-horizon reasoning. However, they typically operate independently of visual perception, preventing synchronized symbolic and visual state prediction. We propose a Hierarchical World Model (H-WM) that jointly predicts logical and visual state transitions within a unified framework. H-WM combines a high-level logical world model with a low-level visual world model, integrating the long-horizon robustness of symbolic reasoning with visual grounding. The hierarchical outputs provide stable intermediate guidance for long-horizon tasks, mitigating error accumulation and enabling robust execution across extended task sequences. Experiments across multiple vision-language-action (VLA) control policies demonstrate the effectiveness and generality of H-WM's guidance.

Biometric footstep recognition, based on a person's unique pressure patterns under their feet during walking, is an emerging field with growing applications in security and safety. However, progress in this area has been limited by the lack of large, diverse datasets necessary to address critical challenges such as generalization to new users and robustness to shifts in factors like footwear or walking speed. The recent release of the UNB StepUP-P150 dataset, the largest and most comprehensive collection of high-resolution footstep pressure recordings to date, opens new opportunities for addressing these challenges through deep learning. To mark this milestone, the First International StepUP Competition for Biometric Footstep Recognition was launched. Competitors were tasked with developing robust recognition models using the StepUP-P150 dataset that were then evaluated on a separate, dedicated test set designed to assess verification performance under challenging variations, given limited and relatively homogeneous reference data. The competition attracted global participation, with 23 registered teams from academia and industry. The top-performing team, Saeid_UCC, achieved the best equal error rate (EER) of 10.77% using a generative reward machine (GRM) optimization strategy. Overall, the competition showcased strong solutions, but persistent challenges in generalizing to unfamiliar footwear highlight a critical area for future work.

Theory of Mind (ToM) assesses whether models can infer hidden mental states such as beliefs, desires, and intentions, which is essential for natural social interaction. Although recent progress in Large Reasoning Models (LRMs) has boosted step-by-step inference in mathematics and coding, it is still underexplored whether this benefit transfers to socio-cognitive skills. We present a systematic study of nine advanced Large Language Models (LLMs), comparing reasoning models with non-reasoning models on three representative ToM benchmarks. The results show that reasoning models do not consistently outperform non-reasoning models and sometimes perform worse. A fine-grained analysis reveals three insights. First, slow thinking collapses: accuracy significantly drops as responses grow longer, and larger reasoning budgets hurt performance. Second, moderate and adaptive reasoning benefits performance: constraining reasoning length mitigates failure, while distinct success patterns demonstrate the necessity of dynamic adaptation. Third, option matching shortcut: when multiple choice options are removed, reasoning models improve markedly, indicating reliance on option matching rather than genuine deduction. We also design two intervention approaches: Slow-to-Fast (S2F) adaptive reasoning and Think-to-Match (T2M) shortcut prevention to further verify and mitigate the problems. With all results, our study highlights the advancement of LRMs in formal reasoning (e.g., math, code) cannot be fully transferred to ToM, a typical task in social reasoning. We conclude that achieving robust ToM requires developing unique capabilities beyond existing reasoning methods.

While recent advances in Reinforcement Fine-Tuning (RFT) have shown that rule-based reward schemes can enable effective post-training for large language models, their extension to cross-modal, vision-centric domains remains largely underexplored. This limitation is especially pronounced in the medical imaging domain, where effective performance requires both robust visual perception and structured reasoning. In this work, we address this gap by proposing VRFT-Aug, a visual reinforcement fine-tuning framework tailored for the medical domain. VRFT-Aug introduces a series of training strategies designed to augment both perception and reasoning, including prior knowledge injection, perception-driven policy refinement, medically informed reward shaping, and behavioral imitation. Together, these methods aim to stabilize and improve the RFT process. Through extensive experiments across multiple medical datasets, we show that our approaches consistently outperform both standard supervised fine-tuning and RFT baselines. Moreover, we provide empirically grounded insights and practical training heuristics that can be generalized to other medical image tasks. We hope this work contributes actionable guidance and fresh inspiration for the ongoing effort to develop reliable, reasoning-capable models for high-stakes medical applications.

