Attention
-
Multi-Scale Latent Temporal Compression Attention and Video-Text Cross Attention Fusion for Video Action Detection
2026-10-21 04:08:49Yicheng Qiu, Keiji YanaiAbstract
In video action detection, modeling the temporal dependencies in continuous video streams is crucial. Previous methods based on deep neural networks can perform frame-level video recognition tasks. However, these methods struggle to model dynamic temporal patterns, facing challenges such as redundant frame information, difficulty balancing global and local features, and insufficient fusion of visual and textual features. To address these issues, we propose a video temporal attention module based on the Multi-Scale Latent Temporal Compression Attention, which aggregates continuous temporal sequences into feature representations across diverse temporal scales while efficiently reducing redundancy. We also introduce a Video-Text Cross Attention Fusion module that leverages cross-attention at both latent and standard feature levels, achieving fine-grained fusion between video and textual features to enhance semantic alignment and feature interaction. Experimental results across multiple video recognition benchmarks demonstrate that our proposed methods achieve superior performance compared to existing methods, validating the effectiveness and generalization capability.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29176
PDF
-
XingChen-UniFE: A Unified Generative Speech Front-End via Complementary Acoustic-Semantic Modeling
2026-10-21 03:55:18Kelan Kuang, Qihan Hu, Jian Kang, Jie Li, Yongxiang Li, Yi LiuAbstract
This paper presents XingChen-UniFE, a unified large audio language model (LALM) for general speech restoration (GSR) and target speaker extraction (TSE). The proposed model employs complementary semantic and acoustic representations through a dual-stream encoder, where a self-supervised learning model provides hierarchical semantic features and a learnable acoustic branch captures fine-grained structures from complex spectra. A bidirectional cross-attention module further facilitates information exchange between the two representations. The fused representation conditions the LALM for discrete speech token generation with task and speaker information. Experiments demonstrate strong performance on both GSR and TSE and robust generalization to unseen degradation conditions.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29164
PDF
-
SICAF: Time-Varying Focus Bottleneck for Self-Supervised Event-Based Optical Flow with Spiking Neural Network
2026-10-21 03:48:08Shuangming Yang, Qing He, Shangqi Guo, Badong ChenAbstract
Event-based optical flow estimation achieves high temporal resolution and dynamic range by exploiting asynchronous events, while self-supervised learning provides an effective solution to the absence of ground-truth labels. Spiking neural networks (SNNs), as brain-inspired models, process information through asynchronous spikes, naturally aligning with optical flow events. However, the temporal irregularity across event streams hinders the compact extraction of spatiotemporal features, while the absence of explicit labels leads to spurious correlations, thereby increasing redundancy within latent representations. In this paper, we propose a novel 'Compactly Compress, Time-varying Focus' training strategy. We present SICAF (Spikes Information Compression And Focus) framework that formulates the learning process as a constrained optimization with temporal modulation. It encourages self-supervised SNNs to extract compact latent representations while preventing over-compression that would discard temporal saliency. To better enable feature selectivity across time steps, we design an estimator with learnable temporal attention, achieving context awareness with cross-time dependencies. Experimental results on several datasets demonstrate that SICAF achieves state-of-the-art performance in self-supervised SNNs, with a 8.14% improvement in AEE on the MVSEC dataset and a 13.49% increase in robustness under white-box adversarial attacks.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29158
PDF
-
DGCM-SSFA: Dual-Granularity Cross-Modulation and Sparse Strip Frequency Attention for Efficient 3T-to-7T MRI Synthesis
2026-10-21 03:21:45Zhichun Fan, Zhiqiang Chen, Lanling Liu, Lingjun Liu, Zhonghua XieAbstract
Synthesizing high-fidelity 7T MRI from routinely acquired 3T scans offers a practical pathway to enhanced neuroimaging without the prohibitive cost of ultra-high-field scanners. While existing deep learning approaches have achieved promising results, state-of-the-art methods typically rely on either large-scale architectures exceeding 45 million parameters or computationally intensive ensemble strategies, limiting their practicality in resource-constrained clinical environments. In this work, we propose a compact multi-domain attention framework, termed DGCM-SSFA, that achieves competitive synthesis quality with only 11M parameters. The proposed architecture integrates two complementary modules into a U-Net Transformer backbone. First, a Dual-Granularity Cross-Modulated Feature Modulator (DGCM) is embedded at multiple hierarchical levels to enhance local edge and texture details through Sobel-guided spatial attention and depthwise frequency-aware channel attention with cross-branch gating. Second, a Sparse Strip Frequency Attention (SSFA) module at the bottleneck captures global spatial context via strip-pooled attention with near-linear complexity, while simultaneously performing FFT-based spectral modulation in the frequency domain. These two modules operate synergistically with the backbone's native channel attention, forming a triple-granularity feature refinement pipeline. Extensive experiments on the UNC and BNU 3T-to-7T MRI datasets across T1w and T2w modalities demonstrate that DGCM-SSFA achieves strong structural fidelity, attaining an SSIM of 0.8431 on BNU T2w---approaching the best benchmark value of 0.8624---and 0.8100 on BNU T1w, while using substantially fewer parameters than Transformer-based alternatives such as ResViT (123.4M) and FS-RWKV (45.4M).
