# Research Papers — 2026-06 · bundle page 188 of 260

> Full-text markdown of papers #9351–9400 of 12993 published in 2026-06, newest first. AI-readable mirrors of the full 12993-paper 2026-06 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-06/page/188
[← Previous bundle](https://aipod.dev/papers/archive/2026-06/page/187.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-06/page/189.md)

### IMUG-Bench: Benchmarking Unified Multimodal Models on Interleaved Understanding and Generation

Authors: Lingyi Meng, Zecong Tang, Haoran Li, Tengju Ru, Zhejun Cui, Weitong Lian, Qi Kang, Hangshuo Cao, Yichen Zhu, Yechi Liu, Kaixuan Wang, Yu-Jie Yuan et al.
Published: 2026-06-08
Categories: cs.AI, cs.CV, cs.MM
arXiv: https://arxiv.org/abs/2606.09169v1

In recent years, unified multimodal models (UMMs) have emerged to support both understanding and generation within a single framework. Mastering dynamic, multi-turn interleaved image-text dialogues is a crucial task for UMMs in real-world applications. However, existing benchmarks fail to evaluate this important task, as they are often limited to single-turn or static settings, and typically overlook exposure bias in multi-turn interactions. To bridge this gap, we propose IMUG-Bench, a comprehensive benchmark for multi-turn interleaved image-text dialogue of UMMs that jointly evaluates their understanding and generation capabilities. Our IMUG-Bench comprises three classes: Static Spatial, Temporal Causal, and Hybrid, covering 3,113 samples and 12,034 interaction turns. It also includes dynamic understanding questions, thereby supporting evaluation that better reflects real-world multi-turn interaction scenarios. Large-scale experiments on IMUG-Bench systematically evaluate mainstream open-source and closed-source UMMs, revealing their capability boundaries and failure modes, and uncovering pronounced exposure bias on the generation side in multi-turn interactions. We further explore several test-time scaling strategies, including Chain-of-Thought, Self-Verification, and Best-of-N Sampling, which effectively improve generation accuracy and mitigate exposure bias in generation tasks. These findings provide insights into enhancing the robustness and multi-turn interaction capability of future UMMs.

---

### Vision-Language Guided Hyperspectral Object Tracking via Semantics Fusion and Contextual Template Updating

Authors: Rui Yao, Yuhong Zhang, Kunyang Sun, Hancheng Zhu, Jiaqi Zhao, Zhiwen Shao, Abdulmotaleb El Saddik
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09167v1

Hyperspectral object tracking (HOT) leverages the rich spectral information provided by hyperspectral videos (HSVs), offering substantial potential for object tracking. However, efficiently extracting and exploiting spectral information from redundant spectral bands remains a fundamental challenge, which severely limits model generalization and tracking performance. Moreover, in dynamic scenes, targets often experience drastic appearance variations due to factors such as occlusion and illumination changes. These variations lead to large deformations between the current frame and the template. Such discrepancies pose major challenges for existing temporal modeling approaches. In this work, we propose VLHTrack, a novel hyperspectral vision-language (VL) joint tracking framework. Specifically, we incorporate language priors to address the fundamental challenge of spectral redundancy by designing a Language-Guided Band Selection Module (LBSM). By leveraging Large Language Model (LLM) descriptions, LBSM establishes a semantic-to-spectral mapping that mitigates redundancy and accentuates discriminative spectral features. A Multi-Modal Vision-Language Fusion Module is then employed to seamlessly integrate visual and linguistic embeddings, harnessing their complementary advantages to learn coherent cross-modal representations. To address target deformation in long-term sequences, we propose a dynamic update template feature strategy implemented via the Dynamic Template Update with Mamba (DTUM) module. By leveraging selective state space modeling, DTUM learns inter-frame dependencies to update template feature, ensuring efficient template feature evolution guided by temporal context. Experiments on HOT2023 and HOT2024 demonstrate that VLHTrack outperforms state-of-the-art (SOTA) methods.

---

### Reliable to Expressive: A Curriculum for Rubric-Following Safety Judges

Authors: Yongtaek Lim, Hyeji Choi, Minwoo Kim
Published: 2026-06-08
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.09165v1

Safety judges are increasingly deployed to evaluate model outputs against evolving criteria, yet recent meta-evaluation work shows they remain brittle under prompt and rubric variation, with false negative-rate swings of up to 0.24 reported for stylistic perturbations alone. We argue that safety judgment is fundamentally a rubric-following problem: a robust judge must apply the given evaluation criteria consistently across rubric formulations rather than memorize one specific template. We propose a training strategy that combines (i) instance-conditioned dynamic rubrics generated from prompt-response-label triples to expose the judge to the variability of evaluation criteria, and (ii) a reliable-to-expressive curriculum that begins with clean fixed-rubric supervision and progressively introduces noisier dynamic-rubric data. We evaluate on a single human-labeled set under three contrasting rubric prompts (HarmBench-style, ShieldGemma-style, and a domain-specific rubric). Our 12B curriculum judge achieves 94.12-94.88% accuracy across the three rubrics with a cross-rubric range of only 0.76, outperforming general-purpose LLMs, dedicated safety classifiers, and reasoning-oriented judges up to 30B in both peak accuracy and stability. An ablation shows that naively mixing dynamic rubrics into SFT increases cross rubric variance (1.44 -> 3.60); only the curriculum schedule recovers and improves on the fixed rubric baseline (variance 0.76).

---

### Zero-Parameter Geometric Gating for Temporally Stable Low-Altitude UAV Video Semantic Segmentation

Authors: Jingpu Yang, Fengxian Ji, Zhengzhao Lai, Juanfan Wu, Mingxuan Cui, Yufeng Wang
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09162v1

Video semantic segmentation for low-altitude UAVs requires temporal consistency, yet dense optical flow introduces spatially structured noise in the planar regions that dominate aerial imagery. We propose a zero-parameter geometric gate that uses RANSAC homography inlier ratios on a $16\times16$ spatial grid to route each region to either homography or optical flow warp before fusion via Semantic Similarity Propagation. The gate requires no learned parameters -- only a median-threshold binary decision on RANSAC statistics -- adding only 211K trainable parameters (the SSP fusion layer) to a frozen backbone. On synthetic UAVid, the method achieves +4.24--4.91\% mIoU improvement over base models across two architectures (SegFormer-b2 and Hiera-S+UPerNet). Mechanism diagnostics reveal that flow residuals in planar regions are spatially autocorrelated (Moran's I = 0.32, $p < 0.001$), predict boundary instability (Spearman $ρ= 0.66$), and that rigidification recovers temporal consistency from 62\% to 92\% (+29.5pp) in homography-valid regions.

---

### Crop Recommendation and Agricultural Query Answering System Using Spatio-Temporal Graph Neural Networks and Hybrid Retrieval Augmentation

Authors: Prajwal Thapa, Yagya Raj Pandeya
Published: 2026-06-08
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.09160v1

This paper presents a unified system designed to support precision agriculture by integrating advanced weather prediction, crop recommendation, and a question-answering tool for farmers. We propose two deep learning models -- a Transformer-based Graph Neural Network and a Spatio-Temporal Graph Convolutional Network (STGCN) -- to forecast weather conditions for the next 30 days using data from 1,359 locations in Nepal. The STGCN outperforms the Transformer-based model in accuracy (MSE ~0.011 vs. 0.013), effectively modeling both spatial and temporal dependencies in climate data. These predictions are combined with static soil properties such as pH, moisture, and organic content to generate localized crop recommendations through a scoring algorithm that matches each crop's optimal growing conditions. Additionally, we develop a Retrieval-Augmented Generation (RAG) chatbot that leverages domain-specific agricultural documents to answer farmers' questions in natural language. The entire system is deployed via a mobile application, offering real-time suggestions and conversational support. User feedback confirms the system's usability and relevance, especially in rural settings where personalized farming guidance is limited. Overall, our approach demonstrates how combining machine learning models with local agricultural data can empower farmers with actionable insights, promoting more informed decisions, better crop yields, and increased resilience to climate variability.

---

### Unified Energy for Invariant and Independent Decoding in Diffusion Language Models

Authors: Yuchen Yan, Minkai Xu, Zaiquan Yang, Yatao Bian
Published: 2026-06-08
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2606.09159v1

Diffusion Language Models (DLMs) enable parallel text generation by iteratively denoising a full sequence, offering attractive flexibility compared to auto-regressive (AR) decoding. However, existing methods fail to fully capture token relationships, leading to a performance gap relative to AR baselines, especially as the degree of parallelism increases. In this paper, we give a systematic analysis of the gap, identifying three key factors: (i) model capacity, (ii) dependency, and (iii) invariance. To address these issues, we first propose an invariant energy (Inv-E) together with an effective sampling-based estimator to handle the invariance issue. By further combining with the independent energy (Ind-E), we obtain a unified energy (Uni-E), that accounts for all these factors. Uni-E enjoys a unique advantage: it can be computed exactly without sampling-based partition estimation. Besides, Uni-E is model agnostic and can therefore be scaled to models of arbitrary size. We further prove that Uni-E can correct the distribution shift caused by dependency and invariance. Extensive experiments across Diffusion Language Models (DLMs) and Diffusion Large Language Models (DLLMs) demonstrate the effectiveness of the proposed Uni-E.

