# Research Papers — 2026-08 · bundle page 110 of 240 > Full-text markdown of papers #5451–5500 of 11995 published in 2026-08, newest first. AI-readable mirrors of the full 11995-paper 2026-08 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-08/page/110 [← Previous bundle](https://aipod.dev/papers/archive/2026-08/page/109.md) [Next bundle →](https://aipod.dev/papers/archive/2026-08/page/111.md) ### Closing the Affective Loop: Multimodal Speaker-Listener Emotion-Dynamics-Aware Empathetic Social Robots Authors: Zi Haur Pang, Casey Kennington, Tatsuya Kawahara Published: 2026-08-17 Categories: cs.HC, cs.CL, cs.RO arXiv: https://arxiv.org/abs/2608.16686v1 Empathetic social robots should respond not only to what users say, but also to how their emotions dynamically evolve during interaction. However, existing empathetic dialogue systems are often text-centered and primarily model empathy as a one-way mapping from the user's emotion to the system response, limiting their ability to capture embodied speaker--listener affective exchange. We present AffectLoop, a multimodal speaker-listener emotion-dynamics-aware spoken dialogue system implemented on the Misty II robot. The system tracks the speaker's verbal and facial affective dynamics, estimates the robot listener's own verbal and behavioral affective state, and conditions LLM-based response generation on both affective streams. The robot then generates a short spoken empathetic response together with emotionally congruent embodied behavior, forming a closed speaker--listener affective loop. We evaluate the system in a pilot within-subject study with five participants, comparing it with an otherwise identical utterance-conditioned baseline that omits the speaker- and listener-affective-state inputs. The proposed system received higher overall impression ratings, especially for empathetic response and user satisfaction. Post-hoc log analysis further showed higher speaker-listener affective alignment and stronger valence-based distress recovery. These preliminary results suggest that explicitly modeling both speaker emotional dynamics and listener affective state can improve embodied empathetic interaction. --- ### Bridging the Gap between Labeled and Unlabeled Data via Unified Flow with Feature Memory Bank Authors: Shanwen Wang, Xin Sun, Danfeng Hong, Junyu Dong, Patrick Le Callet Published: 2026-08-17 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.16681v1 Although semi-supervised semantic segmentation ($\text{S}^4$) utilizes abundant unlabeled data to reduce manual labeling burdens, independent training of labeled and unlabeled data causes the former to dominate, which severely degrades pseudo-label quality. To address this challenges, we propose a novel remote sensing (RS) $\text{S}^4$ method via unified flow with feature memory bank (UFFM). Specifically, UFFM comprises two key innovations: unified flow (UF) and feature memory bank (FMB). The UF is a new training flow that generates less biased pseudo-labels by combining an external visual foundation model (VFM) with an RS domain teacher, and jointly optimizes labeled and pseudo-labeled data under a unified training objective. The FMB is a novel memory module for $\text{S}^4$ that dynamically updates class-specific features during training and reduces the feature discrepancy between labeled and unlabeled data through class-feature alignment. To verify the effectiveness of our model, we conduct extensive experiments on RS datasets. The experimental results show the superiority of our method over SOTA $\text{S}^4$ methods. Moreover, the results demonstrate the effectiveness of our contributions in bridging the optimization and feature representation gap between labeled and unlabeled data. Our code is released at \href{https://github.com/wangshanwen001/RS-UFFM}{https://github.com/wangshanwen001/RS-UFFM}. --- ### How Sampling Strategy Affects Imbalance Mitigation in LiDAR Segmentation: A Study of Structured vs. Random Point-Based Architectures Authors: Antonis Savva, Christos Kyrkou, Theocharis Theocharides Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16673v1 Class imbalance in LiDAR point clouds poses challenges for semantic segmentation in autonomous navigation and urban mapping. While 2D vision has numerous mitigation techniques, their effectiveness in 3D remains unclear. We benchmark six reweighting schemes and five imbalance-aware losses across three datasets (DALES, S3DIS, STPLS3D) using two architectures (KPConv, RandLA-Net). Inverse-frequency weighting degrades performance by up to 12% compared to uniform weighting, with catastrophic failures in minority classes. Uniform weighting performs within 2% of complex losses for structured sampling (KPConv) but benefits less for random sampling (RandLA-Net, up to 4.6% gap). Loss landscape analysis reveals a complex interplay: for structured sampling, imbalance ratio determines landscape geometry on real LiDAR data but decouples from it on synthetic data; for random sampling, landscapes show high sensitivity to dataset geometry regardless of imbalance ratio. For the two evaluated point-based architectures, these results suggest that the interaction between sampling strategy (structured vs. random), imbalance severity, and data acquisition characteristics shapes which mitigation approaches are effective. --- ### Does the LM Head Create a Harmful Gradient Bottleneck? A Causal Test Authors: Anand Murugan Published: 2026-08-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.16671v1 The language-model head maps a hidden state of width D to a vocabulary of size V, so its transpose can return at most D independent directions to the Transformer. Godey and Artzi argue that this severe projection is a harmful optimization bottleneck. We separate the geometry from the causal claim. Our backward-only intervention keeps the ordinary logits and the exact LM-head parameter update while reducing only the rank of the gradient sent into the Transformer. Across five paired seeds on byte-level and BPE-8192 WikiText-2 models, reducing backward rank increases validation loss. An equally ranked factorized forward head, however, increases loss substantially more. At half rank in the larger model, the backward-only loss increase is 0.0586 (95% CI (0.0167, 0.1005)), while the factorized forward head increases loss by 0.1795 ((0.1547, 0.2042)). The vocabulary-space residual also contributes to the ordinary LM-head update, and removing that contribution is harmful. Additional controls show that repeated-token failures are confounded by the number of independently sampled symbols, that adding never-target output classes does not impair learning, and that projection diagnostics do not reliably predict progress in our runs. Tested auxiliary feedback routes do not beat tuned backpropagation. These results confirm strong geometric compression but do not establish that it is a harmful optimization bottleneck. --- ### Concept-based explanation of gene expression prediction from H&E images Authors: Amos Muench, Jonathan Thielmann, Reduan Achtibat, Maximilian Dreyer, Philip Bischoff, Caroline Forsythe, Hamidreza Parand, Thomas Walter, David Horst, Sebastian Lapuschkin, Wojciech Samek, Teresa Gabriela Krieger Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16669v1 Recent advances in pathology foundation models have enabled accurate prediction of spatial transcriptomics (ST) from routine H&E images. However, existing explainability methods for vision transformer (ViT)-based models are largely limited to local heatmaps and do not reveal how morphological concepts contribute to ST predictions. Here, we introduce an explainable framework that combines relevance propagation and concept discovery to link transcriptional programs to tissue morphology. We developed a ViT-based framework for virtual ST from H&E images that combines ViT-aware layer-wise relevance propagation with relaxed archetypal TopK sparse autoencoder-based concept discovery. This approach provides both local explanations and global insights into the morphological patterns associated with transcriptional programs. We applied the framework to colorectal cancer ST data from the HEST-1k cohort and evaluated its generalizability in TCGA COAD. Our architecture accurately predicts clinically relevant ST signatures and accompanying molecular phenotypes. Measured and predicted gene expression profiles reveal substantial spatial heterogeneity of the colorectal cancer subtypes iCMS2 and iCMS3 across a large number of samples. Spatially resolved and aggregated iCMS classification achieve weighted F1 scores of 0.872 and 0.819 (0.770 in TCGA COAD), respectively, and both stratify patient outcome. Beyond prediction, our framework establishes a relevance-based concept atlas linking molecular phenotypes to histopathological representations. Comparison of activation- with relevance-derived concepts demonstrates that relevances provide a more direct link between tissue morphology and downstream predictions. We establish a general strategy for concept-based explanation of spatial prediction, and our framework is readily applicable to a broad range of ViT-based pathology models. --- ### Chronocooked: A Benchmark for Implicit Interval Timing in Reinforcement Learning Agents Authors: Amrapali Pednekar, Alvaro Garrido-Perez, Yara Khaluf, Pieter Simoens Published: 2026-08-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.16666v1 This paper presents Chronocooked, a reinforcement learning (RL) benchmark suite for studying implicit interval timing in RL agents. Inspired by Overcooked, the suite comprises cooking scenarios that require temporal decision making. The tasks and reward functions are designed such that temporal information is unobserved yet critical for optimal performance. The environment is intentionally kept simple to enable controlled experiments and support biologically plausible models. Evaluation metrics are designed to expose limitations in timing abilities of RL agents, and we report baselines using a non-recurrent, a recurrent, and a biologically plausible model. This work ultimately aims to underscore the need to incorporate time perception and temporal processing in artificial agents designed for human robot interaction and deployment in time dependent human societies. --- ### Random Quadratic Form with random forcing: Metastable synchronization by noise Authors: Anna Shalova Published: 2026-08-17 Categories: math.PR, cs.LG, math.DS arXiv: https://arxiv.org/abs/2608.16664v1 We study the Random Quadratic Form (RQF) on a sphere in the presence of