Starlink and other low Earth orbit (LEO) satellite broadband systems exhibit stochastic access dynamics whose effects propagate from network measurements to throughput prediction and adaptive bitrate (ABR) streaming, making reasoning over multiple plausible futures essential. Existing studies mainly characterize realized traces or individual system layers, leaving cross-layer world modeling of these evolving dynamics largely unexplored. To address this gap, we propose StarWorld, a cross-layer world model for Starlink prediction and streaming. StarWorld couples learned stochastic network dynamics with executable prediction, control, and playback processes to imagine multi-step cross-layer futures, while a calibrated Trust Gate determines whether these futures are sufficiently reliable for use. On real Starlink measurements, StarWorld achieves the lowest network-future distributional errors among the comparison methods at the principal horizons and reduces pairwise controller quality-of-experience (QoE) effect error by 76%. These results support StarWorld as a cross-layer imagination environment for studying prediction and streaming outcomes under uncertain Starlink access conditions.
Diagnosis by Composition: Detecting and Localizing Unseen Diseases by Composing Learned Radiological Findings
Kaihua Hou, Wei Liu, Yi Huang, Hao Zhang, Haibo Jin, Yifei Sun, Yuan gao, Zuoou Li 等 13 位作者
AI 速览：提出诊断即组合（DbC）框架，通过学习放射学征象词汇并依据指南组合征象图来诊断和定位未见疾病，在肝脏CT上优于对比方法。
Deep models for radiological diagnosis are typically trained to predict disease labels directly from images, which limits them to diseases seen during training and makes their decisions hard to interpret in radiologists' terms. Radiology, however, proceeds from visual findings to disease through published guidelines that make this decision rule explicit, interpretable, and reproducible. We introduce diagnosis by composition (DbC), a representation learning framework that encodes this findings-first logic. DbC learns a vocabulary of radiological findings by aligning image features with the text embeddings of the findings, each finding represented as a dense map over the scan. A disease is defined by a set of required findings derived from its guidelines, and its diagnosis map is a fixed, parameter-free composition of the corresponding finding maps. Because the finding-to-disease mapping is derived from guidelines rather than learned, DbC is interpretable and auditable by design, can extend to new diseases by specifying their findings without retraining, and can diagnose unseen diseases by composing their findings. On multi-phase liver CT, DbC detects and localizes unseen diseases, rare diseases from a different institution, and an external cohort significantly better than all comparators, including models that align images with disease names. DbC also improves detection and localization in the fully supervised setting and remains superior across training-set sizes. Although demonstrated on liver CT, DbC applies wherever radiological guidelines define diseases through findings.
Ziyang Zeng, Tingyang Chen, Jia Deng, Qiyao Wang, Shuo Tang, Yike Yang, Tao Ren, Wayne Xin Zhao 等 13 位作者
Scaling supervised fine-tuning for working agents requires not only broad coverage of professional tasks, but also expert demonstrations that faithfully carry out professional workflows to completion. However, missing or inconsistent task information can disrupt expert demonstrations, weakening the resulting supervision. We introduce WorkFactory, a scalable task-construction framework that turns professional intents into executable task environments through hybrid intent sourcing, semantic workflow planning, and execution-based repair. Starting from these intents, WorkFactory constructs a Semantic Workflow Graph that explicitly represents information requirements, dependencies, and locations within the task environment. The graph guides the instantiation of task environments, while an adversarial execution-refinement loop exposes missing or inconsistent information and steers targeted repairs that preserve the original intent. Using WorkFactory, we construct 112,686 tasks spanning 612 occupations and train the WorkPilot family at 4B, 27B, and 35B-A3B scales on the resulting teacher trajectories. All three models improve over their respective base models across four professional-work benchmarks, with WorkPilot-35B-A3B rising from 879 to 1565 Elo on GDPval-AA v2.1. Scaling experiments with up to 110k tasks further show that larger models benefit more from additional task data but exhibit diminishing returns at smaller data scales, while WorkPilot's long-horizon execution evolves from verbose trajectories toward more efficient task completion. Controlled component studies quantify the negative effects of task-environment defects on downstream performance and show how semantic workflow planning and execution-based repair yield downstream gains through improved task-environment quality. The subset of data are available at https://github.com/IQuestLab/WorkFactory.
Xiaoke Yan, Jiaying Lei, Danqing Shi, Zichen Wang, Nan Cao
Recent advances in large language models (LLMs) have enabled coding agents to autonomously perform complex software development tasks, yet their long-running execution remains difficult for users to inspect, understand, and steer. We present Vibetrace, an interactive trace visualization and intervention system for agentic programming built upon OpenCode. Vibetrace introduces a human-agent collaboration paradigm of Observe, Understand, Intervene, and Evolve. It (1) visualizes agent workflows as interactive execution traces for inspecting actions, intermediate states, and failures; (2) enables users to fork execution from prior steps and provide targeted corrections without restarting validated progress; and (3) distills traces and human feedback into reusable skills for continual agent evolution. On SWE-bench Verified, distilled skills reduce input tokens by 23.0% and cost by 15.5%, while improving pass rate from 65.0% to 68.0%. User studies further demonstrate improved accuracy and efficiency over conventional log-based interaction.