---

### SEF-CLGC at SemEval-2026 Task 11: Logical Notation Impact on Language Model Performance

Authors: Hanna Abi Akl, Fabien Gandon, Catherine Faron, Pierre Monnin
Published: 2026-06-08
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2606.09157v1

This paper revisits our pipeline called Syllogistic Evaluation Framework-Common Logic Grammar Construction (SEF-CLGC). We combine formal logical notations with Small Language Models (SLMs) to evaluate reasoning performance on the SemEval-2026 Task 11 Subtask 1: Disentangling Content and Formal Reasoning in Large Language Models. Our experiments show that by relying solely on SLMs, trained on a combination of natural and symbolic languages, our best model achieves a content score of 27.80% on the task while significantly lowering the content bias in reasoning.

---

### OmniGen-AR: AutoRegressive Any-to-Image Generation

Authors: Junke Wang, Xun Wang, Qiushan Guo, Peize Sun, Weilin Huang, Zuxuan Wu, Yu-Gang Jiang
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09156v1

Autoregressive (AR) models have demonstrated strong potential in visual generation, offering superior performance with simple architectures and optimization objectives. However, existing methods are typically limited to single-modality conditions, e.g., text, restricting their applicability in real-world scenarios that demand image synthesis from diverse controls. In this work, we present OmniGen-AR, a unified autoregressive framework for Any-to-Image generation. By discretizing various visual conditions through a shared visual tokenizer and text prompts with a text tokenizer, OmniGen-AR supports a broad spectrum of conditional inputs within a single model, including text (text-to-image generation), spatial signals (segmentation-to-image and depth-to-image), and visual context (image editing, frame prediction, and text-to-video generation). To mitigate the risk of information leakage from condition tokens to content tokens, we introduce Disentangled Causal Attention (DCA), which separates the full-sequence causal mask into condition causal attention and content causal attention. It serves as a training-time regularizer without affecting the standard next-token prediction during inference. With this design, OmniGen-AR achieves new state-of-the-art or at least competitive results across a range of benchmark, e.g., 0.63 on GenEval and 80.02 on VBench, demonstrating its effectiveness in flexible and high-fidelity visual generation.

---

### Improved Convergence Analysis of Topology Dependence in Decentralized SGD

Authors: Yuki Takezawa, Anastasia Koloskova, Sebastian U. Stich
Published: 2026-06-08
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.09154v1

Decentralized SGD is a fundamental algorithm in decentralized learning, although the influence of an underlying network topology on its convergence behavior is not yet fully understood. Existing convergence analyses have shown that topologies with a small spectral gap significantly deteriorate the convergence rate of Decentralized SGD in both homogeneous and heterogeneous cases. However, many prior papers have reported that indeed the choice of the topology has a significant experimental impact in the heterogeneous case, but has little experimental impact on training behavior in the homogeneous case. In this paper, we present a tighter convergence analysis of Decentralized SGD, offering a more precise understanding of how topologies affect the convergence rate than the prior analysis. Specifically, unlike existing convergence analyses that used only the spectral gap as a property of the topology, our novel analysis shows that all eigenvalues of the mixing matrix affect the convergence rate. Throughout the experiments, we carefully evaluated the convergence behavior of Decentralized SGD and demonstrated that our novel convergence analysis can more accurately describe the effect of topology on the convergence rate.

---

### Ultra Flash: Scaling Real-Time Streaming Video Generation to High Resolutions

Authors: Luxury, Jie Huang, Zihao Fan, Xiaoxiao Ma, Jun-hao Zhuang, Yuming Li, Zeyue Xue, Siming Fu, Haoran Li, Mingchen Zhong, Guohui Zhang, Shichen Ma et al.
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09150v2

While recent autoregressive video diffusion models achieve remarkable streaming quality, they remain confined to low resolutions (e.g., 480P), leaving efficient, scalable, real-time high-resolution video generation a fundamental open challenge. To bridge this gap, we present Ultra Flash, a cascaded streaming framework capable of real-time high-resolution video generation. Ultra Flash achieves ~30 FPS at 1K resolution and ~18 FPS at 2K resolution on a single GPU through three key contributions: (1) an architecture-preserving T2V-to-TV2V super-resolution training paradigm coupled with an AIGC-oriented data degradation pipeline that effectively preserves the generative capability of the base model, enabling enhanced high-resolution detail when cascaded after mainstream low-resolution generative models; (2) a causal streaming latent upsampler paired with a high-resolution decoder, which enhances spatiotemporal coherence while enabling efficient latent spatial scaling and precise high-resolution decoding with negligible computational overhead; and (3) a cascade high-resolution streaming video generation optimization scheme that first performs hybrid-reward-enhanced sparse causalization and single-step distillation of the super-resolution model, then introduces cascaded streaming self-forcing preference optimization with dynamic cache management, jointly enhancing overall coherence, improving quality, and enabling real-time high-resolution streaming video generation. Extensive experiments demonstrate that Ultra Flash reliably produces ultra-high-resolution streaming video while maintaining state-of-the-art visual quality and superior efficiency. Project Page: https://xin1u.github.io/UltraFlash/

---

### Explicit Representation Alignment for Multimodal Sentiment Analysis

Authors: Baode Wang, Ziming Wang, Huacan Wang, Ronghao Chen, Biao Wu
Published: 2026-06-08
Categories: cs.CL
arXiv: https://arxiv.org/abs/2606.09148v1

Multimodal affective analysis aims to understand human sentiment and emotion by jointly modeling heterogeneous modalities such as text and images. However, multimodal models often fail to consistently outperform strong text-only baselines, with performance varying significantly across fusion strategies. In this work, we identify representation misalignment between independently pretrained modality encoders as a key bottleneck for effective multimodal learning, and show through controlled experiments that alignment prior to fusion is often more important than fusion complexity. To address this issue, we propose a unified multimodal affective analysis framework that leverages vision-language models (VLMs) to convert visual content into structured textual descriptions, projecting heterogeneous modalities into a shared linguistic space and enabling interpretable text-centric reasoning. To further improve robustness, we introduce a hybrid learning strategy that combines semantic token selection with a batch-level uniformity regularization objective, encouraging a more dispersed and stable global feature space while mitigating noise introduced by VLM-generated descriptions. Experiments on multiple multimodal sentiment and emotion benchmarks show that our method consistently outperforms strong unimodal and multimodal baselines, achieving state-of-the-art performance. Our analysis further highlights the critical role of representation alignment in multimodal affective learning.

---

### CAMF-Det: Closure-Aware Multimodal Fusion for LiDAR-Camera 3D Object Detection on UAV Platforms

Authors: Yanze Jiang, Yanfeng Gu, Xian Li
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09143v1

Multimodal 3D object detection based on LiDAR and cameras has demonstrated excellent performance in ground-vehicle scenarios, but has not been explored for Unmanned Aerial Vehicle (UAV) platforms. In UAV top-down scenes, frequent groundobject occlusion dominated by tree canopies causes spatially varying and modality-dependent information degradation. Existing multimodal fusion frameworks neither explicitly model such ground-object occlusion nor embed occlusion awareness into the detection pipeline, limiting their performance in occluded UAV scenes. To address these challenges, we propose CAMF-Det, a closure-aware multimodal fusion framework for LiDAR-camera 3D object detection on UAV platforms, which derives dual-modal occlusion intensity through physics-inspired modeling and embeds them as priors throughout the detection pipeline. First, a dual-modal closure modeling module explicitly constructs occlusion intensity ground truth for both modalities offline via a Beer-Lambert-inspired formulation and building-mask correction. Second, using these ground-truth maps as supervision, a dual-modal prediction network converts the offline modeling results into online occlusion intensity predictions under single-frame inference. Third, both ground-truth and predicted occlusion intensity are injected into data augmentation, feature encoding, multimodal fusion, and detection head, enabling adaptive detection under spatially varying and modality-dependent information degradation. Experiments on two self-built UAV-based multimodal datasets, SI3D-DI and SI3D-DII, demonstrate that CAMF-Det achieves the best performance across all difficulty levels, with hard-level mAP$_{\mathrm{BEV}}$ improvements of 9.43% and 4.88% over the best competing methods, respectively. These results confirm the effectiveness of explicit occlusion prior modeling and exploitation for robust multimodal 3D detection in UAV scenes.