random Brownian forcing. We show that the forcing does not effectively change the law of the process but affects the synchronization properties of the system. While the RQF without forcing exhibits partial synchronization due to the intrinsic symmetries, the introduction of an arbitrarily small forcing results in long-term symmetry breaking and leads to full synchronization. In this work we focus on the small forcing regime and recover the multiscale behavior of the two-point process. We show that in the first stage the model converges to an anti-polar configuration due to the symmetries of the RQF and in the second stage the two clusters meet due to the symmetry breaking phenomenon. The model is motivated by continuous-time machine learning models such as Neural ODEs and continuous-time formulations of transformers. In particular, the results of this work explain the role of the bias and the scale of its initialization. --- ### Bounded Semantic Planning and Deterministic Compilation for Reliable Enterprise Text-to-SQL Authors: Yi Ai Published: 2026-08-17 Categories: cs.DB, cs.AI arXiv: https://arxiv.org/abs/2608.16663v1 Direct text-to-SQL asks a language model to do two jobs: interpret the business question and construct the complete relational query. In enterprise schemas, SQL can execute successfully while using the wrong relationship role or aggregation grain. We study an alternative placement of the stochastic boundary. A multi-turn planner grounds phrases and selects from question-specific governed options; graph traversal, role predicates, grain lowering, SQL construction, and deterministic checks are implemented in code. We evaluate this semantic path compilation (SPC) system against direct DDL-to-SQL generation on the ACME insurance benchmark. On a 38-question adjudicated comparison set with three runs per question, SPC was adjudicated correct on every run for 37 questions (97.4%), compared with 21 (55.3%) for the baseline. The paired discordance was 16 questions in favor of SPC and none in favor of the baseline (two-sided exact McNemar p=3.05x10^-5). SPC answered all 38 questions correctly at least once and produced one refusal and no adjudicated wrong-but-executed run across 114 run outcomes; the baseline produced 29 adjudicated wrong runs and seven additional judge-flagged data-only coincidences on the same set. A strict-equivalence sensitivity analysis increased the paired difference. Additional SPC runs with GPT-5.4 and Gemini-3.6-Flash showed similar question-level robustness, although their per-run verdict artifacts were not preserved. Six additional benchmark items are retained in an all-item analysis and documented separately by failure class. The study supports an end-to-end systems result, not a causal claim that compilation alone produced the gain, because SPC receives governed semantic artifacts that the DDL baseline does not. --- ### Turning spectra into images improves plant trait retrieval with 2D-CNNs Authors: Javier Lopatin, Teja Kattenborn, Eya Cherif, Sebastián Moreno Published: 2026-08-17 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2608.16661v1 Hyperspectral reflectance spectroscopy enables non-destructive estimation of plant functional traits, yet current deep learning approaches process spectra as one-dimensional sequences, which limits how they capture long-range inter-band dependencies. We asked whether transforming 1D spectra into 2D image representations improves multi-trait prediction with convolutional neural networks (CNN). We compared nine transformations using EfficientNet-B0 on the GreenHyperSpectra dataset (7,897 labeled spectra, eight traits, 400-2450 nm), benchmarked against published 1D CNN results on the same split. Trained from scratch, the simplest transformation, a direct Reshape of the spectrum into a 2D grid, performed best ($R^2 = 0.684 \pm 0.001$) and improved on the state-of-the-art 1D baseline ($R^2 = 0.587$, $+0.097$). We then pretrained a 2D masked autoencoder (MAE-2D) on 139,000 unlabeled spectral images. Linear probing, which freezes the encoder and trains only a multilayer perceptron head, reached $R^2 = 0.646$ and exceeded every 1D self-supervised counterpart, including the fine-tuned MAE-1D ($R^2 = 0.641$). Under cross-dataset evaluation all models lost most of their accuracy and none beat the 1D baseline significantly. To identify which wavelengths drive each prediction, we applied Integrated Gradients and Grad-CAM and unfolded band importance back to the spectral axis. Protein ($r = 0.45$) and leaf water ($r = 0.33$) agreed with sensitivities simulated by the PROSAIL radiative-transfer model, while carotenoids ($r = 0.06$) and leaf area index ($r = -0.11$) did not, showing that the model reads established leaf chemistry for traits with sharp absorption features. The representational advantage of 2D spectral images, rather than architectural complexity or ImageNet pretraining, drives the gain over 1D approaches. --- ### Hoeffding adaptive splitting trees for data stream classification with concept drift and ensemble learning Authors: Daniel Nowak Assis, Jean Paul Barddal, Fabrício Enembreck Published: 2026-08-17 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2608.16659v1 Ensembles of decision trees are well-established methods for data stream classification. In ensemble learning, Hoeffding Trees are widely adopted as base learners, performing periodic split attempts according to the Hoeffding bound. Recent studies, however, indicate that this standard splitting mechanism lacks adaptability, while adaptive trees that trigger splits in response to performance degradation have achieved superior results. In this paper, we identify limitations in the use of adaptive-splitting decision trees as ensemble base learners, showing that change detectors often fail to promote sufficient diversity within ensembles. To address this issue, we propose two novel decision tree models, termed Hoeffding Adaptive Splitting Trees. These models combine the periodic splitting strategy of Hoeffding Trees, which fosters ensemble diversity, with adaptive splitting mechanisms that employ change detection algorithms to identify performance decay and determine split points. Experimental results demonstrate that Hoeffding Adaptive Splitting Trees enhance ensemble performance and achieve state-of-the-art results across a comprehensive evaluation, including benchmark comparisons, computational cost analysis, and concept drift adaptation. --- ### X$^2$Localizer: Cross-grained Alignment for Progressive Cross-view Video Geo-localization Authors: Zichao Zeng, Weijia Fan, Yufan Chen, June Moh Goo, Junwei Zheng, Ruiping Liu, Kunyu Peng, Jiaming Zhang, Rainer Stiefelhagen, Jan Boehm Published: 2026-08-17 Categories: cs.CV, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2608.16658v2 Cross-view Video Geo-localization (CVG) aims to localize ground-view videos by retrieving their corresponding geo-tagged aerial images. However, CVG approaches rely on fixed-length inputs and post-hoc refinement, hindering online-oriented localization under partial or dynamic observations. In this work, we formulate Progressive Cross-view Video Geo-localization (PCVG) as a deployment-oriented extension and evaluation protocol of CVG, enabling localization under varying temporal budgets, prefix-based inference, random-start evaluation, and long-range localization with interruptions. To explore PCVG, we introduce X$^2$Localizer, a cross-grained alignment framework that jointly supervises global prefix-to-aerial retrieval and token-aggregated frame--aerial-tile matching with a budget-dependent asymmetric objective. Furthermore, we introduce a Sliding-Window Re-Localization (SWRL) strategy that dynamically refreshes candidate regions for failure recovery and long-range deployment without full-sequence reprocessing. Extensive experiments show that X$^2$Localizer preserves conventional full-video performance, with marginal gains of +0.1 Recall@1 and +0.3 Recall@10, while substantially improving early localization. In the challenging single-frame setting, X$^2$Localizer improves coarse retrieval by +4.7 Recall@1 and +11.5 Recall@10 over the previous state-of-the-art method. With SWRL, our approach further enables robust progressive localization under random-start and long-distance scenarios, narrowing the gap between benchmark evaluation and real-world deployment. --- ### X$^2$Localizer: Cross-grained Alignment for Progressive Cross-view Video Geo-localization Authors: Zichao Zeng, Weijia Fan, Yufan Chen, June Moh Goo, Junwei Zheng, Ruiping Liu, Kunyu Peng, Jiaming Zhang, Rainer Stiefelhagen, Jan Boehm Published: 2026-08-17 Categories: cs.CV, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2608.16658v1 Cross-view Video Geo-localization (CVG) aims to localize ground-view videos by retrieving their corresponding geo-tagged aerial images. However, CVG approaches rely on fixed-length inputs and post-hoc refinement, hindering online-oriented localization under partial or dynamic observations. In this work, we formulate Progressive Cross-view Video Geo-localization (PCVG) as a deployment-oriented extension and evaluation protocol of CVG, enabling localization under varying temporal budgets, prefix-based inference, random-start evaluation, and long-range localization with interruptions. To explore PCVG, we introduce X$^2$Localizer, a cross-grained alignment framework that jointly supervises global prefix-to-aerial retrieval and token-aggregated frame--aerial-tile matching with a budget-dependent asymmetric objective. Furthermore, we introduce a Sliding-Window Re-Localization (SWRL) strategy that dynamically refreshes candidate regions for failure recovery and long-range deployment without full-sequence reprocessing. Extensive experiments show that X$^2$Localizer preserves conventional full-video performance, with marginal gains of +0.1 Recall@1 and +0.3 Recall@10, while substantially improving early localization. In the challenging single-frame setting, X$^2$Localizer improves coarse retrieval by +4.7 Recall@1 and +11.5 Recall@10 over the previous state-of-the-art method. With SWRL, our approach further enables robust progressive localization under random-start and long-distance scenarios, narrowing the gap between benchmark evaluation and real-world deployment. --- ### Orbit-Planner: Towards Latent World Models for On-Orbit Obstacle Avoidance of Satellite Agents Authors: Zhijian Li, Chao Ren, Peijin Wang, Xian Sun Published: 2026-08-17 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2608.16651v1 Satellite agents for on-orbit