Abstract (translated)
URL
https://arxiv.org/abs/2610.29141
PDF
-
Copying Under Repetition: From Prefix Matching to Position Hashing
2026-10-21 02:21:26Anxin Guo, Tianlong Huang, Chenxiao Yang, Zhiyuan LiAbstract
Frontier LLMs can copy random text almost flawlessly, yet they often overrun or truncate long blocks of repeated characters. In repeated strings, matching local content no longer reliably identifies where copying should continue or when it should stop. Nevertheless, we show that a two-layer transformer with $O(\log\log N)$-bit precision and $O(\log N/\log\log N)$ width can copy every string up to a chosen design horizon $N$ and terminate correctly under arbitrary repetition. We use RoPE-style rotations of attention values and outputs (VO-rotations) with small prime periods to encode distances from delimiter anchors modulo those primes. By the Chinese Remainder Theorem, these periodic encodings jointly distinguish all required distances, allowing one retrieval rule to select either the next source token or the end marker. In contrast, we prove a width-independent impossibility result for copying under sequential finite-precision softmax accumulation. It applies to transformers whose only positional input is an additive attention bias that eventually becomes constant, including NoPE and a recursive implementation of ALiBi. Once an identical-token run exceeds the resulting saturation threshold, further repetitions leave the logits unchanged. If this threshold grows sublinearly with $N$, these models cannot both complete a long all-identical copy and stop correctly, a pattern resembling the termination errors of frontier LLMs. In synthetic copying experiments, baseline transformer models fail on long repeated strings even within the training range. Trained VO models, by contrast, generalize beyond the training lengths with near-perfect accuracy up to length $\approx5005$, the selected primes' period product.
Abstract (translated)
URL
https://arxiv.org/abs/2610.26722
PDF
-
PHD-Hand: A Printable Humanoid Dexterous Hand with Linkage-Driven 3-DOA Digits
2026-10-21 01:01:57Neel Adwani, Muhaiminul Islam Akash, Saffia Ouali, Ethan Liu, Cong WangAbstract
Currently, dexterous humanoid robot hands are widely considered the final hardware bottleneck that hampers generalist robots from physically assisting people at home and work. As conventional tendon-driven designs suffer from high cost, demanding maintenance, poor reliability, and short durability, alternative linkage-driven designs have increasingly gained attention. Despite their potential to avoid the weaknesses of tendon-driven hands, integrating multiple degrees of actuation (DOA) in a single digit with linkage drivetrains is a challenge for design and fabrication. Only a few studies have managed to present functional humanoid hands that are completely linkage-driven and have at least three DOAs per digit without compromising actuator placement and other design preferences. Our work introduces a new linkage-driven hand featuring a unique flexure hinge. Together with elastic drive rods made of spring steel, the linkage drivetrains are significantly simplified. Unlike other linkage-driven designs of similar dexterity, our design does not require many miniature moving parts. As a result, the hand can be fabricated easily with a budget 3D printer. This not only significantly lowers the barrier to entry for using and developing dexterous humanoid robot hand functions, but also allows the hands to be sized much closer to regular human hands. The design offers desirable kinematic advantages as well. A prototype is evaluated for grasping and object manipulation. In addition, this paper presents our latest findings on kinematics and control of the hand.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29030
PDF
-
Enhancing Cross-Grade Generalization in Banana Variety Classification via Hybrid CNN-Transformers and Knowledge Self-Distillation
2026-10-21 00:19:09Khush Kataruka, Mohammad Muzammila, Zahid Akhtar, Raghavendra Ramachandra, Praveen Kumar ChandaliyaAbstract
