WorldMAP: Bootstrapping Vision-Language Navigation Trajectory Prediction with Generative World Models

Vision-language models (VLMs) and generative world models are opening new opportunities for embodied navigation. VLMs are increasingly used as direct planners or trajectory predictors, while world models support look-ahead reasoning by imagining future views. Yet predicting a reliable trajectory from a single egocentric observation remains challenging. Current VLMs often generate unstable trajectories, and world models, though able to synthesize plausible futures, do not directly provide the grounded signals needed for navigation learning. This raises a central question: how can generated futures be turned into supervision for grounded trajectory prediction? We present WorldMAP, a teacher--student framework that converts world-model-generated futures into persistent semantic-spatial structure and planning-derived supervision. Its world-model-driven teacher builds semantic-spatial memory from generated videos, grounds task-relevant targets and obstacles, and produces trajectory pseudo-labels through explicit planning. A lightweight student with a multi-hypothesis trajectory head is then trained to predict navigation trajectories directly from vision-language inputs. On Target-Bench, WorldMAP achieves the best ADE and FDE among compared methods, reducing ADE by 18.0% and FDE by 42.1% relative to the best competing baseline, while lifting a small open-source VLM to DTW performance competitive with proprietary models. More broadly, the results suggest that, in embodied navigation, the value of world models may lie less in supplying action-ready imagined evidence than in synthesizing structured supervision for navigation learning.

AutoSOTA: An End-to-End Automated Research System for State-of-the-Art AI Model Discovery

Artificial intelligence research increasingly depends on prolonged cycles of reproduction, debugging, and iterative refinement to achieve State-Of-The-Art (SOTA) performance, creating a growing need for systems that can accelerate the full pipeline of empirical model optimization. In this work, we introduce AutoSOTA, an end-to-end automated research system that advances the latest SOTA models published in top-tier AI papers to reproducible and empirically improved new SOTA models. We formulate this problem through three tightly coupled stages: resource preparation and goal setting; experiment evaluation; and reflection and ideation. To tackle this problem, AutoSOTA adopts a multi-agent architecture with eight specialized agents that collaboratively ground papers to code and dependencies, initialize and repair execution environments, track long-horizon experiments, generate and schedule optimization ideas, and supervise validity to avoid spurious gains. We evaluate AutoSOTA on recent research papers collected from eight top-tier AI conferences under filters for code availability and execution cost. Across these papers, AutoSOTA achieves strong end-to-end performance in both automated replication and subsequent optimization. Specifically, it successfully discovers 105 new SOTA models that surpass the original reported methods, averaging approximately five hours per paper. Case studies spanning LLM, NLP, computer vision, time series, and optimization further show that the system can move beyond routine hyperparameter tuning to identify architectural innovation, algorithmic redesigns, and workflow-level improvements. These results suggest that end-to-end research automation can serve not only as a performance optimizer, but also as a new form of research infrastructure that reduces repetitive experimental burden and helps redirect human attention toward higher-level scientific creativity.

VecAttention: Vector-wise Sparse Attention for Accelerating Long Context Inference

Long-context video understanding and generation pose a significant computational challenge for Transformer-based video models due to the quadratic complexity of self-attention. While existing sparse attention methods employ coarse-grained patterns to improve efficiency, they typically incur redundant computation and suboptimal performance. To address this issue, in this paper, we propose \textbf{VecAttention}, a novel framework of vector-wise sparse attention that achieves superior accuracy-efficiency trade-offs for video models. We observe that video attention maps exhibit a strong vertical-vector sparse pattern, and further demonstrate that this vertical-vector pattern offers consistently better accuracy-sparsity trade-offs compared with existing coarse-grained sparse patterns. Based on this observation, VecAttention dynamically selects and processes only informative vertical vectors through a lightweight important-vector selection that minimizes memory access overhead and an optimized kernel of vector sparse attention. Comprehensive evaluations on video understanding (VideoMME, LongVideoBench, and VCRBench) and generation (VBench) tasks show that VecAttention delivers a 2.65$\times$ speedup over full attention and a 1.83$\times$ speedup over state-of-the-art sparse attention methods, with comparable accuracy to full attention. Our code is available at https://github.com/anminliu/VecAttention.

