# Research Papers — 2026-10 · bundle page 131 of 156 > Full-text markdown of papers #6501–6550 of 7799 published in 2026-10, newest first. AI-readable mirrors of the full 7799-paper 2026-10 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-10/page/131 [← Previous bundle](https://aipod.dev/papers/archive/2026-10/page/130.md) [Next bundle →](https://aipod.dev/papers/archive/2026-10/page/132.md) ### Poisson-GENERIC Neural Operators: Exact Metriplectic Structure in Function Space via Casimir Entropies Authors: Jason Sulskis, Sathya Ravi Published: 2026-10-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.05570v2 Existing thermodynamically consistent neural operators enforce GENERIC degeneracy by projecting the reversible operator, introducing state dependence without guaranteeing Jacobi. We instead use a compatible Lie--Poisson pencil $L=αD+λ(uD+Du)$ for nonlinear transport and constant Poisson multipliers otherwise. Augmenting the observed field $u$ with latent entropy density $s$ makes $S=\int s$ a Casimir, so $L\,δS/δz=0$ without projection. Energy combines a fixed mechanical quadratic, learned gauge-free potential, and convex internal energy. The friction operator $M=AA^\top$ enforces $M\,δE/δz=0$ pointwise and supports diffusion and damping without an independent reversible transport generator. Skewness, positivity, and both degeneracies hold exactly; the continuum pencil satisfies Jacobi, verified numerically on the alias-free band. The parameterization represents the closed-form heat and damped-wave friction operators within its admissible range. Physical energy is bounded under coercivity assumptions, while a discrete-gradient integrator preserves both laws when solved to convergence. Across four PDEs in 1D and 2D and three backbones (FNO, Transolver, CNO), the model wins 61/72 seed-level comparisons against matched unconstrained baselines, recovers exact transport and wave symbols, matches heat and Burgers dissipation within 13%, and exhibits negligible advection dissipation. A constant-$L$ ablation exposes the cost of nonlinear transport, while a learned-energy ablation shows that conserving learned energy need not constrain physical energy. --- ### Factoriax: A GPU-Accelerated Factorio-Style Simulator for Reinforcement Learning Authors: Mickey Beurskens, Tristan Tomilin, Thiago D. Simão Published: 2026-10-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.05569v1 We introduce Factoriax, a GPU-accelerated factory-building simulator written in JAX. In Factoriax, an agent must collect resources, build machines using those resources, and then automate the collection and crafting process by arranging machines on the map to build production pipelines. This paper discusses the structure of the Factoriax simulator and an initial benchmark called Easy Rocket in which an agent is tasked with building a resource-intensive machine called the Rocket in a limited number of game ticks to escape the planet. We also publish results from a number of PPO-based training runs on Easy Rocket. Factoriax is built to be fast. A 1-billion-step PPO training run, equivalent to 500,000 episodes, runs on Easy Rocket in about 8 minutes on a single NVIDIA A100. A standard laptop GPU can complete the same run in about 84 minutes. Our trained PPO agent learns to gather resources, craft machines from those resources, and place them on the map through a curriculum reward directly tied to a manually designed set of achievements. After training, the agent does not place machines in a functional spatial configuration, failing to fully complete the benchmark, and leaving the challenge open for future attempts. --- ### LifeLong Digital Twin: A Unified Modeling Paradigm and Agent Harness for Event-Driven Lifelong Health State Trajectories Authors: Jin Jiang, Sean Yates, Jasper Chong, Raymond Brooks, Alex Lawson, Yuqin Qin, Liangcai Gao Published: 2026-10-04 Categories: cs.AI, q-bio.QM arXiv: https://arxiv.org/abs/2610.05566v1 Human health is a continuous, dynamic trajectory shaped by the cumulative interplay of biological processes, clinical events, behaviors and environmental exposures across the life course. Unifying the full breadth of lifelong health information, including longitudinal records, genetic variation, molecular profiles and environmental histories, is essential for whole-person modeling and remains a major challenge. We introduce LifeLong Digital Twin, a unified, event-driven modeling paradigm that organizes Life Events into daily Health States and accumulates them into Lifelong Health Context. An accompanying Agent Harness incorporates multimodal evidence beyond the language model's textual context. We evaluate four language models across 25 disease endpoints on three tasks: Disease Trajectory Forecasting, Disease Risk Ranking and Multi-horizon Disease Prediction. The approach yields marked gains over the reference condition: model-averaged F1 increases by 22.0% for disease identification in trajectory forecasting and 18.3% for five-year disease outcomes; thyroid-disease F1 reaches 0.669. The framework provides a foundation for whole-person digital twins and research on personalized lifelong disease prevention. --- ### Lend Me Your Eyes: Instruction-Aware Text Embeddings via Attention Relay Authors: Yiyuan Luo, Vaggos Chatziafratis Published: 2026-10-04 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2610.05564v1 Text embedding models trained with contrastive learning learn to follow task instructions from instruction-paired data, while instruction-tuned LLMs already know how to follow them. We show that this instruction-following ability can carry over from an LLM to a Transformer-based embedder without any training. We propose Attention Relay, which passes the attention weights an LLM produces to the embedder's own attention. Across six instruction-tuned LLMs from the Qwen3, Llama 3.1 and OLMo 3 families and ten widely used embedding models that differ in tokenizer, size and pooling