In recent years, automated fruit classification systems have become a more critical component in precision agriculture for efficient yield estimation, quality control, and supply chain optimization. However, they still face significant challenges under varying quality, fruit maturity, illumination, texture degradation, morphological attributes, domain shift, and post-harvest handling conditions. In this study, we address the challenging issue of cross-grade banana variety classification. This is a real-world setting where models that have been trained on specific quality grades must generalize to unseen grades to handle significant domain shift. Under a rigorous Leave-One-Grade-Out evaluation protocol, we present a thorough comparative analysis of CNNs, ViTs, and dual-stream hybrid architectures using a large-scale dataset of 31,678 banana images with three varieties and three quality grades. Specifically, to address the drawbacks of single-paradigm frameworks, we present a dual-stream Swin ResNet hybrid framework enhanced with KL-divergence-based self-distillation to align probabilistic feature representations across the CNN and Transformer streams. Comprehensive experimental analyzes show that the presented hybrid framework with KL-divergence based self-distillation achieves better overall performance, i.e., it attains a mean accuracy of 81.25% and significantly surpasses CNN and Transformer baseline models. Qualitative analysis further confirms that the hybrid framework in this study improves discriminative feature transferability, improves attention localization, and stabilizes optimization.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29008
PDF
-
FlashNext-RL: A System Optimization Study for Long-Horizon Agentic RL of Qwen3.8-Flash-Next
2026-10-20 23:53:09Zhewei Yao, Yuxiong HeAbstract
Long-horizon agentic reinforcement learning (RL) requires both efficient rollout generation and training on long sequences. However, the memory requirements and model-specific operators of Qwen3.8-Flash-Next make this difficult with existing open-source implementations. To resolve these limitations, we introduce FlashNext-RL, a model-specific RL training optimization framework based on prime-rl, with vLLM for software engineering (SWE) rollout generation. Our main focus is BF16 training of the full 48-layer model on a single 8xB200 node, with the first two layers, the n-gram table, the sparse-attention indexer and the multi-token prediction head frozen. We optimize host offload, gradient accumulation and the kernels for mixture-of-experts, sparse attention and the gated residual stream. In the full-model development study, non-padding training throughput increases from approximately 395 tokens/s/GPU in the initial 32K-token configuration to 4,438 tokens/s/GPU in the optimized 131K-token configuration, with hardware matmul throughput (aka without considering the cost of attention components) increasing from approximately 20 to 213 TFLOPs/GPU. A controlled cumulative study on an 8-layer model further quantifies the contributions of individual optimizations. Together with three inference nodes, the system supports a live SWE RL run with a target batch of 1,024 trainable rollouts and a 131,072-token context limit. Over steps 5-15, an iteration takes 22.4 minutes on average, with bounded entropy and trainer-sampler mismatch KL. These results highlight the importance of tailoring RL training to the model's architecture, memory layout and kernel requirements, rather than relying only on general-purpose RL infrastructure.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28998
PDF
-
NV-Generate-Body: 3D Latent Diffusion Model forControllable Anatomy, Tumor, and Demographic Body Mask Synthesis
2026-10-20 23:01:01Can Zhao, Pengfei Guo, Dong Yang, Yufan He, Yucheng Tang, Ahmed Harouni, Benjamin Simon, Mason Belue, Stephanie Harmon, Baris Turkbey, Daguang XuAbstract
Curating large labeled CT datasets is expensive and constrained by patient privacy. We present NV-Generate-BodyMask, a 3D latent diffusion model that synthesizes whole-body CT segmentation masks with precise, joint control over organ anatomy, tumor burden, and patient demographics. The model operates in the compressed latent space of a frozen mask autoencoder, generating $256^3$ volumetric label maps conditioned on a 19-dimensional vector via Adaptive Group Normalization and cross-attention. A size-adaptive ROI loss with spatial halo weighting and a paired tumor injection pipeline enable accurate tumor size conditioning despite the scarcity of annotated tumor cases. Trained on 8,571 masks across 26 cohorts spanning 132 labels, the model achieves Spearman $\rho = 0.70$--$0.92$ for organ size, 85--100\% tumor recall for lung, pancreatic, colon, and bone lesions, and strong demographic alignment (BMI $\rho = 0.78$, sex-conditioned perfect prostate control).
