Continual Reasoning Gym: Diagnosing and Harnessing Shared Reasoning in Continual RLVR

Reinforcement learning with verifiable rewards (RLVR) commonly post-trains reasoning models on multiple tasks, while rerunning multitask RLVR (MTRL) as new tasks are added makes capability expansion costly. We therefore study continual RLVR, which updates the existing model as each task arrives. The central question is whether a model updated this way can perform as well as a jointly trained model. To answer this question, we introduce Continual Reasoning Gym, a continual-RLVR environment that organizes text and visual reasoning tasks into five task sequences. In this setting, we identify two key observations: Sequential RLVR exhibits modest forgetting, yet its final performance remains below that of MTRL. To understand the latter, we decompose final performance and show that forgetting accounts for only part of the gap. To explain the former, we identify shared reasoning: transferable reasoning structure allows training on one task to support others on average. We therefore introduce Continual Prompt Replay (CPR), which harnesses shared reasoning to improve learning on the arriving and future tasks by replaying previous-task prompts and regenerating their responses with the current policy. On average, only CPR reaches MTRL-level performance.

CIGTSurv: Clinical Information Guided Tri-modal Survival Prediction with Local Prototype Association and Global Feature Alignment

Multimodal learning has significantly advanced survival prediction by integrating pathology images with genomic data. However, clinical information, despite its critical role in reflecting a patient' s overall health, remains underutilized due to its discrete, sparse, and low-dimensional nature. Furthermore, the inherent heterogeneity across these modalities pose significant challenges in modeling cross-modal interactions. In this paper, we propose CIGTSurv, a Clinical Information Guided Tri-modal framework for Survival prediction. Specifically, we first design a holistic text template and use pretrained foundation models to transform clinical tabular data into high-dimensional tokenized embeddings. Using clinical information as an anchor, we then introduce a dual-level interaction mechanism: 1) a local prototype association (LPA) module based on cross-attention to explicitly learn token-level correspondences between different modalities, and 2) a global feature alignment (GFA) loss based on Maximum Mean Discrepancy (MMD) to implicitly enhance cross-modal distribution consistency. Extensive experiments on five TCGA cancer cohorts demonstrate that CIGTSurv achieves state-of-the-art (SOTA) survival prediction performance. Our source code is publicly available at https://github.com/Daijing-ai/CIGT-Surv.git.

UniLM-Nav: A Unified Framework for Zero-Shot Last-Mile Navigation

Mobile manipulation requires a robot to navigate to a target object or receptacle and then perform intended manipulation. However, reaching the vicinity of the target does not guarantee a manipulation-ready base pose, a problem known as last-mile navigation. Prior methods for last-mile navigation either rely on manual pose annotation or task-specific training, limiting their scalability to open-vocabulary settings with fine-grained spatial constraints. We propose UniLM-Nav, a unified framework for zero-shot open-vocabulary last-mile navigation. UniLM-Nav decomposes last-mile navigation into view selection, task-conditioned affordance grounding, and geometry-aware base-pose reasoning, all resolved with a shared multimodal large language model (MLLM) backend. Specifically, UniLM-Nav first selects a reference view that best captures the target object or receptacle from recently collected observations. It then grounds task-relevant affordance point in the selected view and lifts the result into the robot-centric coordinate frame. Finally, conditioned on the grounded affordance, task context, and robot geometry, it infers a manipulation-ready base pose for the robot. We evaluate UniLM-Nav on the OVMM benchmark, where it outperforms the previous state-of-the-art method, MoTo, by 3.13 percentage points. Analyses show that the components of our method are crucial to final performance, and that the choice of MLLM also has a substantial effect. We further deploy UniLM-Nav on a Unitree B2 quadruped robot with a 6-DoF Unitree Z1 manipulator, validating its applicability to real-world mobile manipulation tasks.

Are We Ready For An Agent-Native Memory System?