---

### Decoding Pedestrian Crossing Intention from Egocentric Vision via Vision Language Models

Authors: Danya Li, Xiang Su, Yan Feng, Rico Krueger
Published: 2026-06-08
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2606.09142v1

Egocentric vision offers a first-person view of human perception and decision making, yet its potential for traffic-safety prediction remains underexplored. In this work, we study the decoding of pedestrian crossing intentions from short egocentric video clips. We approach this by formulating the task as a closed-ended visual question answering (VQA) problem and leveraging vision language models (VLMs) to predict the pedestrians' intent. We first benchmark three families of state-of-the-art VLMs in a zero-shot setting, finding that they achieve moderate gains over random guessing but exhibit limited higher-level traffic reasoning. Motivated by these findings, we further adapt VLMs to the target task using parameter-efficient fine-tuning. Our results show that the fine-tuned models substantially outperform their zero-shot counterparts and achieve a 9\% accuracy improvement over a specialized transformer-based baseline. Finally, we demonstrate that incorporating additional contextual cues, including ego motion, vehicle motion, and eye gaze, further improves predictive performance. In particular, the fine-tuned Qwen3-VL-2B model guided by eye gaze and ego motion achieves a 14.5% accuracy improvement over the transformer baseline, establishing a new state of the art for egocentric pedestrian intent decoding.

---

### DiffSight-Former: Modeling Structural Differences and Temporal Dynamics for Glaucoma Progression Prediction

Authors: Yi Huang, Lei Bi, Jinman Kim
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09140v1

Glaucoma is a leading cause of irreversible blindness worldwide, and early detection from fundus images is critical for effective disease management. While deep learning has achieved promising performance in fundus image analysis, most existing methods rely on single time-point images and fail to capture longitudinal structural and vascular changes associated with disease progression. Sequential fundus images acquired during clinical follow-up provide valuable temporal information; however, current sequential models often struggle to detect subtle early progression signals and commonly depend on fixed-length inputs or diagnostic cues from already glaucomatous images, limiting their clinical utility for early prediction. To address these limitations, we propose DiffSight-Former, a framework for glaucoma progression prediction from sequential fundus images. It incorporates a time-variant feature extraction module based on a fundus-specific foundation model to obtain robust anatomical representations. A multi-structure difference modeling module is introduced to quantify progression-related changes in the optic disc/cup region and retinal vasculature. These representations are integrated with temporal interval embeddings and processed by a time-aware Transformer to model disease progression and estimate the probability of future glaucoma onset. Experiments were conducted on two longitudinal datasets, SIGF (405 sequences) and GRAPE (263 sequences). On SIGF, DiffSight-Former achieved an AUC of 91.54% and a sensitivity of 92.16% for progression prediction. On GRAPE, it achieved an average accuracy of 87.48% across three clinical visual-field progression criteria. Compared with existing approaches, DiffSight-Former demonstrates strong performance and robustness across different temporal settings, highlighting its potential for longitudinal glaucoma monitoring and early risk prediction.

---

### A Geometric Framework for Absolute Pose and Velocity Estimation with Event Cameras

Authors: Zibin Liu, Shunkun Liang, Banglei Guan, Yang Shang, Qifeng Yu, Ji Zhao
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09139v2

Despite the rapid advancements in event-based motion estimation, current geometric methods primarily focus on velocity estimation. However, absolute pose estimation, which is equally crucial for key applications such as robotic navigation and augmented reality, remains relatively underexplored. Consequently, the simultaneous recovery of absolute pose and velocity from event streams remains an open and challenging problem. To address this gap, we propose a geometric framework for absolute pose and velocity estimation by leveraging 3D lines in the scene and the events they trigger. At the core of the framework lie two key geometric constraints: the orthogonality between a 3D line and the normal vector of its corresponding event plane, and the collinearity of an event with the 2D projection of its associated line. Based on these constraints, we present both linear and polynomial solvers for absolute pose estimation. The former enables efficient computation, while the latter provides a globally optimal solution for rotation. For velocity estimation, we develop an efficient linear solver and a more accurate optimization-based solver to recover both angular and linear velocities. Notably, our methods require a minimum of three event-line correspondences to determine the 6-DoF absolute pose or velocities independently. Extensive experiments in simulation and on real-world datasets demonstrate that our methods achieve state-of-the-art performance, with significant improvements in accuracy and computational efficiency compared to existing methods. The demo code is publicly available at https://github.com/Zibin6/EventPoseVelocity.

---

### Claw-R1: A Step-Level Data Middleware System for Agentic Reinforcement Learning

Authors: Daoyu Wang, Mingyue Cheng, Qingchuan Li, Shuo Yu, Jie Ouyang, Qi Liu
Published: 2026-06-08
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2606.09138v1

Agentic reinforcement learning (RL) has become an important post-training paradigm for turning LLMs from static chatbots into interactive agents, giving rise to representative applications such as OpenClaw. Existing work mainly focuses on policy optimization algorithms and training frameworks, but pays less attention to the full data lifecycle of agent-environment interactions, from data production to training consumption. To bridge this gap, we present Claw-R1, an interactive step-level data middleware system for agentic RL. Claw-R1 connects heterogeneous agent runtimes with RL training backends through two core components: a Gateway Server and a Data Pool. The Gateway Server captures multi-turn interaction steps through a unified LLM API entry point, while the Data Pool organizes them into step-level records consisting of prompt IDs, response IDs, rewards and other metadata. In our demo, users can interactively inspect live trajectories, examine the state, action, and reward of each step, curate data by quality and readiness, and configure training-ready batches for different downstream RL algorithms. Overall, Claw-R1 treats agent interaction traces as managed data assets rather than temporary runtime logs. Through this demonstration, we hope to encourage the community to recognize the importance of data management in agentic RL. Our code is available at https://github.com/AgentR1/Claw-R1 and the demonstration video can be found at link https://youtu.be/Pw47dAOw6B0.

---

### Steganography Without Modification: Hidden Communication via LLM Seeds

Authors: Felix Mächtle, Jonas Sander, Sebastian Berndt, Ben Weimar, Nils Loose, Thomas Eisenbarth
Published: 2026-06-08
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2606.09135v1

We demonstrate that widely deployed Large Language Model (LLM) inference stacks harbor a steganographic channel that requires no modification to model weights, sampling code, or output distributions. The channel exploits a structural property of deterministic decoding: pseudo-random number generators (PRNGs) used in inverse-transform sampling produce a seed-dependent sequence of token-level probability intervals that can be reconstructed from the generated text alone. A sender encodes a secret message in the PRNG seed before generation; a receiver reconstructs the intervals and recovers the seed, and thus the hidden payload, by exhaustive search over the seed space. We formalize two operational modes. In the known-prompt setting, sender and receiver share the prompt, enabling exact interval reconstruction and perfect seed recovery via forced alignment. In the unknown-prompt setting, only the generated text is available; approximate interval reconstruction combined with a maximum-hit-count scoring strategy still permits reliable recovery from sufficiently long outputs. Extensive experiments across six model families and five heterogeneous text domains show that, in the known-prompt setting, full 32-bit seed recovery from the complete 2^32 candidate space achieves up to 100% accuracy, depending on model and text domain, within 300 tokens and under 35 seconds on a single GPU. In the unknown-prompt setting, recovery reaches near-perfect accuracy at 600-800 tokens in about 12 seconds. We further analyze the influence of prompting strategies, tokenization ambiguities, and sampling hyperparameters on channel reliability. Moreover, we discuss several applications of our results: First, it allows for the steganographic transmission of 32 bits, but also shows that ignorance of the prompt is not a valid security assumption.

---

### From USD Scenes to Knowledge Graphs: Zero-Shot Ontology Grounding with LLMs

Authors: Jiangtao Shuai, Zongxiong Chen, Manfred Hauswirth, Sonja Schimmler
Published: 2026-06-08
Categories: cs.RO, cs.AI, cs.CL, cs.CV, cs.GR
arXiv: https://arxiv.org/abs/2606.09134v1

Constructing knowledge graphs from 3D simulation scenes is essential for robot task reasoning, but the key bottleneck, grounding scene objects to formal ontology classes, still relies on manually curated dictionaries that are brittle and do not generalize across assets. We investigate whether large language models (LLMs) can automate this grounding step for Universal Scene Description (USD) scenes as a zero-shot, training-free alternative. On a kitchen scene (125 objects) with SOMA-HOME Ontology, LLMs achieve 90-96% exact-match accuracy with descriptive names and 49-89% with abbreviated names, substantially outperforming dictionary and embedding baselines. Under fully opaque names, context-augmented prompting recovers up to 48%. Feature ablation reveals that LLMs primarily exploit semantic cues in the scene graph (sibling names and parent paths); anonymizing these cues reduces accuracy to 0-6%, while geometry alone yields only 4-17%.

