Whole-body vision-language-action (VLA) policies for humanoid loco-manipulation usually condition only on the current view, so they fail when brief object cues vanish before the robot must act. This paper proposes WholeBodyMem, which uses a frozen SAM3 to track task objects, detects object events (appear, disappear, move, change shape) to causally select up to $B_{\max}=7$ key frames, and feeds those native tokens together with mask-pooled object tokens into a flow-matching action expert via cross-attention and FiLM. On the new HuMeGo benchmark and standard SIMPLE tasks, the method raises the $\Psi_0$ base policy's simulated success from $29.1\%$ to $72.6\%$ and real-world Unitree G1 success from $36.7\%$ to $88.3\%$, while also improving reactive tasks to $86.9\%$.

Existing UAV vision-language navigation benchmarks generate trajectories without kinodynamic constraints, leading to jerky, collision-prone flight. KinoUAV introduces a dataset built from acceleration-driven, kinodynamically feasible trajectories with directional free-space distance labels and a two-stage Cruise–Seek task. The proposed KinoUAV-Agent hierarchically plans waypoints and predicts continuous accelerations, reporting state-of-the-art success and stability on the benchmark.

Diagnosing combined hepatocellular-cholangiocarcinoma (cHCC-CCA) is difficult because tumors contain mixed hepatocellular and cholangiocellular morphology. This work attacks the problem with weakly supervised multiple-instance learning: it classifies HES-stained whole-slide images into HCC, CCA, and cHCC-CCA, and introduces an attention-supervision loss and a dual-branch model to force heatmaps onto tumor regions and separate HCC-like from CCA-like components. The clinical aim is to improve both slide-level diagnosis and characterization of intratumoral heterogeneity without costly patch-level annotations.

Current chest X-ray report generators that concatenate all available patient records into the decoder prompt produce long, variable-length inputs that can dilute attention and introduce image-irrelevant details. ICGF-CXRMate tackles this by conditioning learnable queries on the current radiograph to compress heterogeneous patient data into a fixed number of clinically relevant tokens, then selectively fusing them into the visual stream via gated cross-attention. On linked MIMIC-CXR and MIMIC-IV-ED examinations, this yields shorter prompts and stronger clinical metrics, offering a practical way to embed emergency-department context without overwhelming the language model.

Procedural mistake recognition usually gets tangled up in temporal segmentation and low-level vision features. This paper introduces PRISM, a language-grounded framework that maps video steps to text, optionally structures them with semantic roles, and decodes sequences with a lightweight hidden Markov model trained only on correct executions. By assuming benchmark-provided step boundaries, it isolates the core reasoning problem of detecting deviations in step content and order through a unified edit-based taxonomy.

The paper tackles Zero-Shot Domain Generalization (ZSDG), where a model must recognize unseen classes in unseen target domains without access to any source-domain images during training. The proposed Domain Prompt Learning (DoP) built on CLIP introduces a two-stage pipeline: Text Domain Learning (TDL) synthesizes diverse style tokens via text-space orthogonality and consistency losses, and Visual Domain Learning (VDL) refines the best-matching token per test image using self-supervised entropy minimization at test time. This matters because it removes the costly and often infeasible requirement to collect and annotate source-domain data, making domain generalization practical for new or scarce classes.

Cinematic video generation demands both inferring plausible action and camera trajectories from sparse text or reference inputs, and independently controlling character, scene, action, and camera factors without unwanted side effects. The authors argue that simply feeding MLLM features into a Video DiT fails because the DiT can bypass those latents via native text-encoder and VAE-reference pathways. Factor-Aware Predictive Learning (FAPL) fixes this with a three-stage curriculum—latent-only grounding, factor-span replacement, and fixed-generator predictive learning—to force the DiT to rely on MLLM representations and, in turn, teach the MLLM to output generation-compatible latents.

World models are increasingly used to reduce environment interaction when post-training vision-language-action (VLA) policies, yet they are typically trained to minimize next-frame pixel error rather than to produce useful policy gradients. DRILL introduces a bilevel framework that trains the world model to maximize the policy's downstream return after an imagined GRPO update, instantiated as both a lightweight signed surrogate (DRILL-VWI) and a full meta-gradient method (DRILL-IMG). The work matters because it empirically decouples pixel fidelity from policy utility, showing that a world model with substantially worse pixel MSE can yield far better downstream policies.

This survey tackles long-document visual question answering, where answers require locating and integrating evidence scattered across many pages and modalities including charts, tables, and fine-grained layout. Rather than grouping papers by model architecture, the authors decompose the task into four functional components: document parsing, evidence organization and retrieval, generator organization and learning, and evaluation. The core argument is that task bottlenecks—visual fidelity, evidence dispersion, and budget limits—should drive taxonomy design, not just technical novelty. This reframing matters because the field has grown explosively since 2024 and prior surveys have organized literature by approach rather than by the functional requirements that long documents impose.

AgentDreamer tackles text-controllable physics simulation by replacing ambiguous video diffusion priors with explicit 3D motion plans generated by an off-the-shelf multimodal agent. The agent reasons over scene geometry and text to produce dense parametric trajectories, which guide a differentiable MPM simulator toward physically feasible motion aligned with the instruction. A new metric, Motion-IG, isolates the information that simulated motion contributes to recovering masked text beyond the prompt alone. While the experimental scope is narrow, the framework offers a clean separation between semantic reasoning and physical grounding.