Collaboratively fine-tuning (FT) large language models (LLMs) over heterogeneous mobile devices fosters immense potential applications of personalized intelligence. However, such a vision faces critical system challenges. Existing federated LLM fine-tuning approaches remain limited by prohibitive on-device resource overhead, straggler-prone synchronous aggregation, and the assumption of a unified pretrained backbone across heterogeneous devices. In this paper, we propose EC-MobiLLM, a novel design for efficient collaborative LLM fine-tuning across heterogeneous mobile devices and heterogeneous backbones. EC-MobiLLM adopts a server-assisted side-tuning paradigm to minimize on-device overhead and pioneers non-blocking asynchronous collaboration to accelerate training. Moreover, EC-MobiLLM introduces adaptive feature alignment to support collaboration across heterogeneous backbones, aligning with practical needs in real-world deployments. Extensive experimental results demonstrate that EC-MobiLLM can maintain robust fine-tuning performance while achieving extremely low on-device memory, with at least 95.2\% reduction in computation overhead, 93.2\% reduction in communication costs and $5.1\times$ faster convergence compared to existing methods, validating its efficacy for practical LLM adaptation over heterogeneous mobile devices.
There is a huge gap between numerous intriguing applications fostered by on-device large language model (LLM) fine-tuning (FT) from fresh mobile data and the limited resources of a mobile device. While existing server-assisted methods (e.g., split learning or side-tuning) may enable LLM FT on the local mobile device, they suffer from heavy communication burdens of activation transmissions. In addition, their multiple on-device forward passes over the same samples across training epochs lead to substantial iterative computation on the mobile device. To address those issues, we develop Eco-MobiLLM, an efficient LLM FT method which can be deployed on the mobile device via server-assisted side-tuning. To reduce communication cost, Eco-MobiLLM develops an activation shortcut that transmits only the tokens involved in the loss calculation instead of full activation matrices to guide the side network tuning. Besides, to improve computing efficiency and further accelerate FT convergence, Eco-MobiLLM introduces a novel two-stage training paradigm, which allows the server to reuse historical activations and saves the mobile device from repeatedly computing forward passes for the recurring data samples. Extensive experimental results demonstrate Eco-MobiLLM's superiority.
Junwei Yao, Zike Yang, Zhijun Wang, Yongxi Zhou, Wenbo Ye
LLM agents call tools that delete data or move money through the Model Context Protocol (MCP). The specification asks for a human who can deny tool calls but splits the mechanism between the server, which can annotate a tool or ask the user through the host (elicitation), and the host application, which decides whether to prompt. We measure both sides from source code. In a 1,000-repository sample of the official MCP registry, an estimated 90% of destructive tools (95% CI 82-95) have no server-side confirmation, and most servers that elicit proceed, or fall back to a flag the model can set, when the host cannot relay the question. Of 41 popular open-source hosts and agent SDKs, 32 have no approval step by default, including 16 of 25 applications. Pairing sampled destructive tools flagged by a rule-based extractor with these applications, 60.7% of pairs have no enforced human approval by default. Through four agent SDKs, four of five low-cost models executed an injected deletion in 479 of 960 runs. If hosts prompted for every tool not annotated read-only or non-destructive, the unapproved share of these pairs would fall to 5.4%; universal elicitation support would leave 56.7%. Applications should therefore configure approval.
Programs that use a language model as a classifier act on its most probable label. Text injected into the input can move that label to one the attacker chooses, a hijack. A monitor that does not know the attacker's label can read the model's own confidence and raise an alarm when the top probability is low. We test this monitor on AG News topic classification with five systems, three of them served through a provider's API. Pooled over five families of injected text, low confidence ranks hijacked answers above clean ones on four of the five systems (AUROC 0.86-0.90). Ranking them does not stop many of them. A threshold set so that at most 5% of clean answers to separate calibration articles alarm falls among the least confident clean answers, and every more confident hijack passes. In our main gate run on the three API-served systems, the monitor misses 51-70% of the hijacks caused by a plain injected note (per-system point estimates). When the note adds one handwritten sentence that calls the target certain and asks for full probability, the monitor misses 99-100% of the hijacks, although low confidence still ranks them at AUROC 0.70-0.79. Confidence monitors should be tested at their operating alarm rate, against attackers who claim certainty.
