Research

• General in-context learning for embodied agents1 paper raises it · 0 address it

  Enables models to adapt to new tasks and environments without retraining, improving robustness in real-world deployment.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Cost and latency in embodied control loops1 paper raises it · 0 address it

  High computational and time costs limit real-time applicability of agentic systems in dynamic environments.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Physical-state verification and failure recovery1 paper raises it · 0 address it

  Agents must reliably detect and correct errors in physical interactions to avoid irreversible failures in real-world tasks.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Long-horizon task preservation in embodied agents1 paper raises it · 0 address it

  Maintaining progress over extended sequences is critical for complex, multi-step tasks in robotics and navigation.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Open-source stacks for generalist computer-use agents1 paper raises it · 1 addresses it

  Closed systems hinder reproducibility and systematic study of agent design for real-world computer tasks.

  • OpenGCUA: Toward a Fully Open Stack for Generalist Computer Useraises · addresses

Named by an AI from these papers' abstracts.

• Training & fine-tuning63+6200.0%
• Evals105+3400.0%
• Computer vision49+2350.0%
• Code models23+2200.0%
• Image & video generation36+1100.0%
• Federated learning49+28.9%
• Quantum machine learning240.0%
• Character recognition OCR170.0%

• arXiv cs.CYOct 18

  Assessing the Empirical Evidence for Loss of Control from Agentic General-Purpose AI

  Risto Uuk, Santeri Koivula, Lorenzo Pacchiardi, Rokas Gipiškis, Laura Caroli +1

  Theoretical literature warns that advanced generalist agentic systems might pursue instrumental goals that are misaligned with human preferences, thus posing risks of irreversible loss of human control. The EU's Code of Practice for General-Purpose AI lists loss of control as a systemic risk that must be assessed, yet no structured assessment of the quality of the empirical evidence for this risk exists. This study fills that gap. We evaluate 19 recent empirical studies across eight properties identified from threat models (such as strategic deception, situational awareness, and autonomous replication), assessing the quality of evidence across three dimensions: construct validity, content validity, and external validity. Our central finding is that the current body of evidence supports a judgment of loss of control as "weakly plausible." In our framework, 'plausibility' denotes the strength and direction of available evidence, not a probability estimate. Emerging evidence exists across multiple properties. No study clears the validity thresholds for a stronger judgment that a property is present; a single study clears them only for a negative-direction judgment (against…

  PDF
• arXiv cs.CVOct 18

  Evaluating Cloud and Self-Hosted OCR Systems for Arabic, English, and French Documents: Bounding Box Detection, Text Recognition, and Mathematical Expression Recognition

  Jumana Alsubhi, Ahmed Alhusayni, Abdulrahman Gharawi, Israa Hamdine, Khadija Alansari +3

  Choosing an OCR system for a multilingual archive requires balancing accuracy against deployment constraints. This study compares cloud and self-hosted systems on three tasks: bounding box detection on 48 Arabic, English, and French pages, text recognition on 315 pages across nine document types, and mathematical expression recognition on 83 English and Arabic samples. Detection was assessed by page-level visual grades, text by character error rate (CER) against a Gemini-generated reference transcription, and mathematics by scores against known answers. Main estimates are accompanied by 95% intervals and paired comparisons. Google Document AI and Surya produced Good line boxes on 47 of 48 (98%) and 46 of 48 (96%) pages. Datalab, Azure Document Intelligence, and Document AI had mean CERs of 0.145-0.152, with no difference detected in unadjusted paired comparisons. Surya had the lowest mean CER among self-hosted systems (0.199). No mathematical test had a statistically clear leader. Strict checks showed that near-perfect page scores could conceal a replaced Arabic variable or Eastern Arabic digits rewritten in Western form. Reliable evaluation therefore requires uncertainty…