Abstract (translated)
URL
https://arxiv.org/abs/2610.24153
PDF
-
Building AI Research Teams for Autonomous Discovery of Linear Attention Architectures
2026-10-20 20:09:31Yujian Liu, Li An, Zelin He, Jacob Helwig, Jiabao Ji, Rohit Jain, Siyu Zhu, Shiyu ChangAbstract
Autonomous agents can increasingly propose ideas, modify code, and run experiments, creating an opportunity to accelerate scientific discovery using teams of collaborative AI researchers. To escape local minima in the discovery, we propose to encourage orthogonal exploration by constructing the team using agents with varying expertise. However, realizing this design requires more than diverse backgrounds. When experiments are costly, a team can test only a limited number of ideas, thereby creating the need to balance exploration and exploitation when selecting a subset of researchers to propose and execute new experiments. Furthermore, each result should shape what the team tries next. We therefore ask how AI researchers with different scientific backgrounds can build on one another's findings and sustain progress under a limited experimental budget. To address these challenges, we build a team of persistent AI researchers, each grounded in a different discipline, who keep private research histories, share experimental evidence, and receive research opportunities through adaptive allocation. We demonstrate the effectiveness of this design by deploying the team to search for linear attention architectures, where it outperforms GDN2 at 1.3B scale on language modeling, common-sense reasoning, and synthetic and real-world retrieval tasks. In controlled comparisons, removing any single organizational component from our design yields markedly weaker discoveries: without disciplinary backgrounds, the search plateaus early; without private histories, researchers make less progress despite identical shared evidence; and uniform selection falls short of adaptive allocation within the same budget. These results suggest that sustained autonomous discovery depends not only on stronger agents, but also on who explores, what they remember, and how research effort is allocated.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28887
PDF
-
Joints, Camera, Action! Co-Generating 3D Motion and Video for Arbitrary Rigs
2026-10-20 17:59:56Noam Rotstein, Yoad Tewel, Chen Tessler, Haggai Maron, Gal Chechik, Or Litany, Yoni Kasten, Yuval AtzmonAbstract
Video generation models can now depict remarkably realistic motion. Yet it is unclear how to use this motion beyond pixels: to animate a specific 3D asset and replay the animation in its 3D scene. For instance, a video model can show a monkey jumping over a crossbar, but it cannot return that jump as replayable 3D motion placed in the 3D scene. Here we present a model that directly co-generates the video and 3D motion for any supplied rig, along with a camera that registers this motion to the scene. Prior video-motion models either handle only human skeletons, often adding billions of parameters, or triangulate 3D motion from generated 2D pose maps. We instead keep 3D motion in a separate model that supports arbitrary skeletons, and connect it to a pretrained video model through a small set of cross-attention layers. This separation lets the motion model first learn how skeletons move from motion-only data, which is easier to obtain and augment than paired video. Additionally, our model learns to locate each joint in the video frames, which lets us recover the camera and place the 3D motion in the scene. To train across many skeletons, we use a generative video editor to place animated 3D assets in scenes suited to their actions, curating the edits so their known motion remains valid. Together with human and robot video, this yields nearly 200K training pairs. Our evaluations demonstrate that our model animates diverse rigs from text more faithfully than video-first pipelines, and its motion responds to the surrounding scene. It also turns video into rig motion, rig motion into video, and enables keyframe-based control of the 3D motion. Videos and additional results are available on our project page: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28632
PDF
-
Turning TS2Vec into a Foundation Model for Time Series Classification
2026-10-20 17:47:51Vasilii Feofanov, Zhenghan Tai, Andrei Ermakov, Irina RishAbstract