Zafeiria Moumoulidou, Hamza Elhamdadi, Redwan Ahmed Rizvee, Ke Yang, Subrata Mitra, Cindy Xiong Bearfield, Alexandra Meliou
Analysts often use sampling in visual exploration to overcome the computational and perceptual bottlenecks of working with massive datasets. Yet, existing sampling methods optimize for statistical or geometric data properties, without explicit consideration for how humans perceive a visualization. We propose perception-augmented databases: a database framework that stores perceptual models of data as precomputed metadata and exposes a new query operator---perception-aware sampling---that selects subsets optimized for how humans visually perceive the result. We make the following contributions: (1) We introduce perception-augmented databases and design PAwS, a perception-aware sampling operator for scatterplot visualizations that leverages saliency maps---a computer vision tool for predicting areas of visual attention---combined with density and coverage objectives. (2) We design ApproPAwS: a fast approximate visualization method that works directly over a perception-aware compressed synopsis of the data, achieving up to 100x speed-ups with minimal loss in visual fidelity, and without accessing the underlying data at query time. (3) We contribute a perceptual similarity metric that evaluates the quality of a sample by comparing its saliency map to that of the original dataset, a novel evaluation methodology, applicable to any sampling method for scatterplots. (4) Our extensive experimental evaluation over datasets up to 3.5M points demonstrates that our methods consistently outperform five state-of-the-art sampling algorithms. A user study confirms that PAwS samples are preferred by humans, validating our quantitative findings.
Clinical interpretation of a 12-lead electrocardiogram (ECG) often requires answering targeted diagnostic questions with clear, quantifiable decisions. We introduce ECGJev, the first ECG framework built on a decision decoder, a pretrained language model trained to return a probability for every listed option. A question-conditioned bridge compresses frozen waveform and spectrogram features into 16 ECG tokens, and a two-stage procedure adapts the bridge and low-rank adapters on the decoder's native option readout using diagnostic labels alone, scoring all options for a question in one forward pass. Evaluated across four ECG benchmarks, ECGJev achieves the highest macro-F1 on CPSC2018 (80.46%), PTB-XL (71.68%) and Chapman (96.86%), and generalizes competitively to the small-sample PTB cohort. Concept probes and perturbation analyses further show that its predictions rely directly on the ECG input, remain stable under option-order and wording variations, and transfer zero-shot to unseen waveform concepts. Our code is available at https://github.com/zhuoyuanli-lab/ECGJev.
Xuan He, Sining Zhoubian, Ruozhen Yang, Yuhao Wen, Zhineng Wang, Yuyang Hu, Ke Yang, Yuyu Zhang 等 11 位作者
Memory is increasingly used in long-horizon agents, yet the net value of building memory over direct retrieval from raw history remains largely unclear. Through a large-scale evaluation across multiple benchmarks and model backbones, we find many dedicated memory systems underperform naive RAG in many cases, indicating a negative net value. Taking RAG as a reference, we dissect the net value of memory into the gains and losses over direct retrieval, termed memory benefit and memory tax. These gains and losses arise because memory process changes how the raw information is organized and accessed. We view such changes as the double-edged deviations from the raw history. They are originally designed and introduced to benefit the task, for example by reorganizing facts or distilling preferences, but some may also remove or obscure evidence, giving rise to memory tax. Since memory benefit reflects the expected gains of its design, we then focus more on the overlooked memory tax, study how it arises from the system and evaluation side, and generalize the concept to more application scenarios of memory. We hope the notion of memory tax can provide a new lens and shift more attention from standalone performance to the net value of memory, informing more reliable memory designs and evaluations. We will open-source the code and results to the community. Project Page: https://hexuan21.github.io/MemoryTax/
Ke Yang Institute of Information Engineering, Chinese Academy of Sciences, Dongyang Liang, Jing Yu Institute of Information Engineering, Chinese Academy of Sciences, Shuguang Yuan Institute of Information Engineering, Chinese Academy of Sciences, Chi Chen
The rapid development of Diffusion Language Models (DLMs) raises concerns about watermarking for DLM-generated detection. However, existing sequential LLM watermarking cannot be directly applied to DLMs, as DLMs’ generation order is arbitrary. While emerging studies adapt biased LLM watermarking to DLMs by temporarily predicting the watermark prefix, they suffer from degraded quality and unstable watermarking due to bias accumulation and prediction errors. Besides, they cannot carry multi-bit watermarks. In this paper, we propose unbiased multi-bit watermarking for DLMs. We introduce a stability-aware constraint that allows watermarking only in stable contexts and a bit-controlled, unbiased modulation to preserve the original DLM output distribution, achieving stable watermarking with minimal quality impact. To enhance detection robustness, we design a Regret-based Remasking, which grants a “second chance” for unwatermarked tokens to be regenerated. It can seamlessly integrate into DLM inference with no added diffusion steps and latency. Experiments across DLMs and various tasks show that our scheme is effective, achieving superior generation quality compared to baselines while maintaining high detection accuracy and multi-bit capacity. Our code is available here https://github.com/iieSKLCSDsg/UMR.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表