  PDF
• arXiv cs.SEOct 18

  Noise-Awareness in Quantum Software Testing: An Empirical Landscape Analysis

  Mir Mohammad Yousuf, Shabir Ahmad Sofi, Bisma Majid

  Noisy Intermediate-Scale Quantum (NISQ) hardware has transformed quantum computing into a practical experimental platform while introducing fundamental challenges associated with hardware noise, decoherence, and probabilistic execution. Although quantum software testing (QST) has rapidly emerged as an active research area, it remains unclear to what extent existing testing methodologies explicitly engage with noisy and hardware-constrained execution environments. This paper presents a systematic empirical landscape analysis of noise-awareness in QST research through a deterministic, taxonomy-guided analytical framework applied to 61 papers published between 2018 and 2025 across eight testing paradigms, contrasting how contributions are framed with how they are evaluated. Among the analyzed papers, 47 (77.0%) are noise-oblivious and only 14 (23.0%) explicitly engage with noise-related execution concerns, while only 11 (18.0%) report evaluation on physical hardware or a noisy simulator. Test generation, the largest paradigm, exhibits a noise-awareness rate of only 14.3%, and debugging and repair 18.2%; coverage-based, metamorphic, and differential testing contain no noise-aware…

  PDF
• arXiv cs.CVOct 18

  Map2Vid: Embodied Navigation Video Generation with Human-Friendly Motion Control

  Yulin Dai, Yuqian Fu, Tianwen Qian, Zhen Xing, Bojia Zi +1

  Controllable video generation has recently achieved impressive visual quality, yet existing control paradigms remain poorly suited to embodied navigation. Text prompts are spatially ambiguous, while camera-trajectory control relies on low-level motion specifications that are unintuitive for human users and difficult to manipulate. In this work, we introduce Map2Vid, a novel task that reformulates embodied navigation video generation as a BEV-map-to-video synthesis problem, together with a large-scale benchmark for systematic evaluation. By using BEV maps, which provide a visual representation of global scene layout and intended navigation routes, Map2Vid enables human-friendly navigation video generation. To support this task, we construct a dedicated dataset with over 100K navigation sequences across 61 indoor scenes, each paired with egocentric videos, BEV maps, camera trajectories and language instructions. We further develop a unified framework that integrates multiple control signals, together with a curriculum training strategy that progressively weakens non-map inputs, enabling flexible inference under both full conditioning and map-only input. Extensive experiments across…

  PDF
• arXiv cs.LGOct 18

  DROD: Dual-Ratio Ordered Dropout for Federated Continual Learning on Heterogeneous Devices

  Berk Sezer, Özgür Soysal

  Federated learning for capacity-heterogeneous clients and federated continual learning are each well studied on their own, but their intersection -- clients holding differently sized models while every client's data stream also drifts over time -- remains underexplored. We study this setting directly, training a ResNet-18 fleet from scratch on Split-CIFAR-100 (10 tasks, class-incremental), and introduce DROD (Dual-Ratio Ordered Dropout), which builds on FjORD's Ordered Dropout by using two independent reduction axes to extract sub-models: a shortcut ratio and a branch ratio. On a shared parameter/FLOP budget ladder, DROD achieves the highest mean accuracy among the heterogeneous-model federated learning strategies we compare. Its advantage over the two strongest baseline geometries persists across every replay-buffer size we test and under ER, DER++, and BiC; under EWC, all three geometries collapse alike.

  PDF
• arXiv cs.CLOct 18

  Hidden Cartographer: Hidden State Guided Data Selection for Dual Budgeted Instruction Tuning

  Zixuan Qin, Qi Qin, Erbo Li, Ting Wei, Yifan Sun +1

  Given the high cost of expert annotation and limited fine-tuning resources, effective data selection from large unlabeled pools under strict budgets has become a crucial step in adapting LLMs to specialized domains. Existing label-dependent methods risk distorted selection when using pseudo-labels, while traditional active learning methods do not explicitly characterize instruction-tuning dynamics. To address these limitations, we propose Hidden Cartographer (HC), a two-stage framework that uses hidden states to guide annotation and exact training dynamics to select the final training subset. The first stage is Progressive Candidate Expansion, which utilizes the proxy to iteratively select top unlabeled candidates for expert annotation, merging them to recompute training dynamics and update the proxy until the annotation budget is exhausted; the second stage is Exact Cartography Pruning, which ranks these annotated candidates by their exact dynamics scores to isolate a compact subset under the final training budget. Experiments across the medical, legal, and financial domains demonstrate that HC achieves the highest average accuracy among all budgeted baselines, requiring…