type, Attention Relay makes nearly every combination instruction-aware. Experiments that break the method down into its parts show that the LLM's attention weights track the instruction in its later layers and come largely from instruction tuning. They also show that relaying these weights selects which content in the text matters: it makes the aspect of the text that the instruction asks about dominant in the embedding, or restores that aspect where averaging had diluted it. --- ### G-CARB: Graph-Localized Conformal Agent Risk Budget for Compositional Harm Authors: Zijun Yu, Yu Gu, Vahid Partovi Nia, Masoud Asgharian Published: 2026-10-04 Categories: stat.ML, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.05563v1 Small language model (SLM) agents need safety controls that track consequences across tool calls with little monitoring overhead. A private read, for example, becomes a leak when a later action sends that data outside the system. We introduce CARB (Conformal Agent Risk Budget), which calibrates when to stop an agent using a ledger of harm incurred before stopping. Under exchangeable episodes, standard conformal risk control bounds this declared loss in expectation over calibration and a future episode. G-CARB selects scorer evidence along observable dependencies from private sources to outgoing actions. The ledger still covers the entire executed history, and computing the gate score requires no additional language-model inference. On AgentDojo replay with two 14B backbones, G-CARB roughly halves scorer-input records at intermediate risk budgets while improving autonomous task completion relative to full-prefix scoring; random context of the same size achieves similar gains. Controlled examples show how retaining the relevant dependency can further avoid stopping benign work. --- ### Cut Binary Cross Entropy: Efficient Large-Vocabulary Loss and Gradient Kernels for Sequential Recommendation Authors: Yaoyiran Li, Haowen Ning, Mohamed Hammad Published: 2026-10-04 Categories: cs.LG, cs.AI, cs.AR, cs.IR arXiv: https://arxiv.org/abs/2610.05559v1 Industrial sequential recommender systems operate over massive item catalogs (e.g., 10^5--10^7 items). Multi-label recommendation models are trained with Binary Cross-Entropy (BCE) loss over the full vocabulary, but standard BCE materializes a dense (B, N, V) logits tensor in High Bandwidth Memory (HBM), incurring prohibitive $O(BNV)$ memory and fatal Out-Of-Memory (OOM) errors. While chunked loss optimizations exist for Softmax Cross-Entropy in LLMs, large-scale multi-label BCE optimization remains unexplored across deep learning ecosystems. We propose CutBCE, an exact, hardware-accelerated BCE loss and gradient operator implemented in JAX and Pallas for large-vocabulary workloads. CutBCE introduces (1) an exact fused reformulation evaluating dense background loss and sparse target corrections; (2) a custom Vector-Jacobian Product (VJP) with a dedicated Pallas TPU backward kernel computing logit tiles on-chip in both passes so logits and their gradients never reside in HBM; (3) dynamic VMEM budgeting and sharding-aware collective hoisting for distributed meshes; and (4) count-based zero-overhead training metrics. On single-chip TPU v5e/v6e mini-benchmarks, CutBCE eliminates OOM errors with up to 91.9% speedup. On 8-chip TPU slice training for multi-label SASRec with 876k items (Yambda-50M), CutBCE reduces peak HBM by 65.7% (>14 GiB saved per chip) and increases training speed by 225.9% with comparable accuracy. CutBCE is open-sourced at https://github.com/AI-Hypercomputer/RecML/blob/main/recml/core/ops/binary_cross_entropy_ops.py. --- ### An equality condition for the Dobrushin bound on attention rollout and how often it holds in trained transformers Authors: Przemysław Rola Published: 2026-10-04 Categories: cs.LG, math.PR arXiv: https://arxiv.org/abs/2610.05558v1 The Dobrushin coefficient of each attention-rollout factor satisfies $κ(\frac12(I+A))\le\frac12(1+κ(A))$, and multiplying these inequalities over layers bounds the coefficient of the whole rollout. We characterise exactly when the layerwise bound is tight: equality holds if and only if some token pair attaining $κ(A)$ is mutually self-dominant - each of the two attends to itself at least as strongly as the other attends to it. The condition is far from automatic: uniformly random stochastic matrices satisfy it only 24-30% of the time. When tested on the head-averaged attention of each individual input and restricted to content tokens - image patches, words or tabular features, excluding cls, register and separator tokens - the condition holds for essentially every input at every layer of DINOv2 (three model sizes), RoBERTa and DistilBERT. In the supervised models DeiT-B and ViT-B/16 it holds for 91% and 64% of input-layer pairs respectively, with all failures occurring late in depth. In FT-Transformer trained on two standard tabular benchmarks it holds for only 11-44% of input-layer pairs. The special tokens account for almost all failures in DINOv2 and the language models: when they are included, the condition holds for only 82-97% of input-layer pairs. --- ### Monocular markerless biomechanics for clinically interpretable gait assessment in spinal cord injury Authors: Shreyasvi Natraj, Mathieu Ruepp, Yanke Li, Robert Riener, Inge Eriks-Hoogland, Diego Paez-Granados Published: 2026-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.05552v2 Three-dimensional gait analysis guides rehabilitation after spinal cord injury but depends on marker-based motion capture and force plates, which few clinics have. Monocular markerless pipelines have been established in fewer healthy adult cohorts but not in neurological cohorts. We present the SCAI SCI Gait dataset, comprising 239 adult individuals with spinal cord injury with synchronized video, motion capture, and force-plate measurements, we fitted a parametric body mesh to a single sagittal-view video, driving an anthropometrically