Failures in large-scale cloud systems incur substantial financial losses, making automated Root Cause Analysis (RCA) essential for operational stability. Recent efforts leverage Large Language Model (LLM) agents to automate this task, yet existing systems exhibit low detection accuracy even with capable models, and current evaluation frameworks assess only final answer correctness without revealing why the agent's reasoning failed. This paper presents a process level failure analysis of LLM-based RCA agents. We execute the full OpenRCA benchmark across five LLM models, producing 1,675 agent runs, and classify observed failures into 12 pitfall types across intra-agent reasoning, inter-agent communication, and agent-environment interaction. Our analysis reveals that the most prevalent pitfalls, notably hallucinated data interpretation and incomplete exploration, persist across all models regardless of capability tier, indicating that these failures originate from the shared agent architecture rather than from individual model limitations. Controlled mitigation experiments further show that prompt engineering alone cannot resolve the dominant pitfalls, whereas enriching the inter-agent communication protocol reduces communication-related failures by up to 15 percentage points. The pitfall taxonomy and diagnostic methodology developed in this work provide a foundation for designing more reliable autonomous agents for cloud RCA.

Aerial manipulation (AM) promises to move Unmanned Aerial Vehicles (UAVs) beyond passive inspection to contact-rich tasks such as grasping, assembly, and in-situ maintenance. Most prior AM demonstrations rely on external motion capture (MoCap) and emphasize position control for coarse interactions, limiting deployability. We present a fully onboard perception-control pipeline for contact-rich AM that achieves accurate motion tracking and regulated contact wrenches without MoCap. The main components are (1) an augmented visual-inertial odometry (VIO) estimator with contact-consistency factors that activate only during interaction, tightening uncertainty around the contact frame and reducing drift, and (2) image-based visual servoing (IBVS) to mitigate perception-control coupling, together with a hybrid force-motion controller that regulates contact wrenches and lateral motion for stable contact. Experiments show that our approach closes the perception-to-wrench loop using only onboard sensing, yielding an velocity estimation improvement of 66.01% at contact, reliable target approach, and stable force holding-pointing toward deployable, in-the-wild aerial manipulation.

With the growing employment of learning algorithms in robotic applications, research on reinforcement learning for bipedal locomotion has become a central topic for humanoid robotics. While recently published contributions achieve high success rates in locomotion tasks, scarce attention has been devoted to the development of methods that enable to handle hardware faults that may occur during the locomotion process. However, in real-world settings, environmental disturbances or sudden occurrences of hardware faults might yield severe consequences. To address these issues, this paper presents TOLEBI (A faulT-tOlerant Learning framEwork for Bipedal locomotIon) that handles faults on the robot during operation. Specifically, joint locking, power loss and external disturbances are injected in simulation to learn fault-tolerant locomotion strategies. In addition to transferring the learned policy to the real robot via sim-to-real transfer, an online joint status module incorporated. This module enables to classify joint conditions by referring to the actual observations at runtime under real-world conditions. The validation experiments conducted both in real-world and simulation with the humanoid robot TOCABI highlight the applicability of the proposed approach. To our knowledge, this manuscript provides the first learning-based fault-tolerant framework for bipedal locomotion, thereby fostering the development of efficient learning methods in this field.

Large language models (LLMs) rarely admit uncertainty, often producing fluent but misleading answers, rather than abstaining (i.e., refusing to answer). This weakness is even evident in temporal question answering, where models frequently ignore time-sensitive evidence and conflate facts across different time-periods. In this paper, we present the first empirical study of training LLMs with an abstention ability while reasoning about temporal QA. Existing approaches such as calibration might be unreliable in capturing uncertainty in complex reasoning. We instead frame abstention as a teachable skill and introduce a pipeline that couples Chain-of-Thought (CoT) supervision with Reinforcement Learning (RL) guided by abstention-aware rewards. Our goal is to systematically analyze how different information types and training techniques affect temporal reasoning with abstention behavior in LLMs. Through extensive experiments studying various methods, we find that RL yields strong empirical gains on reasoning: a model initialized by Qwen2.5-1.5B-Instruct surpasses GPT-4o by $3.46\%$ and $5.80\%$ in Exact Match on TimeQA-Easy and Hard, respectively. Moreover, it improves the True Positive rate on unanswerable questions by $20\%$ over a pure supervised fine-tuned (SFT) variant. Beyond performance, our analysis shows that SFT induces overconfidence and harms reliability, while RL improves prediction accuracy but exhibits similar risks. Finally, by comparing implicit reasoning cues (e.g., original context, temporal sub-context, knowledge graphs) with explicit CoT supervision, we find that implicit information provides limited benefit for reasoning with abstention. Our study provides new insights into how abstention and reasoning can be jointly optimized, providing a foundation for building more reliable LLMs.