  PDF
• arXiv cs.SIOct 18

  Do LLMs Really Lack Difference Awareness? Uncovering Heuristic Substitution in Group-Differentiating Fairness

  Jingshen Zhang, Bo Wang, Yanlin Fu, Yuewen Dang, Dongming Zhao +2

  In this paper, we investigate the issue of difference awareness in LLMs, defined as a model's ability to recognize and acknowledge meaningful differences between social groups in fairness decision-making. Previous studies have consistently shown that models perform poorly on such tasks, a finding commonly attributed to a lack of difference awareness. We propose an alternative account, which we term Fairness Heuristic Substitution: difference awareness is not necessarily absent, but the complex demand of simultaneously identifying relevant group differences and determining their implications for fairness decisions may lead models to substitute the simpler, more accessible equal-treatment heuristic, i.e., formal equality. Based on this hypothesis, we introduce TDD, a Two-stage Distinguish-Decide framework that explicitly decomposes fairness reasoning into first distinguishing contextually relevant group differences and then making the corresponding decision, thereby discouraging reliance on formal-equality shortcuts. (i) Across four benchmarks and ten models ranging from billions to trillions of parameters, TDD significantly outperforms one-stage baselines, even one strengthened…

  PDF
• arXiv quant-phOct 18

  LiftDiff: Guided Discrete Diffusion for Quantum Code Construction

  Yotam Peled, Eliya Nachmani

  The realization of fault-tolerant quantum computation relies on Quantum Error-Correcting Codes (QECC). Of particular interest are codes that combine a high encoding rate with sufficient distance. Existing algebraic constructions, such as lifted product (LP) codes, build a large quantum low-density parity-check (qLDPC) code from a classical base matrix. In this work we introduce LiftDiff, a generative approach to the discovery of QECC. A discrete denoising diffusion model is trained to generate base matrices over the group algebra $\mathbb{F}_2[\mathbb{Z}_\ell]$, which are then expanded into LP codes. The reverse diffusion process is guided by an energy function that evaluates quantum code properties, including girth, degree, sparsity and decoding performance, steering the sampler toward matrices that yield LP codes with favorable parameters. Our method discovers codes with parameters such as $[[169,13,11]]$, $[[143,15,9]]$, and $[[260,24,10]]$, including a $[[247,19,14]]$ code that achieves the highest ${kd^2}/{n}$ among the baselines. Moreover, several discovered codes outperform the Gross code across a broad range of the tested physical error rates under both code-capacity and…

  PDF
• arXiv cs.CLOct 18

  Calibration-Free Early Stopping for Chain-of-Thought Reasoning via E-processes

  Dong-Jae Lee, Gyojin Han, Junmo Kim

  Reasoning large language models (LLMs) benefit from test-time scaling, motivating adaptive policies that spend tokens when they help and stop early otherwise. Existing rules halt when a confidence or stability signal crosses a threshold, but such thresholds rarely transfer across models and datasets, and validation-calibrated rules lose their guarantee under shifts in test-time horizon or signal dynamics. We reframe adaptive stopping as sequential evidence accumulation along the reasoning trajectory rather than threshold calibration on held-out data. Our framework runs two e-processes on a shared per-chunk signal: a commit stream tracks the signal's position relative to its slow trend, and an abstain stream tracks its signed velocity, making the two anti-correlated by construction. Under signal-level nulls on the per-chunk increments, Ville's inequality yields an anytime-valid signal-level false-alarm bound at user-specified significance levels, with no held-out calibration. Across various model--dataset combinations, the framework matches or exceeds calibration-based stopping on the token--accuracy Pareto front using zero calibration data, and the same significance level…

  PDF
• arXiv cs.LGOct 18

  NeuroAct: Advancing EEG Analysis and Decoding via Dual-Loop Autonomous Research

  Wei Xiong, Jiangtong Li, Han Zhao, Linxiao Yang, Jie Li +1

  EEG analysis and decoding underpin brain-computer interfaces, cognitive assessment, and sleep monitoring. Fixed pipelines and reward-driven search do not explicitly determine how signal structure, representations, predictor behavior, and training dynamics contribute to performance or generalization gaps, leaving intervention choice underdetermined. We introduce NeuroAct, an autonomous research agent framework for EEG analysis and decoding. NeuroAct implements dual-loop autonomous research across EEG foundation models, task-specific predictors, and model combinations. In an inner loop, research, development, and review roles use evidence on signal structure, representations, predictor behavior, and training dynamics to test hypotheses, select analyses or interventions, and refine candidates using performance, stability, and cost feedback. An outer loop distills outcomes into conditional, testable recommendations with task and model conditions, then validates and revises them on subsequent tasks.We evaluate four settings: (A) EEG General Analysis, (B) Cross-task Model Optimization, (C) Single-task Open Research, and (D) Collaborative Prediction. In Setting A, NeuroAct's 80.88…

