Multimodal retrieval, which seeks to retrieve relevant content across modalities such as text or image, supports applications from AI search to contents production. Despite the success of separate-encoder approaches like CLIP align modality-specific embeddings with contrastive learning, recent multimodal large language models (MLLMs) enable a unified encoder that directly processes composed inputs. While flexible and advanced, we identify that unified encoders trained with conventional contrastive learning are prone to learn modality shortcut, leading to poor robustness under distribution shifts. We propose a modality composition awareness framework to mitigate this issue. Concretely, a preference loss enforces multimodal embeddings to outperform their unimodal counterparts, while a composition regularization objective aligns multimodal embeddings with prototypes composed from its unimodal parts. These objectives explicitly model structural relationships between the composed representation and its unimodal counterparts. Experiments on various benchmarks show gains in out-of-distribution retrieval, highlighting modality composition awareness as a effective principle for robust composed multimodal retrieval when utilizing MLLMs as the unified encoder. Multimodal retrieval, which aims to retrieve semantically relevant contents across multiple modalities such as text, image and audio, is a fundamental task in various information fields. The core ability of multimodal retrieval is to represent multimodal inputs in a shared and comparable embedding space. A prevailing approach to this problem is to adopt unimodal encoders and align the encoded embeddings through contrastive learning (CL).

Running Large Language Models (LLMs) on edge devices is constrained by high compute and memory demands posing a barrier for real-time applications in sectors like healthcare, education, and embedded systems. Current solutions such as quantization, pruning, and retrieval-augmented generation (RAG) offer only partial optimizations and often compromise on speed or accuracy. We introduce HOLA, an end-to-end optimization framework for efficient LLM deployment. Internally, it leverages Hierarchical Speculative Decoding (HSD) for faster inference without quality loss. Externally, AdaComp-RAG adjusts retrieval complexity based on context needs. Together with LoBi, which blends structured pruning (LoRA) and quantization, HOLA delivers significant gains: 17.6% EMA on GSM8K, 10.5% MCA on ARC, and reduced latency and memory on edge devices like Jetson Nano--proving both scalable and production-ready.

Human perception of events is intrinsically tied to distinguishing between completed (perfect and telic) and ongoing (durative) actions, a process mediated by both linguistic structure and visual cues. In this work, we introduce the \textbf{Perfect Times} dataset, a novel, quadrilingual (English, Italian, Russian, and Japanese) multiple-choice question-answering benchmark designed to assess video-language models (VLMs) on temporal reasoning. By pairing everyday activity videos with event completion labels and perfectivity-tailored distractors, our dataset probes whether models truly comprehend temporal dynamics or merely latch onto superficial markers. Experimental results indicate that state-of-the-art models, despite their success on text-based tasks, struggle to mirror human-like temporal and causal reasoning grounded in video. This study underscores the necessity of integrating deep multimodal cues to capture the nuances of action duration and completion within temporal and causal video dynamics, setting a new standard for evaluating and advancing temporal reasoning in VLMs.

In motion simulation, motion cueing algorithms are used for the trajectory planning of the motion simulator platform, where workspace limitations prevent direct reproduction of reference trajectories. Strategies such as motion washout, which return the platform to its center, are crucial in these settings. For serial robotic MSPs with highly nonlinear workspaces, it is essential to maximize the efficient utilization of the MSPs kinematic and dynamic capabilities. Traditional approaches, including classical washout filtering and linear model predictive control, fail to consider platform-specific, nonlinear properties, while nonlinear model predictive control, though comprehensive, imposes high computational demands that hinder real-time, pilot-in-the-loop application without further simplification. To overcome these limitations, we introduce a novel approach using deep reinforcement learning for motion cueing, demonstrated here for the first time in a 6-degree-of-freedom setting with full consideration of the MSPs kinematic nonlinearities. Previous work by the authors successfully demonstrated the application of DRL to a simplified 2-DOF setup, which did not consider kinematic or dynamic constraints. This approach has been extended to all 6 DOF by incorporating a complete kinematic model of the MSP into the algorithm, a crucial step for enabling its application on a real motion simulator. The training of the DRL-MCA is based on Proximal Policy Optimization in an actor-critic implementation combined with an automated hyperparameter optimization. After detailing the necessary training framework and the algorithm itself, we provide a comprehensive validation, demonstrating that the DRL MCA achieves competitive performance against established algorithms. Moreover, it generates feasible trajectories by respecting all system constraints and meets all real-time requirements with low...

Due to excellent mechanism characteristics of high rigidity, maneuverability and strength-to-weight ratio, 6 Degree-of-Freedom (DoF) Stewart structure is widely adopted to construct flight simulator platforms for replicating motion feelings during training pilots. Unlike conventional serial link manipulator based mechanisms, Upset Prevention and Recovery Training (UPRT) in complex flight status is often accompanied by large speed and violent rate of change in angular velocity of the simulator. However, Classical Washout Filter (CWF) based Motion Cueing Algorithm (MCA) shows limitations in providing rapid response to drive motors to satisfy high accuracy performance requirements. This paper aims at exploiting Model Predictive Control (MPC) based MCA which is proved to be efficient in Hexapod-based motion simulators through controlling over limited linear workspace. With respect to uncertainties and control solution errors from the extraction of Terminal Constraints (COTC), this paper proposes a Switchable Model Predictive Control (S-MPC) based MCA under model adaptive architecture to mitigate the solution uncertainties and inaccuracies. It is verified that high accurate tracking is achievable using the MPC-based MCA with COTC within the simulator operating envelope. The proposed method provides optimal tracking solutions by switching to MPC based MCA without COTC outside the operating envelope. By demonstrating the UPRT with horizontal stall conditions following Average Absolute Scale(AAS) evaluation criteria, the proposed S-MPC based MCA outperforms MPC based MCA and SWF based MCA by 42.34% and 65.30%, respectively.