On the Accuracy Limits of Sequential Recommender Systems: An Entropy-Based Approach

Sequential recommender systems have achieved steady gains in offline accuracy, yet it remains unclear how close current models are to the intrinsic accuracy limit imposed by the data. A reliable, model-agnostic estimate of this ceiling would enable principled difficulty assessment and headroom estimation before costly model development. Existing predictability analyses typically combine entropy estimation with Fano's inequality inversion; however, in recommendation they are hindered by sensitivity to candidate-space specification and distortion from Fano-based scaling in low-predictability regimes. We develop an entropy-induced, training-free approach for quantifying accuracy limits in sequential recommendation, yielding a candidate-size-agnostic estimate. Experiments on controlled synthetic generators and diverse real-world benchmarks show that the estimator tracks oracle-controlled difficulty more faithfully than baselines, remains insensitive to candidate-set size, and achieves high rank consistency with best-achieved offline accuracy across state-of-the-art sequential recommenders (Spearman rho up to 0.914). It also supports user-group diagnostics by stratifying users by novelty preference, long-tail exposure, and activity, revealing systematic predictability differences. Furthermore, predictability can guide training data selection: training sets constructed from high-predictability users yield strong downstream performance under reduced data budgets. Overall, the proposed estimator provides a practical reference for assessing attainable accuracy limits, supporting user-group diagnostics, and informing data-centric decisions in sequential recommendation.

Efficient Reasoning with Balanced Thinking

Large Reasoning Models (LRMs) have shown remarkable reasoning capabilities, yet they often suffer from overthinking, expending redundant computational steps on simple problems, or underthinking, failing to explore sufficient reasoning paths despite inherent capabilities. These issues lead to inefficiencies and potential inaccuracies, limiting practical deployment in resource-constrained settings. Existing methods to mitigate overthinking, such as suppressing reflective keywords or adjusting reasoning length, may inadvertently induce underthinking, compromising accuracy. Therefore, we propose ReBalance, a training-free framework that achieves efficient reasoning with balanced thinking. ReBalance leverages confidence as a continuous indicator of reasoning dynamics, identifying overthinking through high confidence variance and underthinking via consistent overconfidence. By aggregating hidden states from a small-scale dataset into reasoning mode prototypes, we compute a steering vector to guide LRMs' reasoning trajectories. A dynamic control function modulates this vector's strength and direction based on real-time confidence, pruning redundancy during overthinking, and promoting exploration during underthinking. Extensive experiments conducted on four models ranging from 0.5B to 32B, and across nine benchmarks in math reasoning, general question answering, and coding tasks demonstrate that ReBalance effectively reduces output redundancy while improving accuracy, offering a general, training-free, and plug-and-play strategy for efficient and robust LRM deployment. Project page and code are available at https://rebalance-ai.github.io .

HDINO: A Concise and Efficient Open-Vocabulary Detector

Despite the growing interest in open-vocabulary object detection in recent years, most existing methods rely heavily on manually curated fine-grained training datasets as well as resource-intensive layer-wise cross-modal feature extraction. In this paper, we propose HDINO, a concise yet efficient open-vocabulary object detector that eliminates the dependence on these components. Specifically, we propose a two-stage training strategy built upon the transformer-based DINO model. In the first stage, noisy samples are treated as additional positive object instances to construct a One-to-Many Semantic Alignment Mechanism(O2M) between the visual and textual modalities, thereby facilitating semantic alignment. A Difficulty Weighted Classification Loss (DWCL) is also designed based on initial detection difficulty to mine hard examples and further improve model performance. In the second stage, a lightweight feature fusion module is applied to the aligned representations to enhance sensitivity to linguistic semantics. Under the Swin Transformer-T setting, HDINO-T achieves \textbf{49.2} mAP on COCO using 2.2M training images from two publicly available detection datasets, without any manual data curation and the use of grounding data, surpassing Grounding DINO-T and T-Rex2 by \textbf{0.8} mAP and \textbf{2.8} mAP, respectively, which are trained on 5.4M and 6.5M images. After fine-tuning on COCO, HDINO-T and HDINO-L further achieve \textbf{56.4} mAP and \textbf{59.2} mAP, highlighting the effectiveness and scalability of our approach. Code and models are available at https://github.com/HaoZ416/HDINO.

Securing the Floor and Raising the Ceiling: A Merging-based Paradigm for Multi-modal Search Agents

Recent advances in Vision-Language Models (VLMs) have motivated the development of multi-modal search agents that can actively invoke external search tools and integrate retrieved evidence through multi-step reasoning. While promising, existing approaches typically rely on large-scale supervised trajectories or expensive reinforcement learning (RL), leading to high training cost, instability, and a severe cold-start problem for standard VLMs. We propose a training-free paradigm to empower VLMs with autonomous search capabilities via cross-modal model merging. By fusing a text-based search agent with a base VLM, we show that multi-modal search capabilities can be effectively composed without any additional multi-modal training data. To mitigate parameter interference during cross-modal integration, we introduce Optimal Brain Merging (OBM), a saliency-aware merging algorithm that identifies task-critical parameters based on their impact on model loss using only a small set of calibration samples. Extensive experiments on search-intensive benchmarks (e.g., InfoSeek, MMSearch) reveal that: (1) Model merging secures a reasonable performance floor as a zero-shot agent, with OBM achieving superior search rates; (2) OBM significantly raises the performance ceiling as a warm-start strategy, achieving faster convergence and higher peak accuracy than standard VLM initialization.