TS2Vec is one of the strongest self-supervised methods for time series classification, introduced in 2022. It is designed to be trained anew on every dataset: self-supervised pre-training followed by linear probing. Since then, the field has shifted towards foundation models that reuse one encoder across datasets. Yet TS2Vec offers a compact convolutional architecture that operates directly on temporal samples and accepts variable-length inputs without resampling. We ask whether these properties can be retained when the encoder is pre-trained only once, and show that they can: TS2Vec can essentially be turned into a foundation model. Three changes suffice: pre-train on a synthetic corpus, feed the encoder a raw amplitude view and an instance-normalized shape view of the input through two small stems, and distill one specialist encoder per view into a single encoder. The result, TS2Vec-FM, is a 1.6M-parameter dilated convolutional network that, kept frozen, reaches 83.3% on UCR-128 and 75.3% on UEA-27 under linear probing, within 0.3 pp of MantisV2, the current state of the art, and above TS2Vec trained separately on each target dataset. What we find most interesting is how different TS2Vec-FM is from other foundation models in design: near-parity is reached without attention, without a patch tokenizer, and without intermediate-layer extraction, using an "old-school" convolutional encoder and a two-layer MLP stem.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28602
PDF
-
WildMAE: Structure-Aware Masked Autoencoding for Wildlife Image Reconstruction
2026-10-20 17:41:31Yicheng Zhang, Yujie Jin, Ke Zhang, Jiaxin WuAbstract
Camera-trap imagery presents substantial variation in animal appearance, illumination, background clutter, and visual degradation. Masked autoencoders (MAEs) offer a label-free framework for modeling such data through reconstruction of randomly hidden image patches. We present WildMAE, a decoder-side extension of MAE that combines local structure-aware refinement with masked-to-visible cross-attention. A zero-gated convolutional residual path introduces a two-dimensional neighborhood prior, while gated cross-attention allows masked positions to retrieve contextual information from visible-position features without replacing the original Transformer decoder. The associated training recipe progressively introduces normalized-pixel, Sobel-edge, and perceptual objectives. In the original Snapshot Serengeti study, covering 4,744 test images at 75% masking, the reported reconstruction scores were 27.90 versus 24.33 dB PSNR, 0.840 versus 0.780 SSIM, and 0.160 versus 0.320 LPIPS for WildMAE and vanilla MAE, respectively. Additional historical records document favorable average differences under four synthetic degradation conditions and describe component variants. We present the method, experimental observations, and their scope together: the retained metrics are historical protocol-dependent results, rather than a newly reproduced, harmonized comparison or a validation of restoration from genuinely unavailable pixels. WildMAE provides a practical architectural case study for combining global token mixing with spatial and targeted contextual refinement in wildlife masked-image modeling.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28590
PDF
-
Cache Engineering for Agents
2026-10-20 17:32:24Tianyu Fu, Ruoxi Wang, Zihan Min, Zhongqi Ren, Zhihao Li, Tengxuan Liu, Shengxu Mu, Hanling Zhang, Yu WangAbstract
Large Language Model (LLM) agents can handle increasingly complex tasks, navigating long trajectories of observations and actions. As these trajectories grow, key-value (KV) cache I/O and storage become the dominant cost. Trajectories can also exceed the LLM's maximum context length, requiring control over active cache growth. Prior work uses context engineering to reduce cache size by trimming or summarizing history. However, modifying the trajectory prefix invalidates subsequent cached states and triggers costly re-prefilling of retained context. Task accuracy can also suffer when too little context is retained to limit re-prefill costs. We propose cache engineering as a new scheme. It allows agents to directly control their active KV cache without modifying the trajectory or re-prefilling retained context. Specifically, agents can proactively use drop to exclude selected cache from future attention and repos to adjust retained keys' position encodings. We formalize these cache operations and identify history dependence as a serving challenge: identical token sequences can yield different cache states under different cache operation sequences, making prefix matching insufficient for correct cache reuse. Thus, we co-design the AKV serving engine. It provides a stateless interface compatible with existing serving APIs, while internally managing stateful cache reuse, eviction, and restoration. Experiments show that a tool-response rolling-drop cache policy improves average benchmark scores by 1.7 points and reduces average cost by 27.7%, compared with conventional threshold-triggered summarization. With AKV, it achieves up to 1.2x the output throughput and a 1.5x speedup in P95 request latency. Recursive self-improvement of cache policies further reduces cost by 7.1% over expert design. Our code is at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28577