scaled OpenSim model via virtual markers. Markerless lower-body kinematics showed state-of-the-art agreement with motion-capture measurements (r = 0.68-0.90, p < 0.001, and RMSE = 4.18-6.49 degrees), and accurate kinematics-based predicted ground-reaction forces closely matched those measured by force plates (r = 0.85-0.87, p < 0.001, and RMSE = 2.13-2.19 Newton per kg). Furthermore, conditional-dependence graph analysis with Markov blankets revealed that waveform components were conditionally associated with functional independence, and speed-stratified clustering revealed distinct mechanical strategies among individuals walking at similar speeds. These findings establish the use of monocular video as a scalable approach for clinically meaningful biomechanical assessment and data-driven phenotyping in patients with spinal cord injury. Github: https://github.com/SCAI-Lab/SCAI-SCI-Gait-Dataset --- ### When Low Prediction Error Misleads Planning: Diagnosing Representation, Dynamics, and Decision Failures in Latent World Models Authors: Rui Min, Xianyao Li, Fang Xu, Jing Du Published: 2026-10-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.05550v1 The component that dominates a latent world model's prediction error need not be the one whose repair most improves action selection. We show this by comparing action sequences from identical physical starts and separating endpoint error into a candidate-pool center and action-relative responses. Across four model families and four tasks, a confirmation pool of 256 new starts per task and 300 shared candidates per start shows that center error dominates MSE in 14/16 model-task cells. Yet in six of these cells, an oracle that corrects only the action-relative responses yields better physical rank correlation and top-30 elite quality than one that corrects only the center, while leaving more latent MSE (family-wise corrected intervals). The preference differs across the evaluated settings: a separate LeWorldModel (LeWM) study that executes oracle-selected actions favors center repair on PushT and on Reacher with a render-matched goal. Matched-candidate tests localize ordering loss: for LeWM, encoding realized endpoints raises physical Spearman from 0.464 to 0.975 on that Reacher setting and from 0.193 to 0.631 on PushT (64 starts per task), while Cube's encoded-goal cost remains uninformative. A 72-run objective study improves selected response diagnostics, while incremental closed-loop planning gains remain unconfirmed. These results separate error magnitude from the decision effects of oracle correction and motivate evaluating representation, prediction, and planning as separate stages. --- ### Joint Estimation of Common-Slope Decay Rates and Spatial Amplitudes Using Parameterized Nonnegative Matrix Factorization Authors: Jeremy B. Bai, Filip Elvander, Sebastian J. Schlecht Published: 2026-10-04 Categories: cs.LG, eess.AS arXiv: https://arxiv.org/abs/2610.05549v1 We formulate joint estimation of common-slope decay rates and amplitudes from room impulse responses (RIRs) as parameterized nonnegative matrix factorization with the Itakura--Saito divergence as the loss function (IS-NMF). Estimation at each short-time Fourier transform frequency bin produces detailed reverberation time (RT) curves directly from RIR powers with no backward integration needed. Standard space-alternating generalized expectation-maximization (SAGE) algorithm yields closed-form amplitude updates and a convex subproblem for each decay rate update. To accelerate estimation, we introduce contribution-weighted SAGE, which emphasizes observations where each component contributes strongly to the modeled power. Experiments with synthetic data show accurate recovery of well-separated decays and faster loss reduction than standard SAGE. Application to measured coupled-room RIRs yields frequency-dependent RT curves and reveals complementary space-time contributions of the shared decay components. --- ### More Claims, Less Evidence: Bounded Verification of AI-Generated Digital Knowledge Artifacts Authors: Feliks Bańka, Jarosław A. Chudziak Published: 2026-10-04 Categories: cs.AI, cs.DL arXiv: https://arxiv.org/abs/2610.05547v1 Digital libraries, repositories, and AI-mediated knowledge services increasingly rely on generative systems to produce summaries, descriptions, and other multi-claim knowledge objects. Yet generation can scale far more easily than verification capacity: a reviewer may need to decide whether an object is suitable for publication or downstream use after checking only a small fraction of its claims. This creates a fundamental gap between claim-level verification and confidence in the object as a whole. The central question is therefore what a successful partial check implies about the reliability of the complete artifact when its size grows but the verification budget does not. This paper contributes a Bayesian model of bounded verification centered on the Predictive Value of Pass (PVP). The model predicts that evidentiary value decreases as artifacts grow under fixed verification capacity, improves with larger verification budgets, and is especially fragile when errors are sparse. Controlled experiments on FEVEROUS and FEVER support these predictions and show that adding supported claims around a fixed number of false or unsupported claims can make passing more likely while making a pass less informative. The model further yields the minimum verification budget required to maintain a target PVP, providing a practical component for AI-assisted quality-assurance workflows in which generated knowledge objects must be checked before publication or downstream use. --- ### Soft Strategy Selection for Batch-Mode Active Learning Authors: Rushil Gupta, Romain Lopez Published: 2026-10-04 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.05544v1 Real-world deployment of active learning typically forces practitioners to choose an acquisition strategy before