  PDF
• arXiv cs.AIOct 18

  BrandEval: A Two-Track Multi-Agent Benchmark for Risk-Sensitive LLM Crisis Communication

  Haotian Gan, Weidong Tang, Hanbin Sun, Yi Luo, Xintian Li +3

  Evaluating LLMs for crisis communication requires more than judging isolated responses, because public feedback and competitor actions can reshape trust over time. Existing benchmarks remain largely static and single-turn, offering limited evidence about long-horizon risk-sensitive communication. We introduce \textbf{BrandEval}, a two-track benchmark for product-crisis communication, focusing on public brand responses to product-related safety, quality, service, and trust incidents. BrandEval-Static evaluates stakeholder-aware strategic analysis, while BrandEval-Dynamic stress-tests response policies in \textbf{BrandPolis}, a controlled multi-agent sandbox with consumer feedback, competitor responses, and trust-risk trajectories. We also construct \textbf{BrandSRD}, a Chinese crisis-response dataset, and instantiate a reference Strategic Agent using \textbf{Strategic Rationale (SR)}. Across proprietary and open-source models, we find that static strategic quality is informative but incomplete: it aligns with some end-state outcomes but does not reliably predict volatility and delayed recovery; dynamic stress tests further expose tail-risk failures under manipulative or weakly…

  PDF
• arXiv cs.CLOct 18

  A Thousand Faces: A Persuasion Benchmark with Experience-Grounded Persuadees Informed by Psychology and Pragmatics

  Ziqing Zhuang, Dingyi Zhang, Deyu Zhou

  People can express the same beliefs and desires yet need different reasons to change their minds. What makes an argument relevant depends on how those views arose: experiences shape what a person finds credible, what matters to them, and what could prompt reconsideration. These personal grounds may also be expressed indirectly, through contextual clues in conversation. Benchmarks that specify beliefs and desires without these connections can reward arguments that match a description while overlooking whether they address the person's reasons. We introduce A Thousand Faces (ATF-Persu), with 1,000 personas across 200 topics and nine domains. Drawing on personality and attitude theories and conversational implicature, we construct five personas per factual context, linking experiences to judgments, expression, and grounds for revision. Evaluations show lower persuasion success than three comparison benchmarks under their respective protocols. Complete backgrounds yield broader use of supplied personal grounds and higher persona faithfulness than simplified specifications; blinded reviewers also rate ATF-Persu above CToMPersu on identifiable personal reasons, their influence on…

  PDF
• arXiv cs.LGOct 18

  Dynamic System Inspired Continuous Large Language Model for AI4Science Tasks

  Zhaoxin Fan, Zheng Zhu, Yuqing Ma, Faguo Wu, Jian Yang +1

  Scientific foundation models increasingly operate on continuous phenomena, yet still use discrete residual computation. We ask whether pretrained residual transformations can support additional latent computation without expanding the backbone. To test this possibility, we introduce Continuum-LM, which reuses these transformations as dynamical operators in a continuous-depth language model. Its hidden states evolve through an ordinary differential equation evaluated by higher-order numerical integration, with solver order and step size controlling computation. Across eight scientific tasks, Continuum-LM trains stably and improves mean exact-match accuracy by 7.77 and 10.96 percentage points over discrete counterparts at the 0.6B and 1.7B scales, respectively. At 0.6B, it also outperforms an Ouro-style recurrent control under matched parameter and residual-function evaluation budgets. Together, these results support extensible computational depth through parameter reuse, identifying continuous hidden-state evolution as a practical design dimension for scientific language models beyond parameter scaling.