PDF
-
Large Flow Language Models
2026-10-20 17:20:29Arnav Chavan, Chanhyuk Lee, Udbhav Bamba, Jaehoon Yoo, Seunghoon Hong, Aditi Raghunathan, Jinwoo KimAbstract
Continuous flow language models offer a path to parallel generation with tools for control and acceleration unavailable to discrete diffusion, yet effective techniques for building them at the billion-parameter scale remain less established. We present Lafa (Large flow language model), a fast, high-performing flow language model scaled to 8B parameters in a data-efficient manner. Lafa is a blockwise flow model with causal attention on the prefix and bidirectional, asynchronous flow within each block. This design enables efficient training by adapting pretrained autoregressive (AR) models and fast inference via prefix caching and blockwise parallel decoding. We train Lafa at 0.5B, 7B, and 8B scales with fewer than 30B tokens, more efficient than previous diffusion models, and evaluate it on instruction following, reasoning, and coding, where the learned flow matches or surpasses the quality of AR and can be distilled into flow maps for one- and few-step generation. Notably, flow training preserves or even improves pretrained AR capabilities via a multi-token prediction effect, which we leverage in a self-speculative decoding scheme that verifies flow-based drafts autoregressively within a single model. This yields 2.54x the throughput of an AR model at comparable quality, outperforming multi-token prediction and discrete diffusion methods. Our findings show that continuous flows can serve as a scalable foundation for language modeling.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28559
PDF
-
Translation and data ethics
Abstract
This chapter introduces data ethics for translation and discusses a number of ethical issues regarding data. These include issues of copyright, ownership and bias, which apply to all types of data, whereas there are further issues particular to minoritised language communities and those on either side of the digital divide, as data is implicated in power imbalances and dynamics. We close with a discussion of responsibility and regulation, and reflect on the growing attention on data ethics from translation studies researchers.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28710
PDF
-
Asymmetric Looped Transformers: Fixed Global Representations for an Iterative Decoder
2026-10-20 16:36:33Cheng Liang, Tengfei Zhang, Wenjie DU, Jianyu Wu, Changlian Ma, Shuchen Pu, Ya Zhang, Yanfeng Wang, Zhouhan LinAbstract
Looped Transformers increase effective depth by repeatedly applying shared layers, offering a way to scale computation without increasing parameter count. However, full-stack looping places context construction and iterative state refinement within the same recurrent computation, without explicitly distinguishing their roles. We introduce the Asymmetric Looped Transformer (AsymLoop), which separates single-pass context encoding from recurrent state refinement. A single-pass causal encoder produces layer-specific Global key-value (KV) caches that remain fixed across iterations, while a weight-tied recurrent decoder generates Local KV from its evolving state. Joint attention over both sources under a single softmax allows each update to integrate persistent contextual information with the current intermediate state. For training, we apply a heavy-tail-guided layerwise learning-rate method to the encoder and decoder. Experiments on FineWeb-Edu across four model scales, from 60M to 1B, show consistently lower validation perplexity and higher average downstream accuracy than Full Loop under matched effective depth and training token budgets. At 1B, AsymLoop improves average zero-shot accuracy across seven commonsense reasoning benchmarks from 49.58% to 50.85%. Further experiments extend these benefits to longer training budgets, an OLMo-style backbone, and sparse mixture-of-experts models. These results support asymmetric context encoding and state refinement as an effective design for recurrent language models.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28489
PDF
-
Claims of emergence in AI should be grounded in information decomposition
2026-10-20 16:29:28Christoph Riedl, Fernando E. RosasAbstract