any data is labeled. This is a daunting task: strategy performance varies widely across settings (e.g. datasets, surrogate models) and cannot be assessed without deployment. Existing strategy selection methods explore one strategy from a portfolio at each round and identify the optimal one using bandit feedback or model retraining. Many acquisition rounds are therefore spent exploring strategies rather than collecting the most informative data. Such overhead is a major barrier to AL-driven design of high-throughput experiments, such as genetic perturbation screens and directed evolution, where AL runs consist of only a few rounds with large batch sizes. This regime permits a natural alternative: acquiring data using multiple AL strategies within a single batch. We refer to this as soft strategy selection and introduce FractAL, a method specifically designed for this task. FractAL infers a per-strategy reward using influence-function-based data attribution, which requires no additional retraining, and then computes budget shares for each strategy in the portfolio using online mirror descent. We benchmark FractAL across 7 setups spanning classification, regression, and genetic perturbation effect prediction. The results highlight that strategy selection is a hard problem: every existing method performs worse than random sampling on at least one setup. FractAL, however, matches or outperforms every baseline, including random sampling, on all 7 setups. Its allocations concentrate budget on the strongest strategies in the portfolio while pruning the weakest. FractAL is therefore a reliable choice for real-world deployments, where the optimal strategy is unknown in advance, an important step towards making AL practical for high-throughput experiments and modern scientific discovery. --- ### Don't Judge an LLM Only by Its Activations: Discovering Suppressed Safety Features via Counterfactual Activation Potential Authors: Swadesh Swain, Sanghamitra Dutta Published: 2026-10-04 Categories: cs.LG, cs.AI, cs.CL, cs.CR arXiv: https://arxiv.org/abs/2610.05541v1 Mechanistic interpretability has emerged as the primary means to understand safety behavior of LLMs. However, existing tools primarily focus on the activating neurons or features of a model. The role of the remaining large set of inactive components is invisible to such methods. This work demonstrates that the inactive set contains safety-critical features that are causally relevant for refusal of harmful prompts. Suppressing such features could turn refusals into compliance, while passing undetected by prevalent interpretability tools. We introduce the Counterfactual Activation Potential (CAP), a metric that quantifies a suppressed feature's latent activation tendency as the product of its encoder alignment (how strongly the input drives it), suppression strength (how strongly active features inhibit it), and safety criticality (how much refusal depends on it). To find suppressed safety features at scale, we propose CAP-guided Safety Feature Discovery (CSFD), a two-stage filtering algorithm that identifies candidate safety features from hundreds of thousands of transcoder features without exhaustive ablation. A significant fraction of trials turn compliant with harmful prompts when a candidate feature is ablated. Under natural jailbreaks, the suppression acting on the highest-CAP features rises 2-4x, and their activation correspondingly falls by up to 80%. Amplifying a feature's suppressors pushes its activation down and raises harmful compliance with prompts related to the suppressed feature, with no such effect for random features. Our experiments span five Gemma, Qwen, and Llama models across various parameter sizes. Our findings indicate that jailbreaks could operate in part by suppressing safety-critical features rather than solely activating harmful ones, and that suppressed features are a necessary complement to activation-focused interpretability of safety behavior. --- ### SALUS: Automated Auditing of NL-to-SQL Benchmarks through Weak Supervision of Multi-Agent Output Authors: Shiyuan Zhou, Ashwin Gerard Colaco, Sainyam Galhotra, Sharad Mehrotra Published: 2026-10-04 Categories: cs.DB, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2610.05540v1 Natural language to SQL (NL-to-SQL) benchmarks are foundational to progress in data analysis research, yet recent work has shown that widely-used benchmarks contain significant annotation errors. These errors silently corrupt evaluation metrics, penalize correct model output, and distort the field's understanding of state-of-the-art performance. We present SALUS, a system that automatically detects annotation errors in NL-to-SQL benchmarks. SALUS frames benchmark auditing as a weakly supervised error detection: SQL generated by multiple LLM agents drive a suite of complementary weak-labeling functions. By passing this noisy vote matrix through a generative label model, we extract high-confidence training samples without requiring human ground truth. These samples train a decision plane that maps gold SQL query features to per-agent trustworthiness, allowing SALUS to intelligently fuse reliability estimates with raw verdicts for rigorous benchmark error detection. We evaluate on BIRD-Clean-xs, a benchmark of 298 BIRD development tasks with manually verified correctness labels. SALUS achieves F1 = 0.9194, significantly outperforming the state-of-the-art baselines. Applying SALUS to the full development sets, we estimate annotation error rates of approximately 37% on BIRD and 27% on Spider. --- ### LiFT: Loop Flow Transformers Authors: Mohammad Mahdi Derakhshani, Pedro M. P. Curvo, Gertjan J. Burghouts, Jan-Willem van de Meent, Cees G. M. Snoek Published: 2026-10-04 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.05538v1 We introduce Loop Flow Transformers (LiFT), a family of looped generative models that scales computation by repeatedly applying a shared Diffusion Transformer (DiT) core, with only light changes to the standard architecture. Rather than asking every recurrent step for the final prediction, LiFT trains each step with a single regression target: a point on a straight path from the model's initial estimate to the flow-matching target. Because we index these targets by a continuous depth coordinate, a trained model can loop far beyond its training depth with no retraining, early exits, or other modifications. In our experiments, these longer rollouts improve generation, so inference computation can grow without adding parameters. On ImageNet at 256x256, LiFT-L/2 achieves an FID 3.34 points lower than our dense DiT-XL/2 baseline while using approximately 60% fewer parameters, 32% fewer training FLOPs, and 52% fewer inference FLOPs. --- ### Dataset Signatures in Human-LLM Interactions and User Modeling Authors: Joseph Suh, Serina Chang Published: 2026-10-04 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.05534v1 Human--LLM interaction datasets shape our understanding of AI use and provide a foundation for downstream research, including training and evaluation of user models. In recent years, a growing number of datasets have sought to capture a representative picture of human--LLM interactions. But how different are the pictures these datasets provide, and what do those differences mean for research built on them? We study these questions across seven conversation datasets, spanning in-the-wild chat logs and human preference data. We begin by revisiting the dataset classification experiment of Torralba & Efros and find that neural network classifiers identify the source of a conversation from user messages alone well above chance, indicating distinctive dataset signatures. This separability persists after matching datasets on the dimensions of human-designed taxonomies, implying subtle differences that these taxonomies do not capture. We then examine the implications for user modeling: how dataset signatures propagate to the outputs of user models trained on these datasets; how dataset choice influences evaluations of user model quality and subsequent evaluations of LLM assistants paired with these user models; and how dataset classifiers can guide data selection for training user models. While each dataset is meant to capture a slice of 'real-world' interactions, our findings reveal the extent to which these slices diverge, and the consequences of those differences for research built on these foundations. --- ### What Does a Harness Repair? A Preregistered Study of Visibility, Baseline Adequacy and Evaluation Defects Authors: Bowen Xu, Boyu Chen Published: 2026-10-04 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.05533v1 Harness search keeps a change to the prompts, reasoning switches, token budgets or parsers around a frozen model if the change raises a score. Such a gain can come from answers the parser could not read before, a weak comparison, or a defect in the evaluation. We preregistered a study of where these gains come from, with three small models, three benchmarks, replication and test partitions, a GEPA search arm and six evaluation defects injected one at a time, and we report all 47 primary endpoints. Turning thinking off raised accuracy over a capped thinking setting in 5 of 9 model-benchmark cells, and in each the gain came mostly from questions where the capped setting gave no readable answer. The thinking-off setting was not meaningfully worse than a rescue configuration or four GEPA-selected harnesses in 11 of 13 comparisons, and lost to the rescue on GSM8K for two models. GEPA repaired its broken starting points, but none of its selected harnesses was more accurate than the thinking-off setting. A thinking budget in the serving engine, which also allows a longer answer, lowered truncation and raised the parse rate in 6 of 9 cells. In 6 of 15 evaluable defect-model pairs, replication through the same pipeline reproduced the defect's distortion instead of revealing it. On the LongevityBench multiple-choice tasks, only the longevity-tuned model beat the strongest constant-label baseline. --- ### DelegationBench: Measuring When AI Agents Should Ask Before Acting Authors: Shiva Pochampally Published: 2026-10-04 Categories: cs.AI, cs.CL, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2610.05532v1 AI agents that send emails, edit files, and make purchases must decide when to act on their own and when to check with the user first. This decision is usually evaluated by showing a model a proposed action, asking whether it should proceed, and scoring agreement with human labels. We introduce DelegationBench to test whether such scores can be trusted. It has 156 scenarios with four possible responses (act, ask for permission, ask for missing information, refuse), and most scenarios come in matched pairs that change a single feature: whether the action was requested, what is at stake, whether it can be undone, or who will see it. Across ten models from five families, agreement scores mislead in three ways. A simple keyword rule, which we wrote after seeing the benchmark, agrees with our annotators more often than eight of the models, yet its decision changes in only 9 of 48 matched pairs. Equivalent ways of asking the same question change how often a model acts by up to 52.5 percentage points. And every model stops to ask the user less often when it must carry out the task with tools than when it judges a proposed action. When rules are stated explicitly, the same models follow them almost perfectly, so the gaps are not explained by a general inability to follow rules. We release the benchmark and evaluation tools and recommend reporting these properties separately rather than as one score. --- ### Deep Prior Learning for Embodied Perception Authors: Yimou Wu, Jiaxin Guo, Yun-hui Liu, Zheng Li Published: 2026-10-04 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2610.05531v1 Embodied systems need geometric perception