navigation tasks need to predict collision risks using limited onboard observations. However, conventional planners often rely on predefined maps and fixed environmental assumptions, limiting their adaptability in dynamic on-orbit scenarios. In this paper, we propose Orbit-Planner, a two-stage latent world model for on-orbit obstacle avoidance. Orbit-Planner learns action-conditioned spacecraft dynamics to perform future-state rollouts in latent space, and introduces a Physics Probe to decode physical state changes from imagined latent trajectories. Experiments demonstrate that Orbit-Planner can perform long-horizon latent rollouts and recover physical states from imagined trajectories. In closed-loop obstacle-avoidance navigation in Isaac Sim, it attains a success rate of 91.7%. Code is available at https://github.com/ZhijianLi2003/Orbit_Planner. --- ### PCA-guided Activation Scaling for Monotonic Bidirectional Control over LLM Sycophancy Authors: Zheng Chen, Zhaoxin Feng, Yip Tin Po, Jianfei Ma, Emmanuele Chersoni, Bo Li Published: 2026-08-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.16650v1 Large language models (LLMs) exhibit sycophancy, a tendency to agree with user beliefs regardless of factual accuracy. This can reinforce misconceptions, but eliminating it entirely risks over-correction against valid opinions. Effective control must therefore both reduce and increase sycophancy with predictable and gradual effect. Yet, existing methods fail to ensure a bidirectional and monotonic relationship between steering strength and behavioral outcome across models and datasets. We introduce PCA-guided Activation Scaling (PAS), an activation steering framework that decomposes residual stream activations into a PCA-identified sycophancy-honesty subspace and an orthogonal residual, then applies distinct scaling exponents to achieve monotonic, bidirectional control. Across three LLMs and three datasets, PAS achieves strong monotonicity (Spearman $ρ$ = +0.92) and an average shift of 15.4% per direction, compared with 8.7% for the baselines. Ablation studies confirm that the decomposition, asymmetric exponents, and layer selection are each essential for maintaining monotonic control. The data and code are available at https://github.com/Bellafc/PCS. --- ### Every Coin Has Two Sides: On the Dual Nature of Generalization in On-Policy Distillation of Large Language Models Authors: Zhaoyi Li, Deyang Kong, Yuan Wei, Evan Yang, Ranran Shen, Mahardika Krisna Ihsani, Ming Yang, Wei Zhang, Chuan Hao, Jian Yang, Ran Tao, Bryan Dai et al. Published: 2026-08-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.16647v2 On-policy distillation (OPD) transfers teacher capabilities by supervising trajectories sampled from the student's own policy, yet its generalization behavior remains poorly understood, as most studies evaluate OPD on a single domain and on benchmarks close to the training data. We present a controlled study that varies one generalization factor at a time, from in-domain distribution shifts to cross-domain transfer and the multi-teacher setting. We find that OPD transfers a teacher's reasoning behavior rather than its answers to particular problems: training difficulty barely matters, and even problems the teacher never solves are useful. Transfer depends strongly on the origin relationship between teacher and student: same-origin pairs bring the student close to the teacher across languages, reasoning horizons, and even other domains, whereas cross-origin pairs mostly fit the trained distribution. This broad reach is a double-edged sword: since routing prompts to domain experts cannot confine each teacher's influence, combining them yields a mixture-dependent seesaw among their capabilities. These results clarify when OPD generalizes and offer a useful perspective for diagnosing multi-teacher OPD. --- ### Every Coin Has Two Sides: On the Dual Nature of Generalization in On-Policy Distillation of Large Language Models Authors: Zhaoyi Li, Deyang Kong, Yuan Wei, Evan Yang, Ranran Shen, Mahardika Krisna Ihsani, Ming Yang, Wei Zhang, Chuan Hao, Jian Yang, Ran Tao, Bryan Dai et al. Published: 2026-08-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.16647v1 On-policy distillation (OPD) transfers teacher capabilities by supervising trajectories sampled from the student's own policy, yet its generalization behavior remains poorly understood, as most studies evaluate OPD on a single domain and on benchmarks close to the training data. We present a controlled study that varies one generalization factor at a time, from in-domain distribution shifts to cross-domain transfer and the multi-teacher setting. We find that OPD transfers a teacher's reasoning behavior rather than its answers to particular problems: training difficulty barely matters, and even problems the teacher never solves are useful. Transfer depends strongly on the origin relationship between teacher and student: same-origin pairs bring the student close to the teacher across languages, reasoning horizons, and even other domains, whereas cross-origin pairs mostly fit the trained distribution. This broad reach is a double-edged sword: since routing prompts to domain experts cannot confine each teacher's influence, combining them yields a mixture-dependent seesaw among their capabilities. These results clarify when OPD generalizes and offer a useful perspective for diagnosing multi-teacher OPD. --- ### Training-Free Reconstruction-Based AI-Generated Image Detectors Are Inherently Vulnerable to Adversarial Examples Authors: Roman Demchenko, Jonas Ricker, Asja Fischer Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16646v1 The impressive visual quality and ubiquity of AI-generated images call for reliable and robust detection methods. Reconstruction-based detectors have emerged as a promising direction for transparent and training-free identification of synthetic images. However, due to their fundamentally different mode of operation (compared to standard, classifier-based methods), little is known about their adversarial robustness. In this work, we propose two novel attack methods targeted at detectors that leverage autoencoder reconstruction error. We find that by constructing imperceptible adversarial examples, the distance between original and reconstruction can be artificially increased, causing fake images to be wrongly classified as real. Our evaluation including images from three state-of-the-art generators and three detectors demonstrates that detection performance is significantly decreased, even if attacked images additionally undergo real-world degradations. Critically, our adversarial examples naturally transfer across detectors, as they all share the same principle, pointing towards an inherent vulnerability of reconstruction-based detectors. --- ### Reconstruction: A Blind Benchmark for Recovering Research Ideas from Pre-Publication Bibliographies Authors: Shaolong Chen, Yanlin Fei, Nazhou Liu, Xinmiao Yu, Lei Li, Rahul Thapa, Madalina Ciobanu, Navan Preet Singh, Qingqing Mao, Ritankar Das Published: 2026-08-17 Categories: cs.AI, cs.CL, cs.MA arXiv: https://arxiv.org/abs/2608.16645v3 Can a language model recover the true research idea of a published paper when given only that paper's pre-publication bibliography? We introduce Reconstruction, a blind idea-recovery benchmark that withholds the seed paper and all contemporaneous or future literature, and asks models to propose hypotheses that an independent large language model judge matches against the held-out ground-truth idea. A strict anti-leakage protocol-temporal citation cutoff, anonymous reference IDs, and frozen per-paper bibliographies, which prevents prompt-time leakage of the seed idea. Across six scientific domains and 643 evaluated papers, seven frontier models achieve only modest Match rates (approx. 3-15%). We then evaluate a reference-only multi-agent (top 4) pipeline that combines cross-model review with a Swiss tournament over aligned hypothesis slots, without external web search. Cross-model review plus tournament selection raises Match rates to approx. 23-42% across all six domains, which is an observed approx. 2.4x lift over the best single-model baseline. This draft reports the protocol, anti-leakage design, and current results as an arXiv timestamp. --- ### Reconstruction: A Blind Benchmark for Recovering Research Ideas from Pre-Publication Bibliographies Authors: Shaolong Chen, Yanlin Fei, Nazhou Liu, Xinmiao Yu, Lei Li, Rahul Thapa, Madalina Ciobanu, Qingqing Mao, Ritankar Das Published: 2026-08-17 Categories: cs.AI, cs.CL, cs.MA arXiv: https://arxiv.org/abs/2608.16645v1 Can a language model recover the true research idea of a published paper when given only that paper's pre-publication bibliography? We introduce Reconstruction, a blind idea-recovery benchmark that withholds the seed paper and all contemporaneous or future literature, and asks models to propose hypotheses that an independent large language model judge matches against the held-out ground-truth idea. A strict anti-leakage protocol-temporal citation cutoff, anonymous reference IDs, and frozen per-paper bibliographies, which prevents prompt-time leakage of the seed idea. Across six scientific domains and 643 evaluated papers, seven frontier models achieve only modest Match rates (approx. 3-15%). We then evaluate a reference-only multi-agent (top 4) pipeline that combines cross-model review with a Swiss tournament over aligned hypothesis slots, without external web search. Cross-model review plus tournament selection raises Match rates to approx. 23-42% across all six domains, which is an observed approx. 