This work presents a consensus-based Bayesian framework to detect malicious user behavior in enterprise directory access graphs. By modeling directories as topics and users as agents within a multi-level interaction graph, we simulate access evolution using influence-weighted opinion dynamics. Logical dependencies between users are encoded in dynamic matrices Ci, and directory similarity is captured via a shared influence matrix W. Malicious behavior is injected as cross-component logical perturbations that violate structural norms of strongly connected components(SCCs). We apply theoretical guarantees from opinion dynamics literature to determine topic convergence and detect anomaly via scaled opinion variance. To quantify uncertainty, we introduce a Bayesian anomaly scoring mechanism that evolves over time, using both static and online priors. Simulations over synthetic access graphs validate our method, demonstrating its sensitivity to logical inconsistencies and robustness under dynamic perturbation.

Large language models (LLMs) show promise in clinical decision support yet risk acquiescing to patient pressure for inappropriate care. We introduce SycoEval-EM, a multi-agent simulation framework evaluating LLM robustness through adversarial patient persuasion in emergency medicine. Across 20 LLMs and 1,875 encounters spanning three Choosing Wisely scenarios, acquiescence rates ranged from 0-100\%. Models showed higher vulnerability to imaging requests (38.8\%) than opioid prescriptions (25.0\%), with model capability poorly predicting robustness. All persuasion tactics proved equally effective (30.0-36.0\%), indicating general susceptibility rather than tactic-specific weakness. Our findings demonstrate that static benchmarks inadequately predict safety under social pressure, necessitating multi-turn adversarial testing for clinical AI certification.

Cervical spine fractures demand rapid and accurate diagnosis for effective clinical management. This study presents an automated, end-to-end pipeline for fracture detection across cervical vertebrae (C1--C7) that assesses the feasibility of fracture recognition from vertebra-level volumes of interest extracted using estimated 3D masks derived from fused orthogonal 2D segmentations. Unlike traditional 3D methods, our approach approximates 3D volumes via optimized 2D axial, sagittal, and coronal projections to reduce input dimensionality of intermediate pre-processing steps while maintaining high diagnostic performance for downstream fracture classification. First, spine regions of interest are localized from multi-view variance projections using a YOLOv8 detector, achieving a 3D mean Intersection over Union of 94.45%. Next, multi-label vertebra segmentation is performed using a DenseNet121-Unet architecture on energy-based sagittal and coronal projections, attaining a mean Dice score of 87.86%. The orthogonal 2D masks are then fused to reconstruct an estimated 3D mask for each vertebra, which is used to extract volumes of interest from the original CT. These extracted vertebra volumes are subsequently analyzed for fractures using an ensemble of 2.5D spatio-sequential CNN-Transformer models, yielding vertebra-level and patient-level F1 scores of 68.15 and 82.26, with area under the receiver operating characteristic curve scores of 91.62 and 83.04, respectively. The framework is further validated through an explainability study using saliency map visualizations and an interobserver variability analysis. Overall, the results indicate that this projection-based strategy delivers clinically relevant performance comparable to expert radiologists, while reducing the dimensionality of intermediate stages, supporting its potential for practical deployment.

Reinforcement learning with verifiable rewards (RLVR) has emerged as a promising paradigm for post-training large language models (LLMs) on complex reasoning tasks. Yet, the conditions under which RLVR yields robust generalization remain underexplored. This paper provides an empirical study of RLVR generalization in the setting of probabilistic inference over causal graphical models. This setting offers two natural axes along which to examine generalization: (i) the level of the probabilistic query -- associational, interventional, or counterfactual -- and (ii) the structural complexity of the query, measured by the size of its relevant subgraph. We construct a dataset of causal graphs and queries spanning these difficulty axes and fine-tune Qwen-2.5-Instruct models using RLVR or supervised fine-tuning (SFT). We vary both the model scale (3B-32B) and the query level included in training. We find that RLVR yields stronger within-level and across-level generalization than SFT, but only for specific combinations of model size and training query level. Further analysis shows that RLVR's effectiveness depends on the model's initial reasoning competence. With sufficient initial competence, RLVR improves an LLM's marginalization strategy and reduces errors in intermediate probability calculations, producing substantial accuracy gains, particularly on more complex queries. These results show that RLVR can improve specific causal reasoning subskills, with its benefits emerging only when the model has sufficient initial competence. Our code and data is available at https://github.com/zhichul/rlcausal.