that exploits available observations beyond images alone. Recent feed-forward 3D models incorporate geometric priors, including camera poses, intrinsics, and depth. However, handling noisy poses, preserving accurate priors, and recovering physical scale require more than simply accepting these inputs. We introduce \emph{Vision-Prior Geometry Grounded Transformer} (VPGGT), a VGGT-based framework that extends OmniVGGT for prior-aware embodied perception. We formulate sensor-motivated pose corruptions from ground-truth trajectories for training and introduce a parameter-free \emph{prior residual connection} (PRC) to mitigate \emph{prior dilution}, where predictions are less accurate than their supplied pose priors. Our noise formulation targets camera poses; supplied intrinsics and depth receive no additional corruption. We further introduce \emph{Metric Global Attention}, which conditions a global scale token on available pose and depth scales and predicts a shared metric scaling factor for the geometric outputs. Experiments across four datasets show that \emph{PRC} improves translation-direction accuracy and joint pose AUC over a matched training baseline when camera priors are provided for all views, under both exact and corrupted poses. These results support explicit prior access during refinement as a useful addition to feature-level conditioning. --- ### DynaMesh: Dynamic 3D Texture Generation Authors: Raj Hansini, Guan Chen, Rana Hanocka, Itai Lang Published: 2026-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.05529v1 We present DynaMesh, a dynamic texture generation method for 3D meshes. Given a textureless shape and a text prompt describing an effect, our method produces an appearance that evolves while the object's geometry remains unchanged. Previous works on dynamic 3D content generation have focused on motion, where an object's geometry and position change while keeping its appearance the same. Methods on texture generation sit on the other side of the problem, painting appearance onto a shape as a fixed surface property and not as an evolving process. Neither addresses a visual effect that propagates on a 3D object. A natural route consists of two generators: a video model that shows the effect from a single view, and an image-to-3D generator that lifts each frame to 3D. However, the latter has no notion of time, so running it per video frame produces a sequence that flickers, loses effect details, and yields a different mesh at every video frame. Our method addresses these failures by conditioning a video model on a render of the mesh and the prompt to obtain a reference video, then running a frozen image-to-3D generator on the video with two changes. The conditioning of each frame is blended over a temporal window, and low-rank adapters are fit per shape to restore the lost details. The mesh is encoded once for the whole sequence, so geometry is constant by construction, and the output is a single mesh with a texture per frame. Applied to various objects and effects, DynaMesh substantially improves over recent video-to-4D and texturing methods, and can generalize its temporal effect to different shapes never seen during training. Our project page is at https://threedle.github.io/dynamesh/. --- ### Have I Scene This Before? Spatially Grounded Conversational Memory for Complex Queries in Egocentric Assistants Authors: Jiazhou Liang, Liam Gallagher, Kiko Chen, David Guo, Armin Toroghi, Yifan Simon Liu, Scott Sanner Published: 2026-10-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.05526v1 Egocentric assistants must connect what users say with what they see across long interaction histories. We formalize this challenge as Spatially grounded Conversational Reasoning (SpaCR): cross-scene, recall-oriented, and counterfactual spatial queries that combine user-stated facts with geometric evidence. Direct vision-language models incur high inference costs and context limits as histories grow, while keyframe selection and retrieval can omit objects or evidence needed for complete recall. We propose Spatially grounded Conversational Memory (SpaC-MEM), an object-centric working memory that uses 3D reconstruction and segmentation to ground conversational information in persistent physical objects. It compresses multimodal histories while preserving spatial evidence and allowing object-specific facts to be updated through dialogue. We also introduce Ego-SpaCR, a benchmark comprising 620 ScanNet video sessions augmented with 95 task-oriented conversations and 3,100 evaluation queries. SpaC-MEM achieves the highest overall answer accuracy among the evaluated methods and improves object recall while requiring substantially fewer reference input tokens than native-video baselines. Removing 3D spatial information substantially degrades performance, highlighting the importance of preserving spatial and conversational evidence together. --- ### Optimization Geometry of QAOA and Variational Quantum Algorithms Authors: Vojtěch Novák, Ivan Zelinka, Silvie Illésová, Swagatam Das, Martin Beseda Published: 2026-10-04 Categories: quant-ph, cs.NE arXiv: https://arxiv.org/abs/2610.05524v1 Variational quantum algorithms turn choices of Hamiltonian, ansatz, and parameterization into a classical nonconvex optimization problem. We study how this objective function can be visualized and characterized in ways that help explain optimizer behavior. We distinguish two properties of the objective: the number of local minima encountered along sampled directions and the differences in quality among local-search endpoints. We then ask how a local optimizer, represented by BFGS, compares with adaptive differential evolution, represented by jSO. Rather than comparing jSO with a single local run, we allow BFGS multiple starts within the same function-evaluation budget. This gives local search repeated opportunities to explore different basins and provides a stronger baseline for asking when a global