---

### Vision Language Model Helps Private Information De-Identification in Vision Data

Authors: Tiejin Chen, Pingzhi Li, Kaixiong Zhou, Tianlong Chen, Hua Wei
Published: 2026-06-08
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.09132v1

Visual Language Models (VLMs) have gained significant popularity due to their remarkable ability. While various methods exist to enhance privacy in text-based applications, privacy risks associated with visual inputs remain largely overlooked such as Protected Health Information (PHI) in medical images. To tackle this problem, two key tasks: accurately localizing sensitive text and processing it to ensure privacy protection should be performed. To address this issue, we introduce VisShield (Vision Privacy Shield), an end-to-end framework designed to enhance the privacy awareness of VLMs. Our framework consists of two key components: a specialized instruction-tuning dataset OPTIC (Optical Privacy Text Instruction Collection) and a tailored training methodology. The dataset provides diverse privacy-oriented prompts that guide VLMs to perform targeted Optical Character Recognition (OCR) for precise localization of sensitive text, while the training strategy ensures effective adaptation of VLMs to privacy-preserving tasks. Specifically, our approach ensures that VLMs recognize privacy-sensitive text and output precise bounding boxes for detected entities, allowing for effective masking of sensitive information. Extensive experiments demonstrate that our framework significantly outperforms existing approaches in handling private information, paving the way for privacy-preserving applications in vision-language models. Our dataset and code can be found here.

---

### Late-Layer Fusion is Enough: Dual-Path Vision Token Routing for Multimodal Large Language Models under Visual Saturation

Authors: Siyuan Liu, Jinyang Wu
Published: 2026-06-08
Categories: cs.AI, cs.CL, cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2606.09131v1

Multimodal large language models (MLLMs) commonly inherit the deep, symmetric Transformer backbone designed for unimodal text modeling, and apply the same computation uniformly to image and language tokens. This design overlooks a key modality asymmetry: image and text tokens differ substantially in information density, redundancy, and required reasoning depth. Through a layer-wise analysis of LLaVA-1.5, we observe that vision tokens tend to saturate in the middle layers. Specifically, text-to-image attention decreases from 0.68 at layer 0 to 0.07 by layer 4, and stabilizes near 0.04 after layer 18, whereas text tokens continue to benefit from deep semantic processing. These findings suggest a mismatch between architectural symmetry and depth-asynchronous modality evolution, resulting in redundant visual computation and possible drift in perceptual representations during deep task-specific adaptation. Motivated by this, we propose Dual-Path Vision Token Routing (DPVR), a modality-asymmetric routing framework for efficient MLLMs. Its core instantiation, DPVR-LF (Late-Layer Fusion), routes vision tokens at the saturation point into a one-layer trainable side branch, runs a thirteen-layer text-only forward that skips image positions in the deep stack, and re-fuses the visual and textual streams only at the final layer. With approximately 3% trainable parameters, DPVR-LF preserves competitive multimodal performance on standard benchmarks while reducing visual computation in the deep Transformer stack. The results challenge the conventional assumption that vision tokens must traverse all deep language-model layers, and indicate that a single late fusion layer can be sufficient for maintaining strong perceptual competence in LLaVA-style MLLMs.

---

### OpenOpt: An Open-Source SRAM Optimizer Based on Equivalent Circuit Model

Authors: Yikai Wang, Yiheng Wu, Can Wang, Bohao Liu, Junhao Ma, Zhuohua Liu, Qinxin Mei, Shan Shen
Published: 2026-06-08
Categories: cs.NE, cs.AR
arXiv: https://arxiv.org/abs/2606.09129v1

This paper proposes a co-optimization framework that jointly optimizes SRAM architecture and transistor sizing using equivalent circuit models. The framework simplifies inactive SRAM cells into equivalent RC loads and static power models, achieving up to 61.4$\times$ simulation speedup while maintaining high fidelity (read/write delay error $<$0.22%, power error $<$1.68%). A joint search space encompassing architecture parameters and device sizing integrates seven algorithms including SA, PSO, Bayesian Optimization variants, and multi-objective evolutionary algorithms. Based on FreePDK45, ablation experiments confirm complementary gains from architecture selection and transistor sizing. Among all algorithms, MOEA/D achieves the best Figure of Merit (8.2721), yielding 6.2% improvement in SNM, 73.6% reduction in area, and 42.3% reduction in peak power. The framework is publicly available at https://github.com/W1Y1K1/OpenOpt.

---

### Unveiling Privacy Risks in Multi-modal Large Language Models: Task-specific Vulnerabilities and Mitigation Challenges

Authors: Tiejin Chen, Pingzhi Li, Kaixiong Zhou, Tianlong Chen, Hua Wei
Published: 2026-06-08
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2606.09125v1

Privacy risks in text-only Large Language Models (LLMs) are well studied, particularly their tendency to memorize and leak sensitive information. However, Multi-modal Large Language Models (MLLMs), which process both text and images, introduce unique privacy challenges that remain underexplored. Compared to text-only models, MLLMs can extract and expose sensitive information embedded in images, posing new privacy risks. We reveal that some MLLMs are susceptible to privacy breaches, leaking sensitive data embedded in images or stored in memory. Specifically, in this paper, we (1) introduce MM-Privacy, a comprehensive dataset designed to assess privacy risks across various multi-modal tasks and scenarios, where we define Disclosure Risks and Retention Risks. (2) systematically evaluate different MLLMs using MM-Privacy and demonstrate how models leak sensitive data across various tasks, and (3) provide additional insights into the role of task inconsistency in privacy risks, emphasizing the urgent need for mitigation strategies. Our findings highlight privacy concerns in MLLMs, underscoring the necessity of safeguards to prevent data exposure. Our dataset and code can be found here.

---

### A Regret Minimization Framework on Preference Learning in Large Language Models

Authors: Suhwan Kim, Taehyun Cho, Geon-Hyeong Kim, Yu Jin Kim, Youngsoo Jang, Moontae Lee, Jungwoo Lee
Published: 2026-06-08
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.09124v1

Reinforcement learning with verifiable rewards (RLVR) has enabled progress on reasoning-intensive tasks by relying on task-specific verifiers that provide automated correctness signals. However, many realistic language tasks are difficult to equip with reliable verifiers, motivating a growing reliance on reinforcement learning from human feedback (RLHF). In this setting, we argue that a closer examination of how human feedback should be interpreted is essential. We introduce Regret-based Preference Optimization $(\textbf{RePO})$, which reframes RLHF through $\textit{regret minimization}$ rather than reward maximization. Human preferences are often shaped by $\textit{prospective}$ anticipation of outcomes and $\textit{counterfactual}$ comparisons to alternative behaviors, rather than by immediate, outcome-independent utility. $\textbf{RePO}$ captures this structure by modeling preferences as behavior-conditioned assessments of relative suboptimality. Experiments on mathematical reasoning benchmarks and human preference datasets demonstrate consistent performance gains, indicating that $\textbf{RePO}$ is an effective and human-aligned approach for training large language models.

---

### An Enhanced Geometric-Spectral Feature Learning Framework for Airborne Multispectral Point Cloud Classification

Authors: Xian Li, Yanfeng, Gu Linghua Xu, Aleksandra Pižurica
Published: 2026-06-08
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2606.09123v2

Multispectral point cloud (MPC) is composed of 3D spatial-spectral information, which holds tremendous potential for accurate land-cover classification. However, the representation power of classification models is limited by inherent high-dimensional and heterogeneous spatial-spectral information, unbalanced sample distribution, and inter-class spectral similarity of airborne MPCs. We build two MPC datasets and propose an enhanced geometric-spectral feature learning framework based on attentions for airborne MPC classification. A key component in our model is a two-stream feature fusion method with attention mechanisms, which enhances the representation capability of spatial-spectral features from high-dimensional heterogeneous MPCs. The first stream aims to extract position-encoded global spectral features with fusion self-attention, and the second stream comprises a multikernel point convolution and feature aggregation attention to extract spectral-guided geometric features. We then develop a residual attention fusion block to integrate the most informative geometric-spectral features from the two parallel streams. Another important contribution of this work is a joint loss function to improve the learning ability on unbalanced and interclass similar samples. Experimental results on two airborne MPC datasets demonstrate the effectiveness of the proposed method compared with the state-of-the-art methods. Furthermore, the codes and datasets used in this paper will be made available freely at https://github.com/HITlixian/TGRS_GSFF.

---

### Autonomous Incident Resolution at Hyperscale: An Agentic AI Architecture for Network Operations

Authors: Arun Malik
Published: 2026-06-08
Categories: cs.SE, cs.AI, cs.ET, cs.MA, cs.NI
arXiv: https://arxiv.org/abs/2606.09122v1

Cloud network infrastructure at hyperscale presents unique operational challenges where traditional human-driven incident response cannot keep pace with the volume, velocity, and complexity of failures. This paper presents an agentic AI architecture for autonomous incident resolution in large-scale network operations. Our system employs a multi-agent orchestration framework where specialized AI agents collaborate to detect, diagnose, and remediate network incidents without human intervention. We describe the architectural principles, including hierarchical agent decomposition, skills-based tool invocation via standardized protocols, structured knowledge encoding from operational runbooks, progressive autonomy with safety boundaries, and closed-loop verification. The architecture has been deployed in production at a major cloud provider, demonstrating that agentic AI systems can achieve autonomous resolution rates exceeding 90% for common incident categories while maintaining safety guarantees through layered authorization and rollback mechanisms. We discuss design tradeoffs, failure modes, and lessons learned from operating autonomous AI agents at scale.