2.4x lift over the best single-model baseline. This draft reports the protocol, anti-leakage design, and current results as an arXiv timestamp. --- ### Toward Better Assessment of LLMs' Performance in Clinical Error Detection Authors: Yifan Zhang, Rahmatollah Beheshti Published: 2026-08-17 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.16643v1 Automated detection of errors in clinical documentation is a promising application of large language models (LLMs), yet decisions to deploy such models rest on benchmarks that evaluate each clinical note in isolation. Error-detection benchmarks are typically constructed by injecting errors into notes, such that each erroneous note has a natural counterpart. Aggregate discriminative metrics (e.g., balanced accuracy or F1) do not exploit this structure. We show that this omission is consequential. In particular, evaluating 15 diverse LLMs on 4 standardized clinical error-detection test sets across 3 languages, we find that 13 of 15 models fall below the level of random pairwise discrimination, even while achieving F1 scores that standard practice would read as moderate. We also observe that the underlying bias patterns differ across languages: the same model can default to "no error" on one language and over-flag errors on another. To diagnose where discrimination breaks down, we further introduce a procedure to score the evidence models cite in their outputs. We find that while models consistently locate error-relevant content, they fail to produce the corresponding correct verdict on the clean counterpart. Finally, we show that F1 and pairwise accuracy are driven in opposite directions by the same underlying bias, so that ranking models by F1 may systematically promote the weakest discriminators. For safety-critical clinical NLP applications, we advocate for supplementing aggregate metrics with paired evaluations in benchmark reporting. Code and analysis scripts are available at https://github.com/healthylaife/paired-clinical-eval. --- ### PDDLCoder: Agentic PDDL Generation for LLM-Assisted Symbolic Planning Authors: Veit Laule, Jiangtao Shuai, Manfred Hauswirth, Sonja Schimmler Published: 2026-08-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.16637v1 LLMs remain unreliable for long-horizon planning, often generating logically inconsistent or non-applicable plans. Recent hybrid methods instead translate natural language into the Planning Domain Definition Language (PDDL), allowing symbolic planners to produce verifiable plans. However, existing methods frequently rely on rigid generation pipelines, a partial PDDL definition, or human feedback. Furthermore, their evaluation is hindered by the lack of standardized benchmarks with automated verification. To address these limitations, we present PDDLCoder, an agentic framework for PDDL generation from natural language that iteratively generates, analyzes, and refines planning specifications. We further introduce NL-pddlgym, a benchmark dataset comprising 711 planning problems across 23 domains with executable gym environments for the automated verification of plan applicability. Experiments on the NL-pddlgym test set containing 106 problems across 4 held-out domains show that PDDLCoder generates applicable plans for 89.6\% of tested planning problems. This improves upon our adaptations of previous PDDL generation methods, which achieved up to 45.3\%, and outperforms direct LLM planning approaches, which reached up to 74.5\% on the same test set. Our work demonstrates the effectiveness of agentic PDDL generation for planning and establishes a reproducible benchmark for future research on LLM-assisted symbolic planning. --- ### DRAFE: Domain-Robust Asymmetric Fusion of Heterogeneous Detection Transformers for Cross-City Fine-Grained Traffic Object Detection Authors: Divine Yao Agbobli, Geoffery Eyram Agorku, Israel Afriyie, Kwadwo Amankwah-Nkyi, Marvin Osei-Kuffour, Richmond Owusu Duah, Bright Seglah, Kelvin Asamoah Terkper, Kwabena Amoako Adjei Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16632v1 Deep learning-based object detectors are fundamental to intelligent transportation systems, enabling traffic monitoring, vehicle analytics, and infrastructure management. However, achieving both fine-grained vehicle recognition and robust cross-city domain generalization remains challenging. We present the Domain-Robust Asymmetric Fusion Ensemble (DRAFE), which combines independently trained LW-DETR and RF-DETR detectors for cross-city fine-grained traffic object detection. DRAFE employs a two-stage training strategy that first pretrains complementary detectors on diverse public traffic datasets using pseudo-label expansion and human-in-the-loop annotation refinement, producing a curated corpus of 6,049 images and 203,619 annotations, before challenge-compliant fine-tuning on the Project Hafnia Track 6 dataset. At inference, DRAFE applies anchor-conditioned class-consistent matching, reliability-weighted coordinate fusion, agreement-aware confidence recalibration, and complementary hypothesis recovery. On AI City Challenge 2026 Track 6, DRAFE achieves 0.4022 mAP, ranks sixth among 25 participating teams, and improves by 0.0553 mAP over a preliminary ensemble evaluated under identical benchmark conditions. --- ### The Working Set of a Coding Agent: Coherence Debt in Repository-Scale Tasks Authors: Bardia Mohammadi, Lars Klein, Aman Chadha, Akhil Arora, Laurent Bindschaedler Published: 2026-08-17 Categories: cs.SE, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.16630v1 Repository-scale coding requires an agent to keep tests, imports, configuration, and migration rules consistent within a bounded context window. We model this as reconstructing a coupled-fact graph: at each edit, a required fact comes from recent context or parametric memory, and the facts covered by neither form coherence debt. We supply and withhold each channel and inject faults across seven models and five harnesses. As expected, no model completes a task on an unseen API with both channels empty, and putting the facts in the prompt restores success. When a rename defeats what models memorized about a real library, all seven fail in the same place, passing and missing the same tests. Availability decides the outcome and distance does not: withholding a fact costs exactly the work it supports, and a supplied fact works as well far from the edit as next to it. Harnesses pay unequal prices for it: configurations that all pass every test differ more than tenfold in tokens consumed because they rebuild the same content at different rates, and spending more recovers nothing when facts are withheld. A missing fact produces wrong work rather than absent work: an agent asked to act acts, fabricating the file or guessing the value, so instruments built on reads look for a hole already filled. How often it says it is blocked instead is a property of the model, from every trial to none. Availability does not settle every edit: where standard and code disagree, agents follow the standard even when it prescribes the worse code, so a stale convention file costs more than no file. Because parametric memory substitutes for reading, on SWE-bench, where models likely know the repositories, reads no longer predict success. Harnesses should keep the facts an edit depends on available when the agent writes, and check that availability against what the agent produces rather than what it reads. --- ### Hypergraph-based Multimodal Retrieval-Augmented Generation with Incremental Refinement Authors: Shenao Chen, Yidan Xu, Xiangmin Han, Rundong Xue, Duanpo Wu, Yuhan Gao, Chenggang Yan, Yue Gao Published: 2026-08-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.16628v1 Modern Multimodal Retrieval-Augmented Generation (M-RAG) systems are fundamentally limited by the binary connectivity paradigm of traditional simple graphs, which fails to capture the intricate, high-order correlations among heterogeneous entities, such as the N-ary relationships between a visual chart, its scattered textual descriptions, and underlying numerical data. Furthermore, existing refinement strategies often rely on exhaustive, full-page reconstruction to align cross-modal information, leading to prohibitive computational redundancy and the introduction of contextual noise in long-form document processing. In this paper, we propose Hyper-M2RAG, a novel framework that redefines multimodal document retrieval through High-order Hypergraph Representation Learning. We first formalize the document structure as a Multimodal Hypergraph, utilizing hyperedges as unified semantic containers to encapsulate multi-way associations across text, images, and tables, thereby transcending point-to-point modeling. To mitigate semantic fragmentation caused by physical pagination, we introduce an Anchor-driven Incremental Refinement mechanism. Rather than performing a global sweep, our approach identifies boundary-crossing anchor nodes and reconstructs their local hyper-topology using one-hop neighborhood contexts. This targeted refinement effectively bridges cross-page knowledge gaps with minimal computational footprints. Extensive evaluations on multimodal benchmarking datasets demonstrate that Hyper-M2RAG significantly outperforms state-of-the-art methods in both retrieval precision and generation coherence. Our code is available at https://github.com/ShenAoChen2001/MMHRAG. --- ### When Do Explanations Help In-Context Learning? A Comparative Study of Natural Language Explanation Types and Faithfulness Authors: Mahdi Dhaini, Adam Dejl, Juraj Vladika, Volkan Özer, Barbara Plank, Gjergji Kasneci Published: 2026-08-17 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.16627v1 Natural language explanations (NLEs) are increasingly used as inputs, for example, as few-shot rationales that influence model behavior in in-context learning (ICL). However, it remains unclear how different types of NLEs compare in their effects on downstream model performance in explanation-augmented prompting. Therefore, we provide a comparative evaluation across six benchmarks and four instruction-tuned models, studying how NLE source (human-written when available, self-generated explanations, generated by an external LLM) and NLE selection (random vs faithfulness-based filtering) affect downstream utility of NLEs when used in ICL settings. Our extensive evaluation shows that, on classification-style benchmarks, adding NLEs to few-shot prompts often improves accuracy over few-shot prompting without explanations; among NLE sources, externally generated LLM-NLEs often provide strong downstream utility and remain competitive with human rationales where both are available, whereas self-NLEs are more sensitive to the selection strategy. On math reasoning, the effects are more model- and source-dependent. We further show that faithfulness-based selection of self-NLEs yields small average gains overall, but can improve or reduce performance depending on the metric, task, and model. Different faithfulness metrics can disagree substantially, affecting which self-NLE examples are selected and their downstream predictive utility. Robustness tests with randomly swapped and out-of-distribution rationales indicate partial robustness, suggesting that semantic alignment contributes to performance gains. Overall, our results provide insights for selecting and reporting explanations that influence model behavior in practical prompting pipelines. --- ### A Shop Floor Production Scheduling Case