evolutionary solver is useful. We study these questions using VQE and QAOA, focusing on how frustra- tion, circuit depth, parameter tying, mixed locality, and nonlinear repa- rameterization change the Hamiltonian expectation-value objective seen by the classical optimizer. Increasing independent circuit depth raises the sampled local-minimum count without making global search more effective. Parameter tying, by contrast, produces both more repeated local structure and much larger differences in quality among local-search endpoints, and in this regime adaptive differential evolution outperforms function-evaluation-matched multistart BFGS. The comparison shows that the number of local minima alone does not determine whether global search is advantageous: the important distinction is whether different basins lead to similarly good solutions or to substantially different objective values. These results provide a practical workflow for connecting model construction, objective- function geometry, empirical diagnostics, and optimizer choice. --- ### What Does Fréchet Distance Measure? A Directional Decomposition Authors: Yunghee Lee, Jaeyeon Kim Published: 2026-10-04 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.05518v1 The Fréchet distance is a de facto standard for evaluating generative models across domains, appearing as FID for images and FVD for videos. It summarizes the discrepancy between generated and reference distributions in a single scalar, with lower values typically interpreted as better generation quality. However, this scalar view can obscure what drives the comparison. For example, in COCO dataset, increasing the number of diffusion sampling steps improves ImageReward scores yet worsens (increases) FID. Motivated by this mismatch, we seek to make the Fréchet distance more interpretable by uncovering where the discrepancy lies. To this end, we introduce directional Fréchet distance, the expected squared projection of the optimal transport displacement onto a given direction. Across our image, video, and protein case studies, we find that a small number of interpretable directions account for much of the distance. We use these directions to explain the FID increase in terms of semantic concepts represented by CLIP embeddings, quantify FVD's bias toward per-frame appearance, and revisit the interpretation of Protein FID. We open-source our codebase at https://github.com/yhlee-add/directional-fd. --- ### Robust 2D Traversability Mapping for Construction AMRs via Failure-Mode-Aware Fusion of LiDAR Geometry and Monocular Semantics Authors: Manoj Karnekar, Om Mandhane, Gautham Ramkumar Published: 2026-10-04 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2610.05505v1 Autonomous Mobile Robots (AMRs) on active construction sites face severe navigational challenges: geometry-based traversability mapping (e.g., LiDAR) misses visually hazardous but geometrically flat surfaces like wet mud and ponding concrete, while abrupt geometry on drivable speed-breakers and inclines produces phantom obstacles. We propose a real-time, failure-mode-aware multimodal traversability pipeline on an NVIDIA Jetson AGX Orin, where LiDAR is the primary geometric safety estimate and monocular semantics act as a selective, class- and confidence-gated corrective signal. The representation retains distinct traversable classes, namely flat road, terrain, and rocky terrain, while flagging construction hazards. We also release a multimodal construction-site dataset from a custom AMR: four closed-loop ROS 2 sequences from two active sites (RGB, depth, LiDAR, IMU, GPS-RTK, odometry) plus 506 annotated frames across 28 semantic classes. By projecting LiDAR onto dense semantic masks, resolving sparsity via morphological in-painting, and applying failure-mode-aware fusion with Patchwork++, the system corrects complementary geometric failure modes for a local AMR costmap. --- ### The Functional Structure of Post-Compression Recovery in Low-Rank LLMs Authors: Zishan Shao, Liang Tian, Georgiy Zemlevskiy, Kangning Cui, Lixun Zhang, Yixiao Wang, Ting Jiang, Jinhee Kim, Yixuan Chen, Rui-Feng Wang, Fan Yang, Hai Li et al. Published: 2026-10-04 Categories: cs.AI, cs.IT arXiv: https://arxiv.org/abs/2610.05504v1 Different low-rank compression methods can produce compressed LLMs that respond differently to the same post-compression recovery procedure, and relative advantages observed between methods at the compression endpoint may shrink, grow, or even reverse after recovery. We ask whether this recovery heterogeneity reflects functional structure beyond scalar loss evolution, and how that structure evolves throughout recovery. Our results establish that this heterogeneity reflects a reproducible compression-induced functional structure, which we formalize as recovery pressure. To characterize this structure consistently throughout recovery, we develop a standardized functional characterization within each backbone that is applicable across heterogeneous low-rank methods. The primary backward characterization reveals reproducible module-wise structure across independent probes, while a complementary forward-only characterization recovers related structure without loss or backpropagation. We further find that recovery pressure measured at the endpoint is associated with subsequent recovery response; during recovery, its module-wise structure is reorganized non-uniformly, and localized updates induce distributed responses beyond directly updated modules. Further evidence indicates that tracking this evolving structure provides a complementary functional view of recovery progress alongside scalar loss. --- ### Lightweight Semantic EEG Foundation Model for Frozen Cross-Disorder Transfer Authors: Rita Huan-Ting Peng, Nhat Bui Published: 2026-10-04 Categories: cs.LG, eess.SP, q-bio.NC arXiv: https://arxiv.org/abs/2610.05503v1 Large-scale EEG foundation models have demonstrated promising transferability across neurological