---

### ComplexConstraints and Beyond: Expert Rubrics for RLVR

Authors: Sushant Mehta, Liudas Panavas, Suhaas Garre, Edwin Chen
Published: 2026-06-08
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.09118v3

Evaluation protocols can lag behind LLM capabilities. Programmatically verified benchmarks cover narrow surface constraints, whereas real-world instruction following and agentic workflows require judging semantic, contextual, and policy-dependent behavior. We study expert-curated rubric-based evaluation as a unified mechanism for measurement and reinforcement-learning rewards across two settings: complex instruction following and enterprise agentic tasks. We identify rubric-design choices that affect reward quality, including maximum viable atomicity, intent-aware criterion design, and LLM-judge calibration. We introduce ComplexConstraints, an expert-curated instruction-following suite comprising a public 75-prompt benchmark with 1,559 rubric criteria and a disjoint 1,000-prompt training set, with 10-40 atomic criteria per prompt. Empirically, rubric rewards improve training in both fixed task datasets, such as ComplexConstraints, and stateful RL environments, such as CoreCraft. Training a 4B model on ComplexConstraints improves mean criterion pass rate by +15.5 pp on a held-out split, bringing it within 0.5 pp of the untrained baseline of a roughly 60x larger Qwen3 model, and the gains transfer to external benchmarks the model never saw during training: +8.4 pp on AdvancedIF and +10.1 pp on MultiChallenge. In CoreCraft, rubric-reward RL likewise transfers to out-of-distribution benchmarks (+4.5 pp BFCL, +7.4 pp tau^2-Bench, +6.8 pp Toolathlon). These results show that expert-authored rubrics provide effective evaluation targets and scalable reward signals for improving LLM instruction following and agentic behavior.

---

### Optimizing Energy-based Neural Network Training with Coherent Ising Machine

Authors: Chen-Rui Fan, Bo Lu, Zhi-Hong Zhang, Run-Qing Zhang, Jing-Wei Wen, Chuan Wang
Published: 2026-06-08
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.09117v1

While Ising machines serve as advanced physical solvers for the Ising model,enabling applications in combinatorial optimization and neural network training,their scalability for large-scale neural networks remains constrained by hardware connectivity limitations and suboptimal training methodologies. In this work,we leverage a Coherent Ising Machine (CIM) to train an energy-based neural network using Equilibrium Propagation, achieving performance comparable to existing software-based implementations. We further enhance the algorithm by integrating the Adam optimizer to solve for the ground state of a Hopfield energy network, significantly improving convergence speed and solution accuracy. Additionally, we demonstrate the scalability of our approach across deeper network architectures and convolutional operations. Our results highlight the potential of CIM dynamics as a scalable platform for training complex neural networks, offering a pathway toward energy-efficient implementations via analog circuits, optoelectronics, or integrated photonics. This work establishes a novel physical framework for next-generation AI hardware development.

---

### Counterfactual Transport Flows for Offline Conservative Trajectory Refinement

Authors: Lena Krieger, Xuan Zhao, Zhuo Cao, Qin Wang, Hanno Scharr, Ira Assent
Published: 2026-06-08
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.09115v1

Offline reinforcement learning (RL) offers a path to policy improvement from logged data alone, using historical returns or other measurable outcomes as world feedback. A key difficulty is improving observed behavior without extrapolating beyond what the offline data supports. We propose \emph{counterfactual transport flows}, a source-conditioned trajectory refinement framework for offline decision-making guided by world feedback. Given a low-feedback candidate trajectory, we construct local preference pairs from offline data by retrieving nearby trajectories in latent trajectory space with higher task-specific feedback, and use them as weak supervision for conservative refinement. The framework learns instance-specific refinement directions: at inference time, a refinement strength parameter controls how far the candidate trajectory is transported, enabling a trade-off between preserving the original behavior and applying stronger improvement. Experiments on D4RL benchmarks, including AntMaze and MuJoCo tasks, show that our method improves behavior from historical returns as world feedback, while providing interpretable trajectory-level refinement paths.

---

### MAAM: Anchor-Preserving Compression and Contextual Calibration for Chinese Discriminatory Language Detection

Authors: Yuxin Fu, Shijing Si
Published: 2026-06-08
Categories: cs.CL
arXiv: https://arxiv.org/abs/2606.09114v1

Chinese discriminatory-language detection is challenging because harmful intent is often implicit and context-dependent. We propose MAAM (Myopia--Astigmatism Anchor Mechanism), a lightweight, model-agnostic framework inspired by functional visual blur: rather than preserving every token equally, MAAM retains discrimination-relevant semantic anchors and calibrates them with C--I--S contextual priors (Contextual Tone, Group Identity, and Stance Polarity). We also introduce ChLGBT, to our knowledge the first Chinese LGBT-focused discriminatory-language dataset, with 8,120 manually annotated samples and three ordinal labels: explicit bias, implicit bias, and emotional intensity. Across strong encoder baselines, MAAM improves all three prediction dimensions, with consistent gains in accuracy, F1, Brier score, and expected calibration error. Compared with frontier LLM baselines under zero-shot and few-shot prompting protocols, MAAM remains competitive while offering stronger compactness and stability. These results suggest that interpretable anchor preservation and contextual calibration provide a practical alternative to heavier model scaling for Chinese discriminatory-language assessment.

---

### Hybridizing Equilibrium Propagation with Ising Machines for Efficient Energy-Based Learning

Authors: Chen-Rui Fan, Bo Lu, Xing-Yu Wu, Tie-Jun Wang, Chuan Wang
Published: 2026-06-08
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.09112v1

The rapid evolution of artificial intelligence has led to substantial advances in deep neural networks. Nonetheless, conventional GPU-based training remains highly energy-demanding, motivating the exploration of physical dynamics and compatible energy-based learning schemes, such as equilibrium propagation (EP). EP-based training, however, frequently suffers from convergence to local minima due to phase-space contraction. Here we introduce an Ising-dynamics-inspired equilibrium-propagation framework in which dissipative Hopfield relaxation is replaced by an extended phase-space dynamics with conjugate variables. The resulting training paradigm keeps the local two-phase learning rule of EP while changing the physical route by which neural states reach equilibrium. We show that this dynamics lowers effective energy barriers, accelerates convergence, improves noise robustness, and trains deep convolutional Hopfield networks on MNIST, FashionMNIST, and CIFAR-10 with performance comparable to backpropagation.

---

### Illumination-Invariant Anomaly Detection for Sub-Canopy UAV Multispectral Point Clouds

Authors: Likun Chen, Yanfeng Gu, Xian Li
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09111v1

Unmanned Aerial Vehicle (UAV) multispectral point clouds (MPC) provide high-dimensional spatial-spectral data for sub-canopy target detection; however, their efficacy is significantly compromised by severe illumination heterogeneity caused by vegetation shadows. To address this, we propose a prior-free anomaly detection framework capable of robustly handling lighting variations. First, we formulate solar angle estimation as an inverse optimization problem. By coupling spectral indices with a ray-tracing model, this strategy achieves Prior-Free Shadow Extraction without relying on flight metadata, effectively distinguishing dark objects from true shadows. Second, to mitigate spectral distortions, we introduce an Illumination-Consistent Sparse Representation mechanism. Unlike standard reconstruction methods, we construct a background dictionary strictly from neighbors sharing the same illumination state. This constraint effectively disentangles spectral reflectance from lighting variations, ensuring that targets are represented solely by physically consistent background points. Experimental results indicate that the proposed method significantly improves the separability between anomalies and background in complex forest environments, demonstrating superior performance over state-of-the-art baselines. This framework is particularly suited for identifying camouflaged military targets, mapping fallen tree trunks, and uncovering archaeological ruins hidden beneath dense foliage.

---

### HDRAgent: An Agentic Framework for Multi-Exposure HDR Imaging

Authors: Weiyu Zhou, Tao Hu, Yijian Wang, Xiaogang Xu, Ruixing Wang, Qingsen Yan
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09110v1

Most existing multi-exposure HDR methods follow a fixed feed-forward reconstruction paradigm, making them prone to ghosting artifacts in complex dynamic scenes. To address this issue, we propose HDRAgent, the first agent-driven framework for HDR imaging, which adaptively selects reconstruction strategies according to the current scene conditions. Specifically, to provide scene-specific prior knowledge, we introduce a fine-grained contextual knowledge matching (FCM) module. This module leverages multimodal large language model (MLLM)-derived scene perception to retrieve relevant historical cases and tool knowledge, organizing them into structured evidence for MLLM-based adaptive tool scheduling. In addition, we propose a perception--distortion feedback mechanism that transforms post-execution quality assessment and artifact diagnosis into structured feedback, which is accumulated in historical memory to help subsequent contextual knowledge refinement and strategy selection. Furthermore, considering that extreme motion can invalidate alignment methods, we design an agent-guided generative alignment strategy that uses MLLM-based dynamic-region parsing to reconstruct unreliable contents in non-reference frames under reference-frame guidance. Experiments demonstrate that HDRAgent effectively reduces ghosting and local artifacts while achieving competitive or superior objective performance and visual quality.