based on RFID-supported Smart Factory Authors: Zhihui Chen, Yize Sun, Yuhao Dong, Zeyu Xiao, Ray Y. Zhong Published: 2026-08-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.16626v1 Radio frequency identification (RFID) technology has been widely implemented for real-time data collection in manufacturing shop floors, which, in turn, can be used to support dynamic shop floor production planning and scheduling. Within such an environment, uncertainty in operation and production processes collectively contribute to the dynamicity in manufacturing, thereby hampering the scheduling system from achieving maximal utility. To highlight the importance of handling such uncertainty, this paper addresses the problem of dynamic shop floor scheduling for a real-life case smart factory equipped with RFID technology. Feasible production sequence mining and real-time processing rate estimation are conducted on RFID-collected production data to quantify the operation and production uncertainties. A deep reinforcement learning approach based on the RFID data analysis is then presented for shop floor production scheduling. Simulation studies based on real-life case data have demonstrated the feasibility and practicality of the proposed dynamic production scheduling framework. Specifically, it is observed that the proposed framework outperforms existing dispatch methods in terms of minimizing operation makespan, including first in first out (FIFO), last in first out (LIFO) and deep Q network (DQN). --- ### HarmTrace: Anchor-Calibrated Decoupled Optimization for Fine-Grained Target Identification in Harmful Memes Authors: Yujia Li, Yiqun Zhang, Zihan Cheng, Yijie Huang, Tenglong Ye, Zihan Wang, Xiaocui Yang, Shi Feng, Yifei Zhang, Daling Wang Published: 2026-08-17 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.16622v1 Multimodal harmful meme detection is typically formulated as image--text harmfulness classification. A model may correctly predict harmfulness while misidentifying the attacked target or its supporting evidence. We therefore extend harmful meme detection with fine-grained target identification, asking what type of target is attacked, who is targeted, and where the target appears in the meme. The model predicts harmfulness for every meme and, for harmful memes, outputs the target category, target entity, textual mention, and visual region. To support this task, we introduce Meme3W, which unifies multiple public harmful meme datasets and provides human-verified annotations for harmful instances. We further introduce Joint Record Accuracy (JRA), a strict record-level metric requiring the harmfulness label and all target-identification fields to be jointly correct. Experiments with representative multimodal large language models reveal a substantial gap between harmfulness accuracy and JRA. To narrow this gap, we propose HarmTrace, an anchor-calibrated decoupled optimization framework. HarmTrace strengthens target-entity supervision through entity-aware supervised fine-tuning. It then applies Conditional Target-identification Policy Optimization (CTPO) to decouple harmfulness and target-identification advantages, restricting target-identification optimization to label-correct responses for harmful examples. CTPO uses a Virtual Positive Anchor (VPA) as a fully correct reference for target-identification advantage normalization. HarmTrace improves both JRA and harmfulness accuracy across the evaluated backbones, with JRA on the Qwen3-VL-8B backbone increasing from 17.58\% to 52.51\%. Our code is publicly available at https://github.com/llly1234/HarmTrace-for-Harmful-Memes. --- ### Cost Scales with Change, Not Corpus Size: Incrementally Maintaining an Evolving Semantic Substrate Authors: Yusuke Takahashi, Kyle Wild, Asako Uraki Published: 2026-08-17 Categories: cs.AI, cs.DB, cs.IR arXiv: https://arxiv.org/abs/2608.16621v1 Retrieval-augmented and agentic question-answering systems increasingly re-derive the meaning of a corpus at query time. Put plainly, instead of re-deriving what a corpus means on every question, the work is done once when a document arrives and is thereafter merely consulted -- a compiler, not an interpreter, of meaning. An alternative is to compile that meaning once, at ingest time, into a compact, queryable semantic substrate and maintain it as the corpus evolves. The central objection is maintenance cost: rebuilding a truncated singular value decomposition (SVD) on every change appears prohibitive, and a change of embedding model seems to force a full re-embedding. We argue and show empirically that maintenance cost scales with the amount of change, not corpus size. On a controlled synthetic pilot (dimension 256, rank 32, a corpus grown from 3,000 to 9,000 documents over 50 update events), incremental low-rank updates were 33.7 times cheaper per update than full re-SVD and 23.8 times cheaper cumulatively, while the incremental subspace tracked the full recomputation to within floating-point precision (maximum principal-angle drift below 1e-11 degrees; recall@10 = 1.0). An orthogonal Procrustes virtual axis update recovered 0.95 mean cosine to truly re-embedded vectors by re-embedding only about 10 percent of the corpus. The results support maintaining, rather than repeatedly reconstructing, a semantic substrate. --- ### Palmyra x6 Technical Report: An Agentic, Tool-Use Model Post-Trained via Anchored Supervised Fine-Tuning Authors: Peng Du, Kiran Kamble, Rakshith Vasudev, Zhizhuo Yang, Rohith Nadimpally, Arjun Krishna, Waseem Alshikh, Daniel M. Bikel Published: 2026-08-17 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.16620v3 Palmyra x6 is a large language model optimized for use with enterprise-oriented agentic tasks. The model was built by post-training a Mixture-of-Experts base model with Anchored Supervised Fine-Tuning on a compact corpus of verified, synthetic tool-use trajectories, optimized with a Muon + Adam hybrid. The recipe is deliberately conservative and deliberately controlled: 626 trajectories, a single epoch, a low learning rate, and a KL anchor to the frozen base. The model shows substantial gains over the previous default model for Writer Agent, and compares favorably with several recent models on public benchmarks, scoring the highest on BFCL Core at $0.785$ and posts the highest six-benchmark mean of the cohort. Furthermore, the model has shown itself to be competitive or leading relative to comparators in our bias and safety evaluations. --- ### Palmyra x6 Technical Report: An Agentic, Tool-Use Model Post-Trained via Anchored Supervised Fine-Tuning Authors: Peng Du, Kiran Kamble, Rakshith Vasudev, Zhizhuo Yang, Rohith Nadimpally, Arjun Krishna, Waseem Alshikh, Daniel M. Bikel Published: 2026-08-17 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.16620v1 Palmyra x6 is a large language model optimized for use with enterprise-oriented agentic tasks. The model was built by post-training a Mixture-of-Experts base model with Anchored Supervised Fine-Tuning on a compact corpus of verified, synthetic tool-use trajectories, optimized with a Muon + Adam hybrid. The recipe is deliberately conservative and deliberately controlled: 626 trajectories, a single epoch, a low learning rate, and a KL anchor to the frozen base. The model shows substantial gains over the previous default model for Writer Agent, and compares favorably with several recent models on public benchmarks, scoring the highest on BFCL Core at $0.785$ and posts the highest six-benchmark mean of the cohort. Furthermore, the model has shown itself to be competitive or leading relative to comparators in our bias and safety evaluations. --- ### Beyond Accuracy: Assessing Calibration of Geospatial Foundation Models and Their Sensitivity to Distribution Shifts Authors: Nils Lehmann, Jakob Gawlikowski, Burak Ekim, Isaac Corley, Xiao Xiang Zhu Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16614v1 Geospatial Foundation Models (GeoFMs) are most commonly ranked and selected by accuracy on standard benchmark conditions via averaged ranks. We show that this protocol is too narrow: the promised deployment in critical EO tasks requires further angles of analysis, mainly calibration, the agreement between a model's confidence and its correctness. Across 16 frozen encoders, four classification and five segmentation datasets, and two orthogonal stress axes, every encoder degrades as corruption intensifies, and the ranking changes as well. Across the four classification benchmarks, EO-pretrained and ImageNet-pretrained encoders are indistinguishable on clean accuracy and clean calibration, and EO pretraining provides no more stability under shift than ImageNet pretraining. Under shift the GeoFMs drift further into overconfidence than the ImageNet-pretrained encoders, at every grade and in every corruption family. A centered kernel alignment (CKA) analysis ties this to representational rigidity: EO-pretrained embeddings move less under corruption while losing just as much task information and remaining overconfident. We apply three commonly explored uncertainty quantification methods and find that temperature scaling and deep ensembles cannot counteract the degradation, while a Gaussian-process probe roughly halves ECE under severe cloud only by tripling it on clean data. In selective prediction experiments, we find that confidence-based abstention cannot defer around confidently wrong predictions, and advocate that benchmark rankings and evaluations should therefore operate across a multitude of conditions and metrics to more holistically evaluate model development progress and close the gap to real world deployment scenarios. --- ### Degradation-Aligned Self-Supervised Learning for State of Health Estimation of Lithium-Ion Batteries under Label Sparsity Authors: Jiaqi Yao, Julia Kowal Published: 2026-08-17 Categories: eess.SP, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.16612v2 An accurate estimation of the state of health (SOH) underpins safe and optimized use of the battery system. Although compelling, data-driven SOH estimation models typically require large amounts of high-quality labeled cycling data, while in practice such labels are often sparse in both quantity and coverage. Therefore, in this work, we propose a degradation-aligned self-supervised learning (SSL) framework based on a convolutional neural network-gated recurrent unit (CNN-GRU) model, which learns aging-consistent representations from unlabeled data through a cycle-order ranking objective as the pretext task for pretraining, thereby enabling robust SOH estimation after