Pinching Antennas for Multiple Access in Multigroup Multicast Communications

This paper aims to design multiple access (MA) schemes to improve the max-min fairness (MMF) for pinching antennas (PAs)-based multigroup multicast communications, where PA placement and resource allocation are jointly optimized. Specifically, three MA schemes are considered to facilitate the multicast transmission: i) treating interference as noise (TIN), ii) non-orthogonal multiple access (NOMA), and iii) time-division multiple access (TDMA) with two PA reconfiguration protocols, namely pinching switching (PS) and pinching multiplexing (PM). i) For TIN, a closed-form solution is derived for optimal power allocation, while a sequential element-wise optimization (SEO) is developed for the PA placement. ii) For NOMA, a recursive power allocation framework incorporating a bisection search is developed, and a hierarchical objective evaluation (HOE) mechanism is incorporated to simplify the SEO process for PA location update. iii) For TDMA, the PS protocol allows the PA locations to be optimized separately using the SEO method, after which the time-power allocation is solved as a convex problem with a global optimum. Under the PM protocol, the PA locations are jointly optimized with the time-power resources through a Karush-Kuhn-Tucker (KKT)-based analytical solution. Numerical results demonstrate that: i) the pinching-antenna system (PASS) architecture significantly outperforms traditional fixed-antenna systems. ii) TDMA-PS achieves superior performance by fully leveraging the flexible PA reconfiguration and benefiting from interference-free transmission, whereas TIN serves as a practical lower-bound solution due to its simplicity despite its limited performance. iii) NOMA consistently outperforms TDMA-PM and, in high transmit power regimes with heterogeneous multicast group distributions, can even surpass the performance achieved by TDMA-PS.

UrbanAlign: Post-hoc Semantic Calibration for VLM-Human Preference Alignment

Aligning vision-language model (VLM) outputs with human preferences in domain-specific tasks typically requires fine-tuning or reinforcement learning, both of which demand labelled data and GPU compute. We show that for subjective perception tasks, this alignment can be achieved without any model training: VLMs are already strong concept extractors but poor decision calibrators, and the gap can be closed externally. We propose a training-free post-hoc concept-bottleneck pipeline consisting of three tightly coupled stages: concept mining, multi-agent structured scoring, and geometric calibration, unified by an end-to-end dimension optimization loop. Interpretable evaluation dimensions are mined from a handful of human annotations; an Observer-Debater-Judge chain extracts robust continuous concept scores from a frozen VLM; and locally-weighted ridge regression on a hybrid visual-semantic manifold calibrates these scores against human ratings. Applied to urban perception as UrbanAlign, the framework achieves 72.2% accuracy ($κ=0.45$) on Place Pulse 2.0 across six categories, outperforming the best supervised baseline by +15.1 pp and uncalibrated VLM scoring by +16.3 pp, with full dimension-level interpretability and zero model-weight modification.

Pinching Antennas-Aided Integrated Sensing and Multicast Communication Systems

A pinching antennas (PAs)-aided integrated sensing and multicast communication framework is proposed. In this framework, the communication performance is measured by the multicast rate considering max-min fairness. Moreover, the sensing performance is quantified by the Bayesian Cramér-Rao bound (BCRB), where a Gauss-Hermite quadrature-based approach is proposed to compute the Bayesian Fisher information matrix. Based on these metrics, PA placement is optimized under three criteria: communications-centric (C-C), sensing-centric (S-C), and Pareto-optimal designs. These designs are investigated in two scenarios: the single-PA case and the multi-PA case. 1) For the single-PA case, a closed-form solution is derived for the location of the C-C transmit PA, while the S-C design yields optimal transmit and receive PA placements that are symmetric about the target location. Leveraging this geometric insight, the Pareto-optimal design is solved by enforcing this PA placement symmetry, thereby reducing the joint transmit and receive PA placement to the transmit PA optimization. 2) For the general multi-PA case, the PA placements constitute a highly non-convex optimization problem. To solve this, an element-wise alternating optimization-based method is proposed to sequentially optimize all PA placements for the S-C design, and is further incorporated into an augmented Lagrangian (AL) framework and a rate-profile formulation to solve the C-C and Pareto-optimal design problems, respectively. Numerical results show that: i) PASS substantially outperforms fixed-antenna baselines in both multicast rate and sensing accuracy; ii) the multicasting gain becomes more pronounced as the user density increases; and iii) the sensing accuracy improves with the number of deployed PAs.