Voice-triggered interfaces rely on keyword spotting (KWS) models that must operate continuously under strict memory, latency, and energy constraints on microcontroller-class hardware. Designing such models therefore requires not only high recognition accuracy but also predictable deployability within limited Flash and SRAM budgets. Bayesian optimization is known to handle accuracy-efficiency trade-offs effectively in multi-objective optimization; however, it is highly sensitive to initialization, particularly in the low-budget regimes of TinyML model optimization. We propose Objective-Aware Surrogate Initialization (OASI), which seeds surrogate optimization with Pareto-biased solutions generated via multi-objective simulated annealing. Unlike space-filling or heuristic warm-start methods, OASI initializes the surrogate conditioning process with a bias toward feasible accuracy-memory trade-offs, thus avoiding SRAM-violating configurations. OASI improves hypervolume and convergence robustness over Latin hypercube, Sobol, and random initializations under the same budget constraints on a TinyML KWS problem. Hardware-in-the-loop experiments on STM32 microcontrollers verify the existence of deployable and memory-feasible models without incurring extra optimization costs.

In many real-world settings, reinforcement learning systems suffer performance degradation when the environment encountered at deployment differs from that observed during training. Distributionally robust reinforcement learning (DR-RL) mitigates this issue by seeking policies that maximize performance under the most adverse transition dynamics within a prescribed uncertainty set. Most existing DR-RL approaches, however, rely on strong data availability assumptions, such as access to a generative model or large offline datasets, and are largely restricted to tabular settings. In this work, we propose a fully online DR-RL algorithm with general function approximation that learns robust policies solely through interaction, without requiring prior knowledge or pre-collected data. Our approach is based on a dual-driven fitted robust Bellman procedure that simultaneously estimates the value function and the corresponding worst-case backup operator. We establish regret guarantees for online DR-RL characterized by an intrinsic complexity notion, the robust Bellman-Eluder dimension, covering a broad class of phi-divergence uncertainty sets. The resulting regret bounds are sublinear, do not scale with the size of the state or action spaces, and specialize to tight rates in structured problem classes, demonstrating the practicality and scalability of our framework.

Interpretability can be implemented to understand decisions taken by (black box) models, such as neural machine translation (NMT) or large language models (LLMs). Yet, research in this area has been limited in relation to a manifested problem in these models: gender bias. In this work, we aim to move away from simply measuring bias to exploring its origins. Working with gender-ambiguous natural source data, this exploratory study examines which context, in the form of input tokens in the source sentence (EN), influences (or triggers) the NMT model's choice of a certain gender inflection in the target languages (DE/ES). To analyse this, we compute saliency attribution based on contrastive translations. We first address the challenge of the lack of a scoring threshold and specifically examine different attribution levels of source words on the model's gender decisions in the translation. We compare salient source words with human perceptions of gender and demonstrate a noticeable overlap between human perceptions and model attribution. Additionally, we provide a linguistic analysis of salient words. Our work showcases the relevance of understanding model translation decisions in terms of gender, how this compares to human decisions and that this information should be leveraged to mitigate gender bias.

Robots interacting with humans must not only generate learned movements in real-time, but also infer the intent behind observed behaviors and estimate the confidence of their own inferences. This paper proposes a unified model that achieves all three capabilities within a single hierarchical predictive-coding recurrent neural network (PC-RNN) equipped with a class embedding vector, CERNet, which leverages a dynamically updated class embedding vector to unify motor generation and recognition. The model operates in two modes: generation and inference. In the generation mode, the class embedding constrains the hidden state dynamics to a class-specific subspace; in the inference mode, it is optimized online to minimize prediction error, enabling real-time recognition. Validated on a humanoid robot across 26 kinesthetically taught alphabets, our hierarchical model achieves 76% lower trajectory reproduction error than a parameter-matched single-layer baseline, maintains motion fidelity under external perturbations, and infers the demonstrated trajectory class online with 68% Top-1 and 81% Top-2 accuracy. Furthermore, internal prediction errors naturally reflect the model's confidence in its recognition. This integration of robust generation, real-time recognition, and intrinsic uncertainty estimation within a compact PC-RNN framework offers a compact and extensible approach to motor memory in physical robots, with potential applications in intent-sensitive human-robot collaboration.