When a language model develops capabilities not traceable to its individual components, when a multi-agent AI system outperforms each agent on its own, or when a human-AI pair succeeds where either fails in isolation, AI researchers describe it as a case of ``synergy'' or ``emergence''. However, these are usually informal assessments not supported by a theoretical framework, making these claims impossible to compare or even meaningfully dispute. Here we argue that \emph{information decomposition can serve as a unifying formal framework to operationalize assessments about synergy in AI systems}. Information decomposition provides a way to distinguish qualitatively different \emph{types} of information, including unique information available from individual components, redundant information shared across multiple components, and synergistic information accessible only from parts taken together. This gives claims of emergence a precise, falsifiable criterion: a system's behavior is emergent only if it carries synergistic information about task-relevant outcomes. We apply this lens to attention heads, autonomous agents, and human-AI teams, showing how it organizes existing knowledge across scales and substrates, while also generating cross-level conjectures that siloed fields would miss.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28479
PDF
-
HGASurv: Hierarchical Granularity-Aware Multimodal Integration for Survival Prediction
2026-10-20 16:17:16Peike Yu, Wuchao Liu, Jihong Guan, Shuigeng ZhouAbstract
Multimodal survival prediction jointly analyzes histopathological whole-slide images (WSIs), genomic profiles, and clinical variables to stratify patients by risk. However, existing approaches often inadequately account for granularity differences across modalities, particularly in spatial organization and feature dimensionality. Within pathology, aggregating WSI patch features without preserving their spatial arrangement can obscure tissue architecture; across modalities, directly concatenating low-dimensional clinical variables with high-dimensional visual and genomic features may limit the contribution of clinical information to prediction. To address these challenges, we propose HGASurv, which introduces a hierarchical WSI encoding module and a clinical-guided graph propagation module for granularity-aware multimodal integration. The hierarchical WSI encoding module maps WSI patch features onto a spatial grid using their original coordinates and applies parallel convolutions to construct representations at patch, local-neighborhood, and regional scales. Genomic signature embeddings then query these scale-specific WSI representations through cross-modal co-attention, producing signature-conditioned morphological features. The clinical-guided graph propagation module uses clinical variables to define patient-similarity edges and propagates fused image-genomic node embeddings with graph attention, integrating clinical information through topology rather than feature concatenation. Across three TCGA cohorts, HGASurv achieved the strongest overall C-index performance, with ablation studies supporting the complementary contributions of the hierarchical WSI encoding and clinical-guided graph propagation modules. This design shows that matching representation and integration mechanisms to modality-specific granularity improves multimodal survival prediction.
Abstract (translated)
URL
https://arxiv.org/abs/2610.26357
PDF
-
Stratified Sparse Attention: Reducing the Sparse-to-Dense Gap at Extreme Sparsity
2026-10-20 16:02:12Sahil Joshi, Prithvi Dixit, Aryan Katiyar, Agniva Chowdhury, Joseph E. Gonzalez, Anshumali Shrivastava, Aditya Desai, Kumar Krishna AgrawalAbstract
Large Language Model inference over long contexts is increasingly bottlenecked by the cost of attending to a growing Key-Value (KV) cache. Sparse attention reduces this cost by accessing only a fraction of the cache, but existing methods face a trade-off between deterministic retrieval, which captures high-scoring tokens but discards residual attention mass, and sampling, which improves coverage at the cost of higher variance. Hybrid methods combine top-$k$ retrieval with uniform sampling, yet still exhibit a substantial gap from dense attention at aggressive sparsity. We introduce \emph{Stratified Sparse Attention (SSA)}, a unified framework combining deterministic retrieval with score-guided sampling. SSA deterministically attends to high-scoring tokens while sampling the residual using proxy attention scores and correcting for their inclusion probabilities, thereby recovering attention mass beyond the retrieval set. SSA encompasses top-$k$, top-$p$, and hybrid top-$k$+sampling as special cases. Across RULER, LongBench, LOFT, and AIME, and five models ranging from 3B to 27B parameters, SSA consistently improves over the baselines at matched sparsity. On RULER-HARD with Ministral-3-8B, SSA improves accuracy over the strongest practical baseline from 81.8\% to 84.2\% at 32K and from 63.8\% to 69.3\% at 128K under $50\times$ attention sparsity, recovering 94\% and 85\% of dense-attention performance, respectively. With optimized CUDA kernels, SSA achieves up to $2.96\times$ decode speedup over FlashInfer at 1M context, demonstrating that its accuracy gains translate into practical long-context inference acceleration. We open-source our code at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.28441
PDF