GM-GReFEL attacks two coupled problems in dynamic facial expression recognition (DFER): the quadratic cost of self-attention in long clips and noisy crowdsourced labels. It replaces a spatiotemporal Vision Transformer with a Modulated GroupMamba backbone that scans six channel groups—four spatially and two temporally—at linear cost $\mathcal{O}(N)$, re-mixing them via a Channel Affinity Modulation gate. A geometry-aware reliability module blends the classifier output with trainable facial-landmark anchors, weighted by per-sample normalized entropy, to fall back on anatomy when the prediction is uncertain. The authors report best-fold results within 0.3 points of the top method on DFEW and ahead on MAFW, while openly noting that five-fold means are more mixed.

Fine-grained image-text retrieval requires aligning local semantics such as objects, attributes, and relations, yet fixed patches and tokens often produce fragmented matching units. This paper proposes HFCMA, a framework that dynamically aggregates low-level patches and tokens into adaptive visual regions and textual phrases, and then aligns these sets with entropy-regularized optimal transport under a frozen CLIP dual-encoder architecture. By making semantic unit construction an explicit learnable step and enforcing globally consistent matching via a transport plan, HFCMA aims to improve discrimination of subtle differences without sacrificing retrieval scalability.

This survey argues that the central bottleneck for advancing remote sensing vision-language models has shifted from architectural innovation to data curation. It introduces GeoChef, an end-to-end framework that maps dataset taxonomy, construction pipelines, fine-tuning recipes, and evaluation protocols onto a culinary metaphor. By cataloging over 120 datasets and launching a community-driven warehouse, the authors aim to turn small academic teams into proactive data chefs rather than passive consumers of foundation models.

MoG-VLM attacks the prohibitive cost of zero-shot video anomaly detection with Vision-Language Models by treating motion not as a feature but as a lightweight gate that decides when to run expensive VLM inference. It introduces Temporal Gradient Magnitude (TGM) for content-aware frame sampling, anchor-triggered hierarchical lookback windows for temporal refinement, and a tier-aware fusion rule that balances frame-level and video-level语义 signals without any learned parameters. The work reports state-of-the-art zero-shot training-free AUC-ROC on XD-Violence (91.91%), UCF-Crime (84.30%), and MSAD (87.88%), alongside a claimed 1.31× speedup over the prior best method.

DLoss tackles objective misalignment in radiology report generation: standard cross-entropy treats diagnostic and narrative tokens identically, so models can sound fluent while omitting critical findings or inserting false ones. The core idea is to pool decoder probabilities across sequence positions with different operators—noisy-OR for supported diagnoses and tempered log-sum-exp for unsupported ones—and add this as an auxiliary loss without changing architecture or inference. Evaluated across six 3D MedVLMs on ~50,000 abdominal CT scans, DLoss improves clinical metrics on most benchmarks, though gains vary widely by backbone.

Inside-JEPA tackles the opacity of latent world models by embedding relational channels for ordered entity pairs directly into the JEPA latent state. The core idea is that an inverse-dynamics loss—recovering actions from changes in these channels—can ground them to geometric relations without any relation labels during training. This matters because it could let world-model states be read and given goals in relational terms, the natural vocabulary for planning and manipulation.

RecallGS tackles online open-vocabulary semantic 3D Gaussian Splatting from streaming images. Rather than learning a heavy 3D U-Net to inject geometric context into 2D CLIP features, it treats the agent's observation history as a non-parametric prior: it retrieves relevant past frames with CLIP, verifies their geometric support via alpha-compositing weights, and fuses the verified historical semantics into the current Gaussians. The core claim is that this training-free recall-ground-fuse pipeline recovers geometry-aware semantic context without the inference overhead of an additional learned 3D branch.

Direct Forcing tackles few-step autoregressive text-to-video generation by eliminating the costly ODE initialization stage that prior distillation methods require. The authors propose a single-stage recipe that jointly learns causal continuation alignment and rollout distribution matching from reference videos, using maximum mean discrepancy and dynamic gradient balancing. Starting from a bidirectional Wan2.1-$1.3$B checkpoint, the method trains in just $250$ updates and achieves $84.63$ VBench Total, offering a markedly shorter path to a fast causal generator.

Open-vocabulary 3D segmentation relies on CLIP features lifted into Gaussian reconstructions, but raw cosine similarity ignores scene-specific directional concentration. This paper proposes a query-time calibration pipeline that subtracts a shared scene direction, whitens text prototypes with a fixed generic reference vocabulary, and recalibrates scores via soft clustering or a CSLS-based relevance margin. The result is a training-free boost in mIoU across seven lifting methods and four datasets, demonstrating that simply changing how features are compared can be as impactful as redesigning the lifting itself.

Accurate intra-hour solar irradiance forecasting from ground-based sky images is essential for grid balancing and storage scheduling as PV penetration rises. GateFormer tackles this by embedding observed global horizontal irradiance directly into a Vision Transformer token stream, then producing two complementary forecasts—a multi-step trajectory and a direct final-horizon displacement—that are adaptively blended and residual-corrected in prediction space. The core claim is that this forecast-space decomposition, rather than conventional latent fusion or ensemble combination, yields consistent gains across three heterogeneous sky-imaging datasets.