  PDF
• arXiv cs.LGOct 18

  ViCoR: Estimating Visual Necessity via Counterfactual Residuals for Multimodal Medical Data Selection

  Haotian Gan, Yudong Li, Xintian Li, Jiarui Han, Weidong Tang

  Medical vision-language models often exhibit hidden stratification, where high overall performance masks failures on subtle visual findings and overlapping clinical conditions. We trace this failure mode to modality imbalance during fine-tuning, where strong linguistic priors can suppress visually grounded learning and induce plausible but weakly grounded rationales. Existing data selection methods usually score image-text pairs as holistic units, without separating visually necessary samples from shortcut-solvable ones. To address this problem, we propose ViCoR, a counterfactual-residual-based data selection method for Med-VLM fine-tuning. ViCoR estimates visual necessity through image-dependent optimization residuals, calibrates the resulting scores with medical supervision compatibility, and constructs a diversity-aware fine-tuning subset. Across two backbones, four external generalization benchmarks, and one multi-disease robustness benchmark, ViCoR-selected 20\% subsets achieve the strongest average performance compared with full-data fine-tuning and selection baselines. Gradient-level analysis and perturbation audits further show that ViCoR enriches image-dependent training…

  PDF
• arXiv eess.SPOct 18

  RAIL-5G: A Challenging Real-World Dataset and Benchmark for AI-Based 5G Indoor Localization

  Jonas Pirkl, Andreas Eidloth, Maximilian Kasparek, Tobias Feigl, Christopher Mutschler

  Reliable radio-based indoor localization is essential for industrial automation, yet existing methods are often evaluated in environments that do not fully reflect the severe multipath and non-line-of-sight (NLOS) conditions encountered in real deployments. This limits our understanding of how AI, i.e., modern geometric, unsupervised, and semi-supervised localization methods generalize to industrial 5G settings. We present RAIL-5G, a public real-world 5G NR channel impulse response (CIR) benchmark recorded in an operational railway maintenance facility. RAIL-5G covers an 811m$^2$ indoor area with eight distributed radio units, robot-mounted user equipment (UE) trajectories, high-accuracy iGPS reference positions, and severe metal-induced multipath caused by train compartments and maintenance infrastructure. We benchmark geometric time-difference of arrival (TDoA), supervised fingerprinting, unsupervised channel charting, and self-supervised wireless pretraining methods. Our results highlight the difficulty of RAIL-5G: on the full dataset, TDoA reaches 7.02m mean average error (MAE), the best unsupervised method reaches 2.46m, and supervised fingerprinting achieves 0.52m.…

  PDF
• arXiv eess.SPOct 18

  ChronoChart: Setup-Agnostic Channel Charting for Label-Free Indoor Localization

  Majid Sharghi Foroushani, Jonas Pirkl, Jonathan Ott, Maximilian Stahlke, George Yammine +2

  Accurate radio-based indoor localization remains challenging in multipath-rich and dynamic indoor environments. Supervised fingerprinting methods achieve high positioning accuracy by learning mappings from radio channel measurements to physical locations, but require costly position labels and frequent recalibration. Channel Charting (CC) addresses this limitation by learning low-dimensional representations of Channel State Information (CSI) without position labels, yielding channel charts in which nearby points correspond to nearby physical locations. However, prominent CC methods rely on setup-specific dissimilarity metrics, such as ADP-, CIRA-, or PDR-based distances, and on precomputed $O(n^2)$ distance matrices, limiting scalability and transferability across radio systems. We propose ChronoChart, a timestamp-supervised CC framework that requires only CSI and timestamps without prior knowledge for training. Our method jointly optimizes complementary objectives: a triplet loss using temporally derived positive and negative samples to establish coarse global structure, a temporal-proximity loss that promotes smooth trajectories in the chart, and a joint-embedding loss that…

  PDF
• arXiv cs.LGOct 18

  Shallow Probes Are Sufficient for Near-Oracle Personalization Depth Allocation in Federated Learning