fine-tuning on sparsely labeled data. Test results showcase that the proposed ranking-based SSL approach proves to endow the pretrained model with degradation awareness from unlabeled data, and after fine-tuning the model can carry out accurate, robust SOH estimation, even when only an extremely limited amount of 1% of unevenly distributed labeled training data is available, where the MAE of 1.718% and RMSE of 2.329% can be achieved on the test cell. In addition, in-depth analyses are presented regarding the influences of label distribution and cross-cell robustness. We believe this work could shed new light on label-efficient SOH estimation of lithium-ion batteries, addressing a practical need in battery management. --- ### Degradation-Aligned Self-Supervised Learning for State of Health Estimation of Lithium-Ion Batteries under Label Sparsity Authors: Jiaqi Yao, Julia Kowal Published: 2026-08-17 Categories: eess.SP, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.16612v1 An accurate estimation of the state of health (SOH) underpins a safe and optimized use of the battery system. Although compelling, data-driven SOH estimation models typically require large amounts of high-quality labeled cycling data, while in practice such labels are often sparse in both quantity and coverage. Therefore, in this work, we propose a degradation-aligned self-supervised learning (SSL) framework based on a convolutional neural network-gated recurrent unit (CNN-GRU) model, which learns aging-consistent representations from unlabeled data through a cycle-order ranking objective as the pretext task for pretraining, thereby enabling robust SOH estimation after fine-tuning on sparsely labeled data. Test results showcase that the proposed ranking-based SSL approach proves to endow the pretrained model with degradation-aligned information from unlabeled data, and after fine-tuning the model can carry out accurate, robust SOH estimation, even when only an extremely limited amount of 1% of unevenly distributed labeled training data is available, where the MAE of 1.718% and RMSE of 2.329% can be achieved on the test cell. In addition, in-depth analyses are presented regarding the influences of label distribution of battery degradation data. We believe this work could shed new light on SOH estimation of lithium-ion batteries under label sparsity in real-world applications. --- ### Interactive Whole Slide Images for RL-based Tumour Segmentation Authors: Mohamad Mohamad, Francesco Ponzio, Maxime Gassier, Nicolas Pote, Xavier Descombes Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16607v1 Whole-slide image (WSI) analysis remains computationally challenging due to the extremely large spatial resolution of slides and the sparse distribution of tumour regions. We propose an end-to-end reinforcement learning framework for sequential tumour segmentation directly on WSIs. Instead of treating the slide as a predefined collection of candidate patches, we formulate the WSI itself as a hierarchical multi-resolution environment through which an agent navigates using movement, zooming, and tumour selection actions. The agent jointly processes local observations and a global thumbnail representation within an actor-critic architecture trained using proximal policy optimization (PPO). Experiments on pulmonary adenocarcinoma WSIs demonstrate the feasibility of direct sequential tumour segmentation on full slides, achieving comparable coarse segmentation quality relative to patch-based approaches operating at similar magnification levels, while reducing inference time to a few seconds per slide. We further analyse the impact of environment design and action-space granularity. Our results suggest that modelling WSIs as interactive environments provides a promising direction for RL-based computational pathology --- ### Variational Outlier-Robust Gaussian Process Regression with Generative Modeling Authors: Arslan Majal, Aamir Hussain Chughtai Published: 2026-08-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.16606v1 Outliers can substantially distort Gaussian process regression (GPR) due to its conventional Gaussian observation likelihood, leading to inaccurate model learning and prediction. To address this limitation, this article introduces a generative GPR model that captures observation-specific contamination and adaptively mitigates the influence of outliers. Subsequently, a variational generalized expectation-maximization procedure is used to learn the latent variables and GPR model parameters. Experiments on synthetic and real datasets under different contamination settings demonstrate that the proposed method remains competitive with-and in several cases outperforms-robust GPR baselines in prediction accuracy. Moreover, the proposed method shares the cubic computational scaling of the compared GPR methods. --- ### GeoPose: Patient-agnostic CTA-to-DSA registration through projection-space calibration Authors: Rudolf L. M. van Herten, Robert Graf, Paula Feldman, Johannes C. Paetzold Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16600v1 Aligning intraoperative biplanar digital subtraction angiography (DSA) to pre-procedural computed tomography angiography (CTA) requires rapid and accurate 3D-to-2D registration. Optimization-based methods are sensitive to initialization and may require hundreds of iterations, whereas learning-based approaches commonly rely on patient-specific training. We propose GeoPose, a population-trained framework that estimates the C-arm pose in a learned canonical frame and transfers it to the native frame of an unseen CTA through projection-space calibration and transform composition. A population-trained residual network refines the pose, followed optionally by low-budget image-driven optimization. GeoPose requires neither patient-specific adaptation nor explicit inter-volume preregistration. On 80 DSA observations from 20 held-out patients, optimization-free GeoPose achieved a carotid mean projected centerline distance (mPCD) of 5.8 mm and a clDice of 0.45, compared with 14.5 mm and 0.28 for the best-performing baseline, while requiring only 0.15 s. After 25 optimization iterations, GeoPose reached an mPCD of 4.6 mm and a clDice of 0.58 in approximately two seconds. Under the same budget, native-initialized optimization achieved 14.6 mm and 0.15, respectively. GeoPose thus provides rapid native-frame registration with fixed population-level weights and the geometric correspondence required for downstream biplanar 3D vascular reconstruction. --- ### CACSurv: Concordance-Aligned Comparative Learning with Large Language Models for Cancer Survival Prediction Authors: Tianqi Xiang, Qixiang Zhang, Xinpeng Ding, Yi Li, Xiaomeng Li Published: 2026-08-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.16594v1 Cancer survival prediction supports treatment planning, risk stratification, and follow-up management. Existing methods use structured clinical variables, whole-slide images, genomic profiles, or multimodal inputs, while patient reports remain underexplored. We study report-centric survival prediction using reports that organize pathological, clinical, and molecular evidence. Large language models (LLMs) can reason over such reports, but case-wise time regression introduces two mismatches. First, a formulation mismatch arises because survival evaluation depends on ordering comparable patients, whereas independent time predictions do not enforce ranking consistency. Second, a supervision mismatch arises because a censored patient's observed time indicates survival beyond that point and cannot serve as an exact regression target, although it still implies orderings relative to patients who died earlier. To address these mismatches, we propose CACSurv, a Concordance-Aligned Comparative framework for report-centric survival prediction. CACSurv reformulates survival modeling as mini-cohort comparative reasoning, where an LLM predicts relative prognostic orderings. We introduce concordance-aligned rewards derived from comparable relations under right censoring, enabling censored outcomes to provide ranking supervision without exact event-time targets. At inference, Monte Carlo Reference Aggregation compares each patient with sampled references and aggregates positions into a cohort-level ranking. We establish TCGA-SurvReport, a benchmark covering six TCGA cancer cohorts. CACSurv achieves the highest C-index on all six cohorts and an average C-index of 0.722, outperforming the strongest published survival model by 6.5 percentage points and the strongest LLM time-regression baseline by 4.2 percentage points. Our code, models, and dataset will be available at https://github.com/xmed-lab/CACSurv. --- ### Towards Zero-Shot Domain Generalization for ID Cards Presentation Attack Detection Authors: Mario Nieto-Hidalgo, Juan M. Espin, Juan E. Tapia Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16591v1 Presentation-Attack Detection (PAD) for national ID cards is limited by the lack of publicly available genuine samples, making it difficult for systems to generalize across countries. This paper introduces two main innovations: (1) a Prototypical Network head using an EfficientNet-V2-b0 backbone that requires only four genuine samples per class to create reliable prototypes; and (2) an episodic training regime that keeps PAD classes fixed while varying the card domain, allowing the network to learn universal attack cues. Evaluated on a large multi-country dataset and the public DLC-2021 benchmark, this method achieves an average Equal Error Rate of around 9\%, outperforming conventional softmax and CLIP zero-shot baselines even with data from a single source country. This approach provides accurate, privacy-preserving PAD while minimizing data collection, facilitating scalable cross-jurisdictional remote onboarding. --- ### Ultra: Unsupervised Cross-Task Optimization for Reliable Restoration Segmentation Collaboration under Adverse Weather Authors: Shiqin Wang, Zhiqian Li, Haoyuan Du, Junming Chen, Jiayuan Li, Tianrun Xu, Haoyang Chen Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16589v1 Unsupervised Domain Adaptation for Adverse Weather Semantic Segmentation (UDA-ASS) aims to transfer semantic knowledge from labeled normal-weather images to unlabeled adverse environments. Existing approaches implicitly assume that restoration and segmentation provide mutually beneficial guidance. However, under severe degradation and without target-domain supervision, the validity of cross-task optimization directions becomes fundamentally unidentifiable, leading