---

### Driving Video Retrieval for Complex Queries with Structured Grounding

Authors: Manyi Yao, Sparsh Garg, Christian Shelton, Amit Roy-Chowdhury, Abhishek Aich
Published: 2026-06-08
Categories: cs.CV, cs.IR, cs.LG
arXiv: https://arxiv.org/abs/2606.09109v2

Video retrieval at scale is central to data curation and safety validation in autonomous driving, where users want to find not only scenes but also dynamic events such as cut-ins and hard braking. Existing vision-language and keyword-based retrieval methods often miss these events because the relevant motion may not be explicitly described in text or captured by lexical overlap. Rule-based retrieval can encode such events more directly, but it is brittle: generated or hand-written rules often fail when their assumptions do not match real driving data. We propose STRIVE-D, a data-calibrated retrieval framework for driving videos. It uses weakly labeled in-domain videos to estimate when a query rule is reliable, adapt rules that mismatch observed data, and fuse calibrated rule scores with vision-language and keyword-based retrieval signals. Across three driving benchmarks, including newly released human-annotated event data on DrivingDojo, STRIVE-D delivers up to 84% relative improvement in top-1 accuracy over state-of-the-art methods.

---

### Driving Video Retrieval for Complex Queries with Structured Grounding

Authors: Manyi Yao, Sparsh Garg, Christian Shelton, Amit Roy-Chowdhury, Abhishek Aich
Published: 2026-06-08
Categories: cs.CV, cs.IR, cs.LG
arXiv: https://arxiv.org/abs/2606.09109v1

Video retrieval at scale is central to data curation and safety validation in autonomous driving, where users want to find not only scenes but also dynamic events such as cut-ins and hard braking. Existing vision-language and keyword-based retrieval methods often miss these events because the relevant motion may not be explicitly described in text or captured by lexical overlap. Rule-based retrieval can encode such events more directly, but it is brittle: generated or hand-written rules often fail when their assumptions do not match real driving data. We propose STRIVE-D, a data-calibrated retrieval framework for driving videos. It uses weakly labeled in-domain videos to estimate when a query rule is reliable, adapt rules that mismatch observed data, and fuse calibrated rule scores with vision-language and keyword-based retrieval signals. Across three driving benchmarks, including newly released human-annotated event data on DrivingDojo, STRIVE-D delivers up to 84% relative improvement in top-1 accuracy over state-of-the-art methods.

---

### RAM: Reachability Across Morphologies

Authors: Tim Walter, Xinyu Chen, Jonathan Külz, Matthias Althoff
Published: 2026-06-08
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2606.09108v1

Many stages of the robotic lifecycle, from morphology synthesis to operation, rely fundamentally on the reachable workspace. However, current methods for approximating workspaces are slow, imprecise, or tied to a single morphology. We introduce Reachability Across Morphologies (RAM): a morphology-conditioned, implicit neural representation that acts as a fast, differentiable surrogate for pose reachability, generalising to unseen morphologies while inherently accounting for self-collisions. To train RAM, we publish a large-scale dataset of $3\cdot10^{10}$ samples generated solely from forward kinematics. Experiments show that our model achieves an $ F_1$-score of $86\%$ at nanosecond inference, outperforming the baseline by $14\%$ while reducing inference time by three orders of magnitude. We further demonstrate speed-ups of one and two orders of magnitude for gradient-based morphology and trajectory optimisation, respectively. Website: https://timwalter.github.io/ram.

---

### Graph2Idea:Retrieval-Augmented Scientific Idea Generation with Graph-Structured Contexts

Authors: Xu Li, Hanzhe Tu, Xun Han
Published: 2026-06-08
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.09105v3

Generating novel, feasible, and high-quality research ideas is an important yet challenging task in scientific discovery. Recent Large Language Model (LLM)-based methods often ground idea generation with retrieved literature, but the retrieved evidence is usually provided as flat text, such as titles, abstracts, or summaries. Such flat contexts may contain redundant or weakly relevant information, while making cross-paper relations among problems, methods, mechanisms, and findings difficult to identify and trace. To address this challenge, we propose Graph2Idea, a knowledge graph-guided framework for retrieval-augmented scientific idea generation.Graph2Idea first retrieves papers according to the input topic, transforms them into structured knowledge triples, and dynamically constructs a target-centered knowledge graph to make literature relations explicit. It then extracts compact graph-derived contexts that retain target-relevant relational evidence while reducing noisy textual input. Based on these contexts, a two-stage generation process first identifies promising research directions and then guides the LLM to synthesize candidate ideas from graph-grounded evidence. Experiments on a scientific idea generation benchmark show that Graph2Idea outperforms representative baselines under the automatic evaluation protocol. Compared with the strongest baseline scores, it improves Novelty from 0.45 to 0.52, Quality from 0.24 to 0.29, and Feasibility from 0.22 to 0.28. These results suggest that graph-structured evidence helps LLMs generate research ideas through more explicit, compact, and traceable recombination of prior scientific knowledge.

---

### Addressing Market Regime Changes and Heavy-Tailed Returns in Portfolio Optimization via Bayesian VAR and Elliptical Black-Litterman

Authors: Daniil Mikriukov, Ruoyu Sun, Angelos Stefanidis, Jionglong Su, Zhengyong Jiang
Published: 2026-06-08
Categories: cs.LG, cs.AI, q-fin.PM
arXiv: https://arxiv.org/abs/2606.09104v1

Deep reinforcement learning (DRL) frameworks for portfolio optimization have shown promise for their ability to learn allocation rules dynamically from market data. However, these models fail to account for fat-tailed returns, which characterize actual market behavior with more frequent extreme events. Furthermore, historical data is treated homogeneously, without accounting for temporal importance, leading models to fail during regime changes. We propose a new BAVAR-BLED algorithm that combines methods derived from Bayesian-Averaging Vector Autoregressive (BAVAR) and the Black-Litterman model using Elliptical Distributions (BLED) within a TD3 architecture. BAVAR captures a set of vector autoregressive representations that consider multi-scale temporal features, enabling adaptive allocation decisions based on regime-aware estimates of return expectations and dispersion matrices. These estimates serve as prior inputs to BLED, a model that uses Student's t-distributions, allowing for more realistic fat tail return estimates. The BAVAR-BLED algorithm uses transformer networks for view construction and CNNs for risk-aversion estimates, which modify dynamic allocation decisions based on market conditions. An evaluation of 29 Dow Jones Industrial Average constituents over a decade-long market period shows that BAVAR-BLED significantly outperforms state-of-the-art methods, achieving Sharpe and Sortino ratios of 1.72 and 2.70, respectively, and total returns of 57.26%.

---

### Alcmean's: Unsupervised community detection using local Laplacian, automatic detection of the number of centers

Authors: Shahin Momenzadeh, Rojiar Pir Mohammadiani
Published: 2026-06-08
Categories: cs.SI, cs.LG
arXiv: https://arxiv.org/abs/2606.09100v1

Community detection is a fundamental problem in the analysis of complex networks. It has applications across social, biological, and financial domains. Traditional algorithms such as Louvain, LPA, and modularity optimization often require manual parameter tuning. They also suffer from inaccurate cluster center selection and struggle with scalability. To address these challenges, we propose Automatic Laplacian Centrality Means (ALCMeans), a novel community detection algorithm. ALCMeans combines Laplacian energy-based automatic center identification with DeepWalk embeddings for robust node representation. Unlike existing Laplacian-based and clustering methods, ALCMeans eliminates the need to predefine the number of communities, enhances cluster center selection using structural importance, and leverages representation learning for more accurate and stable assignments. Experimental results on benchmark datasets demonstrate 10 to 20 percent higher NMI and ARI scores compared to Louvain, Newman-Girvan, LPA, Fast-Greedy, and a recent GNN-based competitor (MAGI, KDD 2024). Additional evaluations with modularity and F1-scores confirm the superiority of ALCMeans. Ablation studies highlight the critical contributions of each component. Despite its reliance on DeepWalk parameters and increased runtime relative to lightweight heuristics, ALCMeans consistently outperforms state-of-the-art methods. This makes it a promising tool for real-world network analysis.