Memory for large language model (LLM) agents has rapidly evolved from simple retrieval-augmented mechanisms into a data management system that supports persistent information storage, retrieval, update, consolidation, and dynamic lifecycle governance throughout agent execution. Despite this evolution, existing evaluations still benchmark agent memory mainly through end-to-end task success metrics (e.g., F1, BLEU), while treating the underlying system as a monolithic black box. As a result, critical system-level concerns, including operational costs, architectural trade-offs across memory modules, and robustness under dynamic knowledge updates, remain insufficiently explored. In this paper, we present a systematic experimental study of agent memory from a data management perspective. We propose an analytical framework that decomposes agent memory into four core modules: memory representation and storage, extraction, retrieval and routing, and maintenance. Under this framework, we evaluate 12 representative memory systems and two reference baselines across five benchmark workloads spanning 11 datasets. Our extensive end-to-end evaluation shows that no single architecture dominates across all scenarios; instead, effectiveness depends heavily on how well the memory structure aligns with the workload bottleneck. Furthermore, through fine-grained ablation studies, we quantify their individual effects on representation fidelity, retrieval precision, update correctness, and long-horizon stability. Finally, we reveal cost-performance trade-offs under realistic workloads, showing localized maintenance is more cost-efficient than global reorganization. Based on these findings, we identify promising directions towards building truly agent-native memory systems. The code is publicly available at https://github.com/OpenDataBox/MemoryData.

SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learning

In deep reinforcement learning (DRL), an agent is trained from a stream of experience. In a continual learning setting, such agents can suffer from plasticity loss: their ability to learn new skills from new experiences diminishes over training. Recently, Mixture-of-Experts (MoE) networks have been reported to enable scaling laws and facilitate the learning of diverse skills. However, in continual reinforcement learning settings, their performance can degenerate as learning proceeds, indicating a loss of plasticity. To address this, building on Neural Tangent Kernel (NTK) theory, we formalize the plasticity loss in MoE policies as a loss of spectral plasticity. We then derive a tractable proxy for spectral plasticity, one expressible in terms of individual expert feature matrices. Leveraging this proxy, we introduce SPHERE, a practical Parseval penalty tailored for MoE-based policies that alleviates the loss of spectral plasticity. On MetaWorld and HumanoidBench, SPHERE improves average success under continual RL by 133% and 50% over an unregularized MoE baseline, while maintaining higher spectral plasticity throughout training.

HGP-Mamba: Integrating Histology and Generated Protein Features for Mamba-based Multimodal Survival Risk Prediction

Recent advances in multimodal learning have significantly improved cancer survival risk prediction. However, the joint prognostic potential of protein markers and histopathology images remains underexplored, largely due to the high cost and limited availability of protein expression profiling. To address this challenge, we propose HGP-Mamba, a Mamba-based multimodal framework that efficiently integrates histological with generated protein features for survival risk prediction. Specifically, we introduce a protein feature extractor (PFE) that leverages pretrained foundation models to derive high-throughput protein embeddings directly from Whole Slide Images (WSIs), enabling data-efficient incorporation of molecular information. Together with histology embeddings that capture morphological patterns, we further introduce the Local Interaction-aware Mamba (LiAM) for fine-grained feature interaction and the Global Interaction-enhanced Mamba (GiEM) to promote holistic modality fusion at the slide level, thus capture complex cross-modal dependencies. Experiments on four public cancer datasets demonstrate that HGP-Mamba achieves state-of-the-art performance while maintaining superior computational efficiency compared with existing methods. Our source code is publicly available at <a href="https://github.com/Daijing-ai/HGP-Mamba.git">this https URL</a>.

Video2LoRA: Unified Semantic-Controlled Video Generation via Per-Reference-Video LoRA

Achieving semantic alignment across diverse video generation conditions remains a significant challenge. Methods that rely on explicit structural guidance often enforce rigid spatial constraints that limit semantic flexibility, whereas models tailored for individual control types lack interoperability and adaptability. These design bottlenecks hinder progress toward flexible and efficient semantic video generation. To address this, we propose Video2LoRA, a scalable and generalizable framework for semantic-controlled video generation that conditions on a reference video. Video2LoRA employs a lightweight hypernetwork to predict personalized LoRA weights for each semantic input, which are combined with auxiliary matrices to form adaptive LoRA modules integrated into a frozen diffusion backbone. This design enables the model to generate videos consistent with the reference semantics while preserving key style and content variations, eliminating the need for any per-condition training. Notably, the final model weights less than 150MB, making it highly efficient for storage and deployment. Video2LoRA achieves coherent, semantically aligned generation across diverse conditions and exhibits strong zero-shot generalization to unseen semantics.