to hallucination-driven error propagation. In this work, we propose a novel Unsupervised Restoration-Segmentation Collaborative Learning Framework (Ultra), which reframes cross-task interaction as direction selection under uncertainty and causal effect estimation, enabling reliable collaboration through candidate direction generation and intervention-based filtering. In detail, we propose CTDN and CMIL. The former exploits complementary visual structures and semantic information to generate candidate optimization directions and performs cooperative direction selection between restoration and segmentation. The latter reformulates cross-task information transfer from correlation-based propagation into causal effect assessment, suppressing hallucination propagation. Extensive experiments on three widely used UDA-ASS benchmarks demonstrate state-of-the-art segmentation performance. Beyond segmentation, our framework achieves better unsupervised restoration results than existing UDA-ASS restoration methods and generalizes to unsupervised restoration and object detection collaboration tasks. Code and models will be available at https://github.com/Wang-Shiqin/Ultra. --- ### SQuad: Sub-Quadratic Attention Distillation for Efficient Video Generation Authors: Animesh Karnewar, Denis Korzhenkov, Amirhossein Habibian, Mohsen Ghafoorian Published: 2026-08-17 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.16585v1 Video Diffusion Transformers (DiTs) spend most of their compute inside the Self-Attention operation, whose cost grows quadratically, $\mathcal{O}(n^2)$, with the number of latent tokens $n$. For the task of video generation, the token count is large, so this term dominates runtime and memory, and thereby caps the resolution and duration we can generate. Linear $\mathcal{O}(n)$ and low-rank $\mathcal{O}(nk)$ surrogates of Self-Attention trade the full softmax $QK^T$ for cheaper kernels, but rarely recover the original's expressivity, leaving a stubborn quality gap. Motivated by this, we propose SQuad, a Sub-Quadratic Attention Distillation framework that achieves a complexity of $\mathcal{O}(n\sqrt{n})$ in the resulting distilled Attention, naturally balancing the efficiency v/s expressivity trade-off. Instead of training our own Video DiT from scratch, which is prohibitively expensive, we fit a pretrained full softmax Self-Attention DiT into our proposed SQuad-Attention one by distilling the former in two stages: Flow-Matching Supervised Fine-Tuning (SFT), followed by improved Distribution Matching Distillation (DMD2) which additionally makes the sampling more efficient. On the Wan~2.2 5B text-to-video model, SQuAD matches the quadratic teacher on VBench ($83.20$ v/s $83.08$) while cutting the per-step per-block attention FLOPs by $\sim$$67\times$ and attention latency by $\sim$$11\times$, and end-to-end DiT latency by 2$\times$, all while also generating a video in only $6$ Neural Functional Evaluations (NFEs) instead of the default $100$. --- ### Physics of Agents: Statistical Mechanics Predicts Collective Behavior of AI Agents Authors: Batu El, Jinhee Paeng, Fatih Dinc, Shiye Su, Mete Erdogan, Aneesh Pappu, Haotian Ye, Wanjia Zhao, Surya Ganguli, James Zou Published: 2026-08-17 Categories: cs.AI, cs.MA, cs.SI arXiv: https://arxiv.org/abs/2608.16578v2 AI agents increasingly operate as part of interacting systems rather than in isolation. As agents exchange information and jointly make decisions, their interactions can improve collective reasoning but may also produce herding, polarization, or amplify shared biases. Understanding and predicting these collective dynamics is therefore important for designing effective and aligned multi-agent systems. Here, we study over 10,000 communities of language-model agents that repeatedly exchange messages and revise their opinions across objective mathematics questions and subjective political statements. Despite substantial diversity in possible behavior, the individual and group dynamics can be represented by three characteristic regimes: indifference, polarization, and consensus. AI agents start indifferent and build conviction as they interact. On objective questions, communication improves collective accuracy, while on subjective questions it often drifts group opinions toward the right in the political spectrum. We explain these observations with a statistical-mechanics formalism in which agents stochastically favor lower social pressure. Given only initial opinions, our model predicts individual trajectories, outperforms all standard baselines, generalizes to unseen community graphs, and reproduces the observed group archetype distributions. Our fitted model parameters reveal the mechanics underlying our key observations: i) communities operate below the critical social temperature, which explains conviction buildup; ii) attractive ties outweigh repulsive ones, which favors consensus; and iii) agents holding the correct answer exert the strongest pull, which drives truth-seeking. Overall, our results demonstrate that collective behavior of AI agents, like that of other complex systems, follows compact and predictive dynamical laws. --- ### Physics of Agents: Statistical Mechanics Predicts Collective Behavior of AI Agents Authors: Batu El, Jinhee Paeng, Fatih Dinc, Shiye Su, Mete Erdogan, Aneesh Pappu, Haotian Ye, Wanjia Zhao, Surya Ganguli, James Zou Published: 2026-08-17 Categories: cs.AI, cs.MA, cs.SI arXiv: https://arxiv.org/abs/2608.16578v1 AI agents increasingly operate as part of interacting systems rather than in isolation. As agents exchange information and jointly make decisions, their interactions can improve collective reasoning but may also produce herding, polarization, or amplify shared biases. Understanding and predicting these collective dynamics is therefore important for designing effective and aligned multi-agent systems. Here, we study over 10,000 communities of language-model agents that repeatedly exchange messages and revise their opinions across objective mathematics questions and subjective political statements. Despite substantial diversity in possible behavior, the individual and group dynamics can be represented by three characteristic regimes: indifference, polarization, and consensus. AI agents start indifferent and build conviction as they interact. On objective questions, communication improves collective accuracy, while on subjective questions it often drifts group opinions toward the right in the political spectrum. We explain these observations with a statistical-mechanics formalism in which agents stochastically favor lower social pressure. Given only initial opinions, our model predicts individual trajectories, outperforms all standard baselines, generalizes to unseen community graphs, and reproduces the observed group archetype distributions. Our fitted model parameters reveal the mechanics underlying our key observations: i) communities operate below the critical social temperature, which explains conviction buildup; ii) attractive ties outweigh repulsive ones, which favors consensus; and iii) agents holding the correct answer exert the strongest pull, which drives truth-seeking. Overall, our results demonstrate that collective behavior of AI agents, like that of other complex systems, follows compact and predictive dynamical laws. --- ### BabelSteering: Multilingual Safety Alignment via English Steering Vectors Authors: Emma V. Stein, Dominik Meier, Terry Ruas, Jan Philip Wahle, Bela Gipp Published: 2026-08-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.16577v1 Large language models (LLMs) are deployed globally in high-stakes settings, yet most safety research and alignment efforts remain concentrated on English. Thus, users interacting with LLMs in other languages may encounter weaker safeguards despite relying on the same systems for similarly sensitive tasks. In this work, we investigate whether safety signals learned from a high-resource language, like English, can improve multilingual safety. We propose BabelSteering, an activation steering method that acts as a lightweight inference- time intervention, using refusal directions derived from English safety supervision to generalize across languages. Our evaluation includes eight languages and jointly measures refusal of harmful requests, over-refusal, and general task utility. The results show that BabelSteering increases the refusal of harmful requests across languages, with only a marginal to no reduction in task utility but with some increase in refusal of pseudo-harmful prompts. For example, for Gemma 7B, we see an average increase in the refusal of harmful prompts across languages of 11 percentage points (pp), with individual languages like Bengali seeing an increase of 17 pp, with no loss of utility on Global MMLU, while pseudo-harmful refusals increase by 13 pp on average. We also introduce a multilingual translation-and-evaluation pipeline to facilitate future work on cross-lingual safety interventions. Overall, our findings suggest that activation steering may provide a practical, low- cost mechanism for extending English-derived safety signals to other languages. Warning: this paper contains examples with unsafe content --- ### Learning Generalizable Reconstruction of High-Dimensional Neural Dynamics Authors: Anima Kujur, Zahra Monfared Published: 2026-08-17 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.16569v2 Accurate reconstruction of long-duration neural recordings is challenging because local field potentials (LFPs) are high-resolution, multichannel, transient, and variable across subjects. We present PCA-DMD, a scalable operator-theoretic framework that segments LFP recordings into overlapping windows, projects them into a compact PCA space, learns linear Koopman evolution in the latent space, and reconstructs continuous signals through inverse projection and overlap-add aggregation. On 200,000-sample hippocampal recordings, PCA-DMD outperformed Classical DMD, SpDMD, MrDMD, and HODMD, achieving KLD=0.0761 and HD=0.0847. In all-pair cross-subject zero-shot generalization at 300,000 samples, correlations were 0.9504-0.9800, with HD=0.0010-0.0072 and KLD=0.0005-0.0022, without target-subject fine-tuning. The prediction showed close one-step agreement on temporally held-out LFP segments across the unseen interval and multiple channels. Scalability analysis from 400,000 to 900,000 samples showed stable zero-shot reconstruction, with mean correlation remaining about 0.965-0.968 while computational cost increased predictably. External validation on an independent 