---

### From Shortcuts to Reasoning: Robust Post-Training of Theory of Mind with Reinforcement Learning

Authors: Jike Zhong, Yuxiang Lai, Ming Li, Yuheng Li, Wuao Liu, Behzad Dariush, Konstantinos Psounis, Shao-Yuan Lo
Published: 2026-06-08
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.09092v1

Theory of Mind (ToM) is a must-acquire skill for modern foundation model systems to operate effectively and safely in the real world. Recent works have explored honing ToM via post-training; however, we show that such progress is confounded by a pervasive "shortcut" issue: tasks can reach up to 99% accuracy by simply exploiting spurious causal correlations, leading to a false sense of ToM. Motivated by this, we first develop a framework to systematically examine ToM datasets for shortcuts and provide guidance for future development. We find that questions reducible to pure state tracking, such as "belief," are especially shortcut-prone compared to mind questions, such as "intention," where reasoning beyond tracking is required. Using four shortcut-free datasets across three ToM contexts, we then comprehensively study whether Reinforcement Fine-Tuning with verifiable rewards and explicit reasoning chains, called Thinking-RFT, elevates ToM beyond Supervised Fine-Tuning, or SFT. Our key findings are as follows. First, Thinking-RFT effectively improves ToM in all scenarios, with a 6% improvement over SFT, particularly in complex higher-order reasoning, with a 10% improvement over SFT, and multimodal cases, with a 7% improvement over SFT. It also generalizes notably better to unseen domains and higher-order queries while being more robust to counterfactuals. Second, ToM benefits specifically from the joint effect of reasoning and RL: Thinking-RFT outperforms Non-Thinking-RFT by 7% on average. Third, RFT works by learning to ground its reasoning on anchor cues, such as keywords and state changes, that correspond to causal factors. We believe our study is useful for developing effective and robust ToM post-training datasets and advancing critical ToM capabilities.

---

### Stabilizing On-Policy Distillation for MLLM Reasoning with Global Normalization

Authors: Dongze Hao, Zhiwei Jin, Chen Chen, Haonan Lu
Published: 2026-06-08
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2606.09091v1

On-policy distillation (OPD) has recently emerged as an important post-training paradigm. By using a stronger teacher model to provide dense, fine-grained supervision for sampled trajectories, OPD offers a clear advantage over reinforcement learning with verifiable rewards (RLVR), which typically depends on sparse binary or outcome-based environmental feedback. However, naive token-level distillation can suffer from gradient instability, due to magnitude misalignment in outlier states. To address this issue, we propose Globally Normalized Distillation Policy Optimization (GNDPO), a practical method that stabilizes optimization by transforming raw KL scores into batch-level relative advantages. This normalization effectively mitigates gradient explosions while retaining the benefits of token-level guidance. Experimental results show that GNDPO substantially improves training robustness and downstream performance across multimodal reasoning tasks. The code is released at https://github.com/OPPO-Mente-Lab/GNDPO.

---

### Context Rot in AI-Assisted Software Development: Repurposing Documentation Consistency for AI Configuration Artifacts

Authors: Christoph Treude, Sebastian Baltes
Published: 2026-06-08
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2606.09090v1

Developers increasingly provide AI coding assistants with persistent context through configuration files such as CLAUDE.md, AGENTS.md, and .cursorrules. These files describe code elements, architecture, and development conventions, forming the context that guides AI tool behavior across sessions. As software evolves, this context can become stale, a phenomenon we call context rot. While AI configuration artifacts are new, the underlying consistency problem connects to decades of software documentation research. Researchers have built tools to check consistency between documentation and code, spanning README files, code comments, API documentation, architecture descriptions, and installation instructions. We argue that this existing toolbox is an immediate starting point for detecting context rot, and we present a research roadmap mapping documentation consistency approaches to corresponding problems in this new setting. As preliminary evidence, applying an existing README/wiki consistency checker to a statistically representative sample of 356 repositories identifies stale code element references in 23.0% of repositories, showing that traditional documentation consistency tools can already surface context rot.

---

### DynaOD: Dynamic Origin-Destination Flow Generation with Discrete-to-Continuous Temporal Semantic Modeling

Authors: Jie Zhao, Xianqi Dai, Jie Feng, Huandong Wang, Yong Li
Published: 2026-06-08
Categories: cs.AI
arXiv: https://arxiv.org/abs/2606.09086v1

Dynamic origin-destination (OD) flow generation seeks to synthesize realistic mobility dynamics from temporal context alone, without relying on historical OD observations. A key challenge is to translate semantic temporal signals into temporally coherent OD patterns while preserving the inherent spatial heterogeneity of urban regions. We propose DynaOD, a semantic-driven framework that models temporal dynamics through two complementary perspectives: discrete directional trends that characterize qualitative shifts in urban activity patterns, and continuous temporal evolution that captures how such shifts unfold over time. By jointly encoding these temporal semantics, the framework constructs time-varying region representations that condition pretrained static OD generators in a lightweight and plug-and-play fashion. This modular design further supports scalable deployment and cross-city transferability. Extensive experiments on large-scale real-world datasets show that our method consistently outperforms representative baselines in both predictive accuracy and distributional fidelity. Code is publicly available at https://github.com/csjiezhao/DynaOD.

---

### Context-Fractured Decomposition Attacks on Tool-Using LLM Agents: Exploiting Artifact Provenance Gaps

Authors: Xiaofeng Lin, Yukai Yang, Daniel Guo, Sahil Arun Nale, Charles Fleming, Guang Cheng
Published: 2026-06-08
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2606.09084v1

Tool-using LLM agents interact with the world through actions that persist state in artifacts (e.g., workspace files or logs). Consequently, jailbreak defenses must reason about cross-step composition rather than isolated text. Yet most existing attacks and defenses, including ``multi-turn'' jailbreaks such as Crescendo and Tree of Attacks,still assume a single contiguous conversation visible to the defender. This assumption breaks down in real agent pipelines, where enforcement is fragmented across tools, modules, and time, and where artifact provenance is often not tracked. We operationalize a deployment failure mode for tool-using LLM agents, the \emph{provenance gap}, and study reproducible triggers for it: \emph{Context-Fractured Decomposition} (CFD), a family of cross-context multi-step jailbreaks that preserve benign-looking intermediate artifacts from an early interaction and elicit harmful behavior much later, potentially in a different agent instance or workflow stage, via individually innocuous tool actions whose risk emerges only under delayed artifact-mediated composition. We instrument the failure mode with trace-level diagnostics and outline a verifiable mitigation direction (provenance lineage tagging). Across agent-system jailbreak benchmarks, CFD improves success rates by up to 28.3 percentage points over state-of-the-art baselines, even against strong single-turn judges. Disclaimer: This paper contains examples of harmful or offensive language.

---

### Edge-Constrained UAV Small-Object Detection with P2 Enhancement and Quantum-Inspired Lightweight Structure Search

Authors: Wuming Lei, Yanbin Gao, Mingyan Sun, Xiaobin Li, Xuechen Liang
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09081v1

Unmanned aerial vehicle (UAV) object detection requires compact detectors that retain small-object details under onboard computation and memory constraints. Repeated downsampling inlightweight networks weakens shallow spatial information, while manually adding attention orfusion modules may increase cost without stable gains. This study analyzes YOLOX-Nano underedge-deployment constraints by combining a P2 high-resolution detection branch with a quantum-inspired evolutionary algorithm (QIEA) for lightweight structure screening. The search space isdefined by lightweight priority and task specificity, and the evaluation jointly considers accuracy,floating-point operations (FLOPs), latency, memory consumption, and recall. On VisDrone, theP2 branch increases APamall by 31.10% over the YOLOX-Nano baseline. Compared with NanoDet-Plus with similar model size, YOLOX-Nano+-P2 improves APs0.ss by 17.5% and APamal by 44.9%.The QIEA-selected candidate obtains the highest Recallso, but +P2 remains the strongest AP-oriented variant after full training. Full 100-epoch verification of Random-best, GA-best, andSA/QUBO-best candidates further shows that proxy rankings do not necessarily transfer to finalAPse9s. These results support using P2 as the main small-object enhancement path and QIEA as alightweight tool for candidate screening and accuracy-cost analysis. The source code, configurationfiles, diagnostic scripts, and summarized results are available at https://github.com/Ming23233/UAV-QIEA-Edge-Detection

---

### Beyond FLOPs: Benchmarking Real Inference Acceleration of LLM Pruning under a GEMM-Centric Taxonomy

Authors: Haozhe Hu, Hao Wu, Anhao Zhao, Longwei Ding, Peiran Yin, Yunpu Ma, Xiaoyu Shen
Published: 2026-06-08
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2606.09080v2

Pruning has emerged as a dominant paradigm for accelerating large language model (LLM) inference, spanning a broad spectrum of methods that remove computation across tokens, layers, heads, dimensions, and attention patterns. Despite sharing the same objective, these pruning approaches induce fundamentally different execution behaviors, causing realized speedups to depend heavily on hardware and kernel implementations. Consequently, the practical acceleration benefits of different pruning families remain poorly understood. In this work, we introduce a GEMM-centric taxonomy that reorganizes existing pruning methods according to the logical \textbf{M}, \textbf{N}, and \textbf{K} dimensions of general matrix multiplication (GEMM). Leveraging this abstraction, we build a unified benchmarking framework that enables implementation-consistent comparison across the pruning design space and systematically characterizes the acceleration--quality Pareto frontier. Our results on Llama3.1-8B show that static depth pruning remains the strongest Pareto-optimal baseline and stays closest to its theoretical acceleration upper bound in memory-bounded scenarios. During prefill, the frontier transitions from static depth at low quality loss (0\%--4\%), to dynamic depth at moderate loss (5\%--16\%), and finally to static width pruning at higher loss levels (17\%--26\%). These findings establish the first unified view of the practical limits of pruning-based LLM acceleration and provide guidance for future pruning research. Code is available at https://github.com/EIT-NLP/LLM-Pruning/tree/main/PruningInferSim