Decentralized partially observable Markov decision processes with communication (Dec-POMDP-Com) provide a framework for multiagent decision making under uncertainty, but the NEXP-complete complexity renders solutions intractable in general. While sharing actions and observations can reduce the complexity to PSPACE-complete, we propose an approach that bridges POMDPs and Dec-POMDPs by communicating only suggested joint actions, eliminating the need to share observations while maintaining performance comparable to fully centralized planning and execution. Our algorithm estimates joint beliefs using shared actions to prune infeasible beliefs. Each agent maintains possible belief sets for other agents, pruning them based on suggested actions to form an estimated joint belief usable with any centralized policy. This approach requires solving a POMDP for each agent, reducing computational complexity while preserving performance. We demonstrate its effectiveness on several Dec-POMDP benchmarks showing performance comparable to centralized methods when shared actions enable effective belief pruning. This action-based communication framework offers a natural avenue for integrating human-agent cooperation, opening new directions for scalable multiagent planning under uncertainty, with applications in both autonomous systems and human-agent teams.

Biological intelligence is more sample-efficient than artificial intelligence (AI), learning from fewer examples. Here we answer why. Given data, there can be many policies which seem "correct" because they perfectly fit the data. However, only one correct policy could have actually caused the data. Sample-efficiency requires a means of discerning which. Previous work showed sample efficiency is maximised by weak-policy-optimisation (WPO); preferring policies that more weakly constrain what is considered to be correct, given finite resources. Biology's sample-efficiency demonstrates it is better at WPO. To understand how, we formalise the "multiscale-competency-architecture" (MCA) observed in biological systems, as a sequence of nested "agentic-abstraction-layers". We show that WPO at low levels enables synthesis of weaker policies at high. We call this "multiscale-causal-learning", and argue this is how we might construct more scale-able, sample-efficient and reliable AI. Furthermore, a sufficiently weak policy at low levels is a precondition of collective policy at higher levels. The higher level "identity" of the collective is lost if lower levels use an insufficiently weak policy (e.g. cells may become isolated from the collective informational structure and revert to primitive behaviour). This has implications for biology, machine learning, AI-safety, and philosophy.

While we do not focus on this particular problem, our method can partially address this issue as it will also cause hue variance in the background area and help the model to rely less on the foreground bias information as well. Knowledge Distillation is proposed to train a student network to mimic the behavior of a larger teacher model Hinton et al. (2015). To avoid the extra costs of teacher network in previous methods Park et al. (2019); Ahn et al. (2019); Tian et al. (2019), researchers have developed selfdistillation approaches that allow models to transfer their own knowledge into themselvesZhu et al. (2018); Xu & Liu (2019); Yun et al. (2020b); Zhang et al. (2019). Among them, CS-KD Yun et al. (2020b) and data distortion Xu & Liu (2019) are relevant to our work as both of them construct training pairs and encourage similar predictions. However, CS-KD uses different training samples within the class to construct the training pair, and data distortion applies the same augmentation to both training samples. In contrast, our method mainly focuses on the appearance variation in videos and utilizes the same sample with different appearances to learn the invariant representations.

Numerous prognostic factors are currently assessed histopathologically in biopsies of canine mast cell tumors to evaluate clinical behavior. In addition, PCR analysis of the c-Kit exon 11 mutational status is often performed to evaluate the potential success of a tyrosine kinase inhibitor therapy. This project aimed at training deep learning models (DLMs) to identify the c-Kit-11 mutational status of MCTs solely based on morphology without additional molecular analysis. HE slides of 195 mutated and 173 non-mutated tumors were stained consecutively in two different laboratories and scanned with three different slide scanners. This resulted in six different datasets (stain-scanner variations) of whole slide images. DLMs were trained with single and mixed datasets and their performances was assessed under scanner and staining domain shifts. The DLMs correctly classified HE slides according to their c-Kit 11 mutation status in, on average, 87% of cases for the best-suited stain-scanner variant. A relevant performance drop could be observed when the stain-scanner combination of the training and test dataset differed. Multi-variant datasets improved the average accuracy but did not reach the maximum accuracy of algorithms trained and tested on the same stain-scanner variant. In summary, DLM-assisted morphological examination of MCTs can predict c-Kit-exon 11 mutational status of MCTs with high accuracy. However, the recognition performance is impeded by a change of scanner or staining protocol. Larger data sets with higher numbers of scans originating from different laboratories and scanners may lead to more robust DLMs to identify c-Kit mutations in HE slides.

Generative multimodal models based on diffusion models have seen tremendous growth and advances in recent years. Models such as DALL-E and Stable Diffusion have become increasingly popular and successful at creating images from texts, often combining abstract ideas. However, like other deep learning models, they also reflect social biases they inherit from their training data, which is often crawled from the internet. Manually auditing models for biases can be very time and resource consuming and is further complicated by the unbounded and unconstrained nature of inputs these models can take. Research into bias measurement and quantification has generally focused on small single-stage models working on a single modality. Thus the emergence of multistage multimodal models requires a different approach. In this paper, we propose Multimodal Composite Association Score (MCAS) as a new method of measuring gender bias in multimodal generative models. Evaluating both DALL-E 2 and Stable Diffusion using this approach uncovered the presence of gendered associations of concepts embedded within the models. We propose MCAS as an accessible and scalable method of quantifying potential bias for models with different modalities and a range of potential biases.