93-channel Allen Neuropixels recording yielded mean and median channel-wise correlations of 0.7427 and 0.7990, respectively. Koopman spectral and mode analyses revealed dominant eigenvalues concentrated near the unit circle. PCA-DMD therefore provides an interpretable, generalizable, and computationally scalable framework for reconstructing high-dimensional neural dynamics. --- ### Probabilistic Circuits as Reasoning Machines in Artificial Intelligence (Part I) Authors: Robert Peharz Published: 2026-08-17 Categories: cs.AI, cs.LG, math.PR arXiv: https://arxiv.org/abs/2608.16565v1 This cumulative habilitation thesis studies probabilistic circuits (PCs) as a powerful and tractable framework for reasoning and learning under uncertainty in artificial intelligence (AI). It first advocates for probability as a core language for AI, emphasizing its connections to logic and information theory; the conceptual simplicity of probabilistic reasoning---based primarily on the sum and product rules; the parallels between probabilistic inference and human cognition; and the role of probability in optimal decision making. However, probability also faces significant computational challenges, as probabilistic inference is NP-hard in almost all probabilistic models. PCs address these challenges through structural constraints that ensure exact computation of a wide range of inference queries in polynomial time, such as marginals, conditionals, most probable explanations, expectations, and more advanced inference tasks. This thesis synthesizes a decade of research across foundations, algorithmic developments, and empirical validation of PCs. Key contributions highlighted in this work are foundational theory of PCs, Bayesian approaches for learning PCs, scalable implementations and integration with deep learning, hybrid models that combine PCs with intractable models, and connections with symbolic machine learning paradigms. This is the first part of my Habilitation Thesis. The second part is omitted, as it comprises the cumulative part of the thesis and has been published at various venues (see Chapter 5). --- ### CUBICS: Situation-aware performance estimation for safety-relevant ML components Authors: Benjamin Herd, Jessica Kelly, Mario Trapp Published: 2026-08-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.16564v1 Machine learning (ML) is a key technology driving innovation today, but ensuring ML safety remains a major challenge for safety-related applications. A promising idea is to build proven-in-use arguments from field data, e.g. by running ML components (MLCs) in shadow mode or within safety envelopes so that their outputs can be monitored as 'safe probes' without affecting safety. These probes can then be used to build a statistical argument about field performance in a Bayesian way. However, many Bayesian field-data approaches in safety engineering model failures as a simple Bernoulli (or binomial) process with a single global failure probability and i.i.d. trials, which is rarely adequate for MLCs whose performance depends strongly on context. Statistical evidence is also about coverage of relevant situations, including edge cases, and building a single integrated statistical model for the entire system is usually not feasible. To address these challenges, this paper introduces CUBICS, a context-modular framework for per-component, situation-aware performance estimation of safety-relevant ML components. CUBICS partitions the operational design domain into situations and, for each safety-relevant component, defines a set of situation-specific assumptions and probabilistic guarantees that are represented and updated in a Bayesian manner using Subjective Logic (SL). By combining these guarantees with beliefs about how often each situation occurs, CUBICS derives an overall risk estimate for each component without requiring a monolithic system-level statistical model, and thus provides a building block for modular, field-data based safety assurance. --- ### DeepInsight II: One Trace from Benchmark to Robot Authors: Siyi Li, Yuchen Kang, Wuliang Wang, Zhengjie Zhang, Jiangpin Liu, Jianhao Yao, Jie Chen Published: 2026-08-17 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.16556v1 Across a Physical AI stack, evaluation maturity is inversely aligned with deployment risk: foundation models enjoy mature, standardized harnesses, while the embodied layers on which deployment actually turns remain fragmented across benchmark-specific simulators, embodiments, and interfaces. The first DeepInsight report (v1) unified evaluation across this stack behind three abstractions---task, resource, and result---but its quantitative evidence centered on the foundation-model layer; navigation and manipulation (System 1) and whole-body control (System 0) remained simulation case studies, and physical execution was outside its empirical scope. DeepInsight II keeps that substrate fixed and quantifies the embodied half. First, it reproduces released-checkpoint references across two navigation and four manipulation benchmarks under their native protocols. Second, MotionBench places four released whole-body controllers under one workload and metric contract, then carries a qualified within-family cohort from parallel simulation to matched real-robot trials in which simulated and physical rollouts share a parent trace identity while retaining execution-domain-specific records, making the sim-to-real gap a native reduction rather than a reconciliation across toolchains. Third, a composed System 2--1--0 study extends trace localization into five evidence-grounded handoff labels, each mapped to a concrete repair action, with a measured repairability criterion and physical episodes testing the same attribution under hardware-observable state. The contribution is therefore not a new evaluation architecture, but empirical continuity from benchmark execution to matched robot evidence and repair-oriented diagnosis. --- ### Ask, Condition or Abstain: Reinforcement Learning for Missing-Premise Reasoning Authors: Yongqi Tong, Zhenyu Zhang, Zimi Liu, Kewei Fu, Mingli Song, Haofei Zhang, Junshao Zhang, Hong Zhu, Jiang-Ming Yang, Xin Zhang, Jianshe Li Published: 2026-08-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.16554v2 Answer-only reinforcement learning (RL) trains reasoning models to solve fully specified problems, but many realistic queries omit a premise needed for a unique answer. In this setting, the useful response is not always refusal: the model should ask for the missing premise, condition its answer on the unknown quantity, or abstain when no informative conditional response is available. We present \emph{Ask-Condition-Abstain Reinforcement Learning} (ACA-RL), a data-augmented RL framework for this setting. Its reasoning-graph-guided pipeline converts well-posed problems into missing-premise training instances with localized gap annotations; ACA-RL then trains on these instances with a structured reward over five observable response behaviors. We also introduce the \emph{Missing-Premise Benchmark} (MPB), a 274-instance human-verified benchmark spanning mathematical, logical, and real-world word problems. Across Qwen3 and Llama models, ACA-RL consistently improves on MPB while preserving competitive performance on well-posed reasoning tasks. Together with the released code, MPB, and training data, this work supports a new mission for NLP evaluation: measuring whether models can recognize when a task is underdetermined and handle uncertainty, not only whether they can answer fully specified questions. --- ### STAGE: Controlled Objective Admission for Multi-Preference LLM Alignment Authors: Yongqi Tong, Zhenyu Zhang, Ruirui Wang, Kewei Fu, Shaoqing Lin, Sijie Dong, Jiang-Ming Yang, Xin Zhang, Jianshe Li Published: 2026-08-17 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.16553v1 Multi-preference alignment is often framed as scalarization: combine reward dimensions, then optimize. This leaves a temporal decision underspecified: when should each preference dimension enter policy optimization? We propose \methodname, a stability-guided active-set controller for controlled objective admission. \methodname starts from a small active set, retains admitted objectives, and expands when reward-deviation gates indicate low recent deviation or a patience budget is exhausted. A probing phase estimates a hard-to-easy order, and adaptive weighting emphasizes underperforming active dimensions. Automatic evaluations with 15 training preferences and 16 held-out benchmark columns show that \methodname obtains higher averages than simultaneous scalarization and shared-budget adapted baselines. Component ablations and expansion dynamics further support cumulative retention, gated admission, and probing-derived ordering as useful design choices in this setting. These results position objective-entry timing as a concrete control variable in reward-vector RLHF. --- ### Supervising the Path to Fine Scales: GalerkinFlow for Scientific-Field and Image Super-Resolution Authors: Zikang Zhan Published: 2026-08-17 Categories: cs.CV, cs.CE, cs.LG arXiv: https://arxiv.org/abs/2608.16546v1 Most super-resolution models learn from paired data by supervising only the final high-resolution output. This provides little control over how the prediction should evolve between the downsampled observation and its fine target. We introduce GalerkinFlow, an equation-agnostic framework that turns each coarse--fine pair into supervision along an entire reconstruction path. At a random sample of intermediate states on the reconstruction path, the model predicts the coarse-to-fine residual velocity and uses coarse-anchor point to define a pseudo-endpoint. We show that the reconstruction loss of this pseudo-endpoint is exactly related to the intermediate velocity loss through a known time-dependent weight. Consequently, every intermediate state contributes supervision toward the same fine target, rather than serving only as an internal step toward an endpoint loss. Because intermediate states already reveal part of the missing fine-scale structure, we additionally supervise the coarse endpoint used during one-step inference. A finite-difference objective further constrains local spatial variation. GalerkinFlow combines convolutional features with scale-conditioned Galerkin operator mixing and requires no governing equation or physical metadata. It achieves the lowest raw-space errors among the evaluated equation-agnostic baselines on Navier--Stokes and Darcy Flow, while remaining competitive on DIV2K. ---