---

### Beyond FLOPs: Benchmarking Real Inference Acceleration of LLM Pruning under a GEMM-Centric Taxonomy

Authors: Haozhe Hu, Hao Wu, Anhao Zhao, Longwei Ding, Peiran Yin, Yunpu Ma, Xiaoyu Shen
Published: 2026-06-08
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2606.09080v1

Pruning has emerged as a dominant paradigm for accelerating large language model (LLM) inference, spanning a broad spectrum of methods that remove computation across tokens, layers, heads, dimensions, and attention patterns. Despite sharing the same objective, these pruning approaches induce fundamentally different execution behaviors, causing realized speedups to depend heavily on hardware and kernel implementations. Consequently, the practical acceleration benefits of different pruning families remain poorly understood. In this work, we introduce a GEMM-centric taxonomy that reorganizes existing pruning methods according to the logical \textbf{M}, \textbf{N}, and \textbf{K} dimensions of general matrix multiplication (GEMM). Leveraging this abstraction, we build a unified benchmarking framework that enables implementation-consistent comparison across the pruning design space and systematically characterizes the acceleration--quality Pareto frontier. Our results show that static depth pruning remains the strongest Pareto-optimal baseline and stays closest to its theoretical acceleration upper bound in memory-bounded scenarios. During prefill, the frontier transitions from static depth at low quality loss (0\%--4\%), to dynamic depth at moderate loss (5\%--16\%), and finally to static width pruning at higher loss levels (17\%--26\%). These findings establish the first unified view of the practical limits of pruning-based LLM acceleration and provide guidance for future pruning research.\footnote{Code is available at https://github.com/EIT-NLP/LLM-Pruning/tree/main/PruningInferSim}

---

### FlashMemory-DeepSeek-V4: Lightning Index Ultra-Long Context via Lookahead Sparse Attention

Authors: Yan Wang, Qifan Zhang, Jiachen Yu, Tian Liang, Dongyang Ma, Xiang Hu, Zibo Lin, Chunyang Li, Zhichao Wang, Miao Peng, Nuo Chen, Jia Li et al.
Published: 2026-06-08
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2606.09079v3

Conventional LLMs keep the full KV cache loaded during decoding, causing a severe GPU memory bottleneck for ultra-long context serving. In this report, we propose \textbf{Lookahead Sparse Attention (LSA)}, a novel inference paradigm powered by a Neural Memory Indexer built upon the DeepSeek-V4 architecture. Rather than passively attending to all historical tokens, LSA proactively predicts future context demands and preserves only the query-critical KV chunks in the GPU memory. Crucially, we instantiate this architecture via a \textbf{backbone-free decoupled training} strategy. By formulating the indexer as a standard dual-encoder architecture, we train it independently using standard retrieval training frameworks without ever loading the massive backbone model into GPU memory. We demonstrate that this ``less is more'' paradigm significantly maximizes serving efficiency while acting as an effective attention denoiser in tasks that rely on long-term global memory. Across primary long-context evaluation suites (e.g., LongBench-v2, LongMemEval, and RULER), \texttt{FM-DS-V4} compresses the average physical KV cache footprint down to merely 13.5\% of the full-context baseline, while consistently preserving or slightly elevating downstream accuracy (+0.6\% absolute margin on average). At 1M context, per-decode-token compute drops to 0.30$\times$ of the baseline and GPU KV cache shrinks by 90\% (3.73$\to$0.37 GB), translating into \textbf{2.8$\times$ aggregate throughput and 2.7$\times$ concurrency gains} in PD-disaggregated serving on 8$\times$H20 GPUs.

---

### The Hidden Bias of Process Reward Models:PRISM for Rewarding the Right Reasoning

Authors: Aakriti Agrawal, Souradip Chakraborty, Armin Saghafian, Nihal Sharma, Rizal Fathony, Nam H Nguyen, C. Bayan Bruss, Amrit Singh Bedi, Furong Huang
Published: 2026-06-08
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.09078v1

Process Reward Models (PRMs) improve credit assignment for reasoning by providing step-level feedback. However, we identify a hidden bias in PRMs caused by severe imbalance in step-level training data. Standard cross-entropy training amplifies this bias, causing PRMs to overcredit plausible but incorrect steps and produce high false-positive rates. We show that these false positives have an asymmetric downstream effect: false negatives mainly slow exploration, whereas false positives actively steer Best-of-N selection, guided decoding, and policy optimization toward flawed reasoning. This suggests that PRM training should shift from pointwise label fitting to reliable relative comparisons. To address this, we propose PRISM (Precision Ranking for Improved Step Modeling), a policy-aware PRM training framework that learns from contrastive step-level comparisons and hard negatives generated by a temporal lookahead strategy, requiring no new human labels. We further use a difficulty-aware curriculum to optimize the contrastive step margin. Across PRMBench and ProcessBench, PRISM substantially reduces false positives (22% on PRMBench) and improves macro F1 over strong discriminative PRMs. When applied to policy optimization and search tasks, including guided decoding and Best-of-N selection, it consistently improves accuracy (up to 22% for guided decoding and 33% for Best-of-N) and robustness. More broadly, trustworthy process supervision is not just about assigning high rewards, but about rewarding the right reasoning for the right reasons.

---

### Neural Legendre-Fenchel transform with Hessian Preconditioning

Authors: Basile Plus-Gourdon, Frank Nielsen
Published: 2026-06-08
Categories: cs.LG
arXiv: https://arxiv.org/abs/2606.09077v1

The Legendre-Fenchel (LF) transform is a fundamental tool in convex analysis and machine learning that maps lower semi-continuous functions to their convex conjugates. In practice, when closed-form formula are not available for expressing convex conjugates of given functions, one must approximate them using various techniques. One recent such versatile numerical method is the deep Legendre transform method which relies on neural networks although it remains challenging particularly for tackling ill-conditioned functions. This work builds on the reformulation of the LF transform as a projective polarity. A notable property of this framework is its affine invariance. We leverage this affine invariance to introduce a Hessian-based preconditioning strategy. Specifically, we apply an affine deformation around a minimizer so that the second-order Taylor approximation of the function coincides with the canonical paraboloid, whose conjugation map is the identity. A residual network initialized near the identity can then learn this simplified mapping, while the original conjugation map is recovered through the inverse deformation. The proposed preconditioning incurs only a modest computational overhead, consisting of a single eigendecomposition during initialization and two matrix-vector multiplications per query. Experiments on a diverse set of convex functions, including high-dimensional benchmarks, demonstrate improved convergence rates and enhanced numerical accuracy of the conjugation, with particularly significant gains for ill-conditioned problems. Finally, we discuss the scope of applicability of our proposed method and highlight several of its limitations.

---

### Z-Reward: Beyond Scalar Rewards by Internalizing Reasoning into Score Distributions

Authors: Xin Jin, Huanqia Cai, Zhen Li, Zechao Zhan, Dengyang Jiang, Aiming Hao, Yuming Jiang, Xiangpeng Yang, Chunle Guo, Peng Gao, Ming-Ming Cheng, Steven C. H. Hoi
Published: 2026-06-08
Categories: cs.CV
arXiv: https://arxiv.org/abs/2606.09076v3

Reward models are central to text-to-image post-training, but visual preference is subjective and better represented as a distribution over rubric scores than as a deterministic scalar. Existing scalar, score-token, and pairwise reward models over-compress uncertainty and fine-grained score differences, while reasoning-based generative rewards provide stronger judgments but are costly to deploy and difficult to use as direct optimization signals. We propose Z-Reward, a teacher-student reward modeling framework that decouples reasoning-heavy judgment from efficient reward deployment. The teacher is a large VLM that uses reasoning to infer rubric-aligned score distributions, and is trained with Group-wise Direct Score Optimization (GDSO), which combines policy-gradient rewards from distribution expectations with direct pointwise and pairwise supervision on score distributions and score gaps. The student is trained with Reasoning-Internalized Score Distillation (RISD), which transfers the teacher's reasoning-conditioned score distribution into a compact VLM without requiring explicit reasoning chains at inference time. On our internally annotated evaluation set, the 27B GDSO teacher reaches 89.6% human preference accuracy, outperforming SFT, RewardDance, and GRPO, while the 9B RISD student reaches 88.6%, outperforming the OPD baseline and closely matching the larger teacher. We further show that Z-Reward can serve as a differentiable reward signal for text-to-image optimization, yielding a 41.3% net human-preference improvement over the SFT baseline.

---