  Beyazit Bestami Yuksel, Emrah Dikbiyik, Ayse Yilmazer Metin

  Rather than proposing a new federated optimizer, we study whether a small client-specific measurement budget can support cost-aware personalization-depth decisions under a utility that trades Macro-F1 against deployment cost. Exhaustive evaluation of six depths defines an offline oracle that is not deployable. We audit the Probe Sufficiency Hypothesis: three shallow probes (P0--P2) may contain most decision-relevant information for approaching that oracle. FedPBA is a secondary learned instantiation for deep-action extrapolation. On CIFAR-10/100 and patient-disjoint PTB-XL (45 sessions; 900 clients; 2700 evaluations; GroupKFold), training-free greedy probing achieves regret 0.0071 and 98.3 percent oracle recovery, beating metadata-only policies and FedPBA (0.0110 / 97.3 percent). Aggregate shallow-best versus oracle correlation is 0.9818 but falls to 0.578 on deep-oracle rows; PTB-XL deep-oracle cases remain a high-regret boundary without a validated escalation gate.

  PDF
• arXiv cs.CLOct 18

  Acharya 2.0: An AI Tutor for Self-Learning Subjects in Hindi

  Shantipriya Parida, Debasish Dhal, Sonal Khosla, Sambit Sekhar, Muhammed Abdur Rahmaan Kamaldeen

  Text processing is crucial for low-resource languages as it enables the development of tools and technologies that bridge the digital divide and provide equitable access to information and services. By empowering low-resource languages with robust text processing, we can preserve linguistic diversity, enhance education, improve communi- cation, and foster inclusivity in a rapidly digitizing world. Due to the lack of sufficient annotated datasets and lin- guistic resources, it is challenging to develop accurate and reliable language technologies for these low-resource languages. Large Language Models (LLMs) have attracted sig- nificant attention with their performance and the value they produce. However, with the ever-evolving language, adapting LLMs for a specific task is time-consuming and resource-intensive. Thus, Retrieval-Augmented Genera- tion (RAG) has overcome this limitation by augmenting LLMs with external knowledge, thereby improving their efficiency and extensive task-specific adaptation. Most applications of the RAG technique have been limited to English and has seen limited application in In- dic languages. This paper introduces the RAG-powered AI-Tutor application,…

  PDF
• arXiv cs.LGOct 18

  Sampling and Curvature in First-Exit Comparisons for Momentum SGD

  Haiyi Li

  First-exit prediction for momentum SGD depends on sampling information that pointwise gradient laws omit. Strongly convex finite pools with identical pointwise laws and matched iid momentum dynamics can have different reshuffled exits; even analytic pools with uniformly close pointwise laws can have separated exit probabilities. For persistent sampling in one dimension, we establish an explicit $O(h)$ mean-exit comparison with a finite-rate switching process under nonvanishing component velocities and step-size conditions; iid sampling instead averages to an ODE. Retaining realized minibatches and sampled curvature yields finite-window exit brackets beyond this regime. In 32 trained ridge-logistic runs, 29 meet prespecified targets with rounding-certified brackets; an enclosure control explains most of the width advantage. Exact-real certificates establish finite means at the original tested radii for three ResNet-18 checkpoints and give local finite-mean conditions for tanh and ViT networks; their upper bounds remain loose, and original-radius ViT mean guarantees remain unresolved despite 128 fresh continuation paths. A two-network replication gives 32/32 successful seeds in…

  PDF
• arXiv cs.CLOct 18

  Reranking Answers of Large Language Models with Knowledge Graphs

  Mikhail Salnikov, Hai Le, Olga Tsymboi, Ivan Lazichny, Dmitrii Iarosh +4

  Answering natural language questions over knowledge graph data is challenging due to the vast number of facts, which can be difficult to process and navigate. One potential solution for this issue is to use mined subgraphs related to the query, although this process still requires extracting these subgraphs. This research presents a solution for extracting subgraphs related to entity candidates from a question-and-answer set, which can be obtained by inferring a large language model by calculating the shortest paths between entities. The proposed approaches detail various features that can be extracted from the subgraphs and reranking models to select the most probable answers from a list of candidates. Experiments were conducted on Wikidata to evaluate the effectiveness of the proposed approaches. This involved enumerating all the main feature types that can be extracted from mined subgraphs and a detailed analysis of the proposed features and reranking method combinations. In addition, a public web application that provides a useful web tool for studying the graph space between question and answer entities has been developed to work with subgraphs. This includes visualization…

  PDF