In this work, we investigate the potential of weights to serve as effective representations, focusing on neural fields. Our key insight is that constraining the optimization space through a pre-trained base model and low-rank adaptation (LoRA) can induce structure in weight space. Across reconstruction, generation, and analysis tasks on 2D and 3D data, we find that multiplicative LoRA weights achieve high representation quality while exhibiting distinctiveness and semantic structure. When used with latent diffusion models, multiplicative LoRA weights enable higher-quality generation than existing weight-space methods.

Quality-Diversity (QD) algorithms constitute a branch of optimization that is concerned with discovering a diverse and high-quality set of solutions to an optimization problem. Current QD methods commonly maintain diversity by dividing the behavior space into discrete regions, ensuring that solutions are distributed across different parts of the space. The QD problem is then solved by searching for the best solution in each region. This approach to QD optimization poses challenges in large solution spaces, where storing many solutions is impractical, and in high-dimensional behavior spaces, where discretization becomes ineffective due to the curse of dimensionality. We present an alternative framing of the QD problem, called \emph{Soft QD}, that sidesteps the need for discretizations. We validate this formulation by demonstrating its desirable properties, such as monotonicity, and by relating its limiting behavior to the widely used QD Score metric. Furthermore, we leverage it to derive a novel differentiable QD algorithm, \emph{Soft QD Using Approximated Diversity (SQUAD)}, and demonstrate empirically that it is competitive with current state of the art methods on standard benchmarks while offering better scalability to higher dimensional problems. Source code is available at https://github.com/conflictednerd/soft-qd.

High-quality main content extraction from web pages is a critical prerequisite for constructing large-scale training corpora. While traditional heuristic extractors are efficient, they lack the semantic reasoning required to handle the structural heterogeneity of the modern web. Conversely, well-pretrained generative Large Language Models (LLMs) offer superior document comprehension but are prohibited by excessive computational costs, limited context windows, and hallucination risks when applied at web scale. We present \textbf{Dripper}, a lightweight framework that resolves these bottlenecks through four contributions: (1) We reformulate extraction as a \textbf{constrained sequence labeling} task using SLMs (Small Language Models). This paradigm eliminates generative hallucinations and achieves exceptional efficiency, reaching a throughput of 3.08 pages per second on a single A100 GPU. (2) We construct \textbf{WebMainBench}, a rigorous benchmark of 7,809 human-annotated pages covering 5,434 unique domains and multiple languages. Evaluations show our Dripper-0.6B model \textbf{outperforms} heuristics like Trafilatura and rivals massive models like DeepSeek-V3.2(685B), GPT-5 and Gemini-2.5-Pro, offering an optimal efficiency-accuracy trade-off. (3) We demonstrate infrastructural value by \textbf{pre-training a 1B model} on a Dripper-curated corpus (63B tokens). This model significantly outperforms baselines in downstream tasks, proving the critical role of extraction quality and the effectiveness of our framework. (4) We \textbf{open-source} the Dripper-0.6B weights and codebase to facilitate the construction of high-quality datasets.

The rapid development of large vision-language model (VLM) has greatly promoted the research of GUI agent. However, GUI agents still face significant challenges in handling long-horizon tasks. First, single-agent models struggle to balance high-level capabilities and low-level execution capability, facing prevalent issues of responsibility coupling and capability conflicts. Second, agents lack awareness of the task state, leading to progress loss in long-horizon tasks. To address these challenges, we propose a staged execution-feedback reinforcement learning algorithm. Unlike training a unified policy model, we focus on training high-level scheduling models. Specifically, we propose and train two agents: a Coordinator, responsible for the strategic planning and task decomposition; and a State Tracker, responsible for context compression and information management to maintain the task's state and coherence. Based on this, we built the Coordinator-Executor-State Tracker (CES) multi-agent framework, which can be integrated with any low-level Executor model, assisting the Executor in solving long-horizon tasks through task scheduling and state management. Experiments on long-horizon task benchmarks demonstrate that CES significantly enhances the system's planning and state management capabilities. Furthermore, analysis confirms that our trained high-level scheduling module is a generalizable, plug-and-play module that significantly enhances the long-horizon capabilities of various Executors. Code can be available at https://github.com/hehehahi4/CES.