MMNavAgent: Multi-Magnification WSI Navigation Agent for Clinically Consistent Whole-Slide Analysis

Recent AI navigation approaches aim to improve Whole-Slide Image (WSI) diagnosis by modeling spatial exploration and selecting diagnostically relevant regions, yet most operate at a single fixed magnification or rely on predefined magnification traversal. In clinical practice, pathologists examine slides across multiple magnifications and selectively inspect only necessary scales, dynamically integrating global and cellular evidence in a sequential manner. This mismatch prevents existing methods from modeling cross-magnification interactions and adaptive magnification selection inherent to real diagnostic workflows. To these, we propose a clinically consistent Multi-Magnification WSI Navigation Agent (MMNavAgent) that explicitly models multi magnification interaction and adaptive magnification selection. Specifically, we introduce a Cross-Magnification navigation Tool (CMT) that aggregates contextual information from adjacent magnifications to enhance discriminative representations along the navigation path. We further introduce a Magnification Selection Tool (MST) that leverages memory-driven reasoning within the agent framework to enable interactive and adaptive magnification selection, mimicking the sequential decision process of pathologists. Extensive experiments on a public dataset demonstrate improved diagnostic performance, with 1.45% gain of AUC and 2.93% gain of BACC over a non-agent baseline. Code will be public upon acceptance.

MVR: Multi-view Video Reward Shaping for Reinforcement Learning

Reward design is of great importance for solving complex tasks with reinforcement learning. Recent studies have explored using image-text similarity produced by vision-language models (VLMs) to augment rewards of a task with visual feedback. A common practice linearly adds VLM scores to task or success rewards without explicit shaping, potentially altering the optimal policy. Moreover, such approaches, often relying on single static images, struggle with tasks whose desired behavior involves complex, dynamic motions spanning multiple visually different states. Furthermore, single viewpoints can occlude critical aspects of an agent's behavior. To address these issues, this paper presents Multi-View Video Reward Shaping (MVR), a framework that models the relevance of states regarding the target task using videos captured from multiple viewpoints. MVR leverages video-text similarity from a frozen pre-trained VLM to learn a state relevance function that mitigates the bias towards specific static poses inherent in image-based methods. Additionally, we introduce a state-dependent reward shaping formulation that integrates task-specific rewards and VLM-based guidance, automatically reducing the influence of VLM guidance once the desired motion pattern is achieved. We confirm the efficacy of the proposed framework with extensive experiments on challenging humanoid locomotion tasks from HumanoidBench and manipulation tasks from MetaWorld, verifying the design choices through ablation studies.

Multi-modal Knowledge Decomposition based Online Distillation for Biomarker Prediction in Breast Cancer Histopathology

Immunohistochemical (IHC) biomarker prediction benefits from multi-modal data fusion analysis. However, the simultaneous acquisition of multi-modal data, such as genomic and pathological information, is often challenging due to cost or technical limitations. To address this challenge, we propose an online distillation approach based on Multi-modal Knowledge Decomposition (MKD) to enhance IHC biomarker prediction in haematoxylin and eosin (H\&E) stained histopathology images. This method leverages paired genomic-pathology data during training while enabling inference using either pathology slides alone or both modalities. Two teacher and one student models are developed to extract modality-specific and modality-general features by minimizing the MKD loss. To maintain the internal structural relationships between samples, Similarity-preserving Knowledge Distillation (SKD) is applied. Additionally, Collaborative Learning for Online Distillation (CLOD) facilitates mutual learning between teacher and student models, encouraging diverse and complementary learning dynamics. Experiments on the TCGA-BRCA and in-house QHSU datasets demonstrate that our approach achieves superior performance in IHC biomarker prediction using uni-modal data. Our code is available at https://github.com/qiyuanzz/MICCAI2025_MKD.