By leveraging tool-augmented Multimodal Large Language Models (MLLMs), multi-agent frameworks are driving progress in video understanding. However, most of them adopt static and non-learnable tool invocation mechanisms, which limit the discovery of diverse clues essential for robust perception and reasoning regarding temporally or spatially complex videos. To address this challenge, we propose a novel Multi-agent system for video understanding, namely VideoChat-M1. Instead of using a single or fixed policy, VideoChat-M1 adopts a distinct Collaborative Policy Planning (CPP) paradigm with multiple policy agents, which comprises three key processes. (1) Policy Generation: Each agent generates its unique tool invocation policy tailored to the user's query; (2) Policy Execution: Each agent sequentially invokes relevant tools to execute its policy and explore the video content; (3) Policy Communication: During the intermediate stages of policy execution, agents interact with one another to update their respective policies. Through this collaborative framework, all agents work in tandem, dynamically refining their preferred policies based on contextual insights from peers to effectively respond to the user's query. Moreover, we equip our CPP paradigm with a concise Multi-Agent Reinforcement Learning (MARL) method. Consequently, the team of policy agents can be jointly optimized to enhance VideoChat-M1's performance, guided by both the final answer reward and intermediate collaborative process feedback. Extensive experiments demonstrate that VideoChat-M1 achieves SOTA performance across eight benchmarks spanning four tasks. Notably, on LongVideoBench, our method outperforms the SOTA model Gemini 2.5 pro by 3.6% and GPT-4o by 15.6%.

Physically-based rendering (PBR) materials are fundamental to photorealistic graphics, yet their creation remains labor-intensive and requires specialized expertise. While generative models have advanced material synthesis, existing methods lack a unified representation bridging natural image appearance and PBR properties, leading to fragmented task-specific pipelines and inability to leverage large-scale RGB image data. We present MatPedia, a foundation model built upon a novel joint RGB-PBR representation that compactly encodes materials into two interdependent latents: one for RGB appearance and one for the four PBR maps encoding complementary physical properties. By formulating them as a 5-frame sequence and employing video diffusion architectures, MatPedia naturally captures their correlations while transferring visual priors from RGB generation models. This joint representation enables a unified framework handling multiple material tasks--text-to-material generation, image-to-material generation, and intrinsic decomposition--within a single architecture. Trained on MatHybrid-410K, a mixed corpus combining PBR datasets with large-scale RGB images, MatPedia achieves native $1024\times1024$ synthesis that substantially surpasses existing approaches in both quality and diversity.

Artificial neural networks are increasingly powerful models of brain computation, yet it remains unclear whether improving their performance in downstream tasks also makes their internal representations more similar to brain signals. To address this question in the auditory domain, we quantified the alignment between the internal representations of 36 different audio models and brain activity from two independent fMRI datasets. Using voxel-wise and component-wise regression, and representation similarity analysis, we found that recent self-supervised audio models with strong performance in diverse downstream tasks are better predictors of auditory cortex activity than previously studied models. To assess the quality of the audio representations, we evaluated these models in 6 auditory tasks from the HEAREval benchmark, spanning music, speech, and environmental sounds. This revealed strong positive Pearson correlations (r > 0.8) between a model's overall task performance and its alignment with brain representations. Finally, we analyzed the evolution of the similarity between audio and brain representations during the pretraining of EnCodecMAE, a recent audio representation model. We discovered that brain similarity increases progressively and emerges early during pretraining, despite the model not being explicitly optimized for this objective. This suggests that brain-like representations can be an emergent byproduct of learning to reconstruct missing information from naturalistic audio data.

Human pose forecasting predicts future poses based on past observations, and has many significant applications in areas such as action recognition, autonomous driving or human-robot interaction. This paper evaluates a wide range of pose forecasting algorithms in the task of absolute pose forecasting, revealing many reproducibility issues, and provides a unified training and evaluation pipeline. After drawing a high-level analogy to the task of speech understanding, it is shown that recent speech models can be efficiently adapted to the task of pose forecasting, and improve current state-of-the-art performance. Finally, the robustness of the models is evaluated, using noisy joint coordinates obtained from a pose estimation model, to reflect a realistic type of noise, which is closer to real-world applications. For this a new dataset variation is introduced, and it is shown that estimated poses result in a substantial performance degradation, and how much of it can be recovered again by unsupervised finetuning.