disorders, but often require millions of parameters and substantial computational resources. In this paper, we present the Universal Semantic EEG Foundation Model (USE-FM), a lightweight EEG foundation model that learns transferable neural representations through self-supervised signal reconstruction on the Temple University Hospital EEG Corpus (TUEG). After pretraining, the encoder is frozen and evaluated on two clinically distinct downstream tasks, abnormal EEG detection (TUAB) and epileptic seizure recognition (TUEP), using a unified frozen-transfer protocol against recent EEG foundation models, including LUNA-Base and CBraMod. With only 1.46 million parameters, approximately one-fifth the size of existing models, USE-FM achieves competitive overall performance, including strong sensitivity and F1-score on TUEP (SEN $75.00 \pm 14.14$, F1 $70.37 \pm 4.01$), while maintaining competitive performance on TUAB (AUC $85.24 \pm 5.61$). Beyond downstream classification, latent representation analysis using $k$-means clustering together with PCA and t-SNE demonstrates that USE-FM learns organized semantic EEG representations comparable to substantially larger foundation models. These results suggest that large-scale self-supervised pretraining enables lightweight architectures to learn transferable semantic EEG representations, providing a computationally efficient foundation for cross-disorder analysis and future clinical decision support in neurological disorders. --- ### Logic-Logit: A Logic-Based Approach to Choice Modeling Authors: Shuhan Zhang, Wendi Ren, Shuang Li Published: 2026-10-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.05501v1 In this study, we propose a novel rule-based interpretable choice model, Logic-Logit, designed to effectively learn and explain human choices. Choice models have been widely applied across various domains---such as commercial demand forecasting, recommendation systems, and consumer behavior analysis---typically categorized as parametric, nonparametric, or deep network-based. While recent innovations have favored neural network approaches for their computational power, these flexible models often involve large parameter sets and lack interpretability, limiting their effectiveness in contexts where transparency is essential. Previous empirical evidence shows that individuals usually use heuristic decision rules to form their consideration sets, from which they then choose. These rules are often represented as disjunctions of conjunctions (i.e., OR-of-ANDs). These rules-driven, consider-then-choose decision processes enable people to quickly screen numerous alternatives while reducing cognitive and search costs. Motivated by this insight, our approach leverages logic rules to elucidate human choices, providing a fresh perspective on preference modeling. We introduce a unique combination of column generation techniques and the Frank-Wolfe algorithm to facilitate efficient rule extraction for preference modeling---a process recognized as NP-hard. Our empirical evaluation, conducted on both synthetic datasets and real-world data from commercial and healthcare domains, demonstrates that Logic-Logit significantly outperforms baseline models in terms of interpretability and accuracy. --- ### LEON: Location Embeddings from OSM Neighborhoods via Hexagonal Graph Masked Autoencoders Authors: Szymon Soltysiak, Radoslaw Malek, Jedrzej Kusnierz, Milosz Chojecki, Piotr Szymanski, Aleksandra Kawala-Sterniuk Published: 2026-10-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.05497v1 Geographic information systems increasingly rely on sophisticated spatial representation learning techniques to extract meaningful patterns from complex geospatial data. This paper introduces LEON, a novel self-supervised framework that adapts Graph Masked Autoencoders (GraphMAE) for geospatial region representation learning. Our method leverages the inherent spatial structure of geographic data by constructing hexagonal grid graphs using H3 indexing and applying masked autoencoding techniques to learn robust spatial embeddings from OpenStreetMap (OSM) amenity distribution patterns. We evaluate LEON on multiple real-world datasets including EuroSAT satellite imagery classification and various geographic prediction tasks (housing prices, crime prediction, and urban analytics). Experimental results demonstrate that LEON achieves significant improvements in spatial understanding, with up to 1.87% accuracy improvement on EuroSAT classification and consistent performance gains across geographic prediction benchmarks. The learned embeddings exhibit highly structured and distinct properties, making them particularly suitable for downstream spatial analysis tasks. Our findings suggest that self-supervised learning provides an effective paradigm for geospatial region representation learning using widely available crowdsourced data. --- ### When Does Retrieval Help? A Study of In-Context Adaptation in Vision-Language-Action Models Authors: Zixuan Liu, Joris Köster, Zizhan Zheng, Siavash Khajavi Published: 2026-10-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.05492v1 Vision-language-action (VLA) models have shown strong potential as generalist robot policies, but adapting them to unseen tasks often requires costly parameter updates. Recent work such as RICL introduces in-context adaptability by retrieving expert demonstrations based on the current VLA observation and providing them as additional context at test time. The effectiveness of this adaptation therefore depends critically on the retrieval mechanism. In this work, we systematically study how different retrieval methods affect both retrieval quality and task performance within the RICL framework. Specifically, we compare four different methods: image-based retrieval, retrieval augmented with VLA's state, retrieval using features from the VLA backbone, and random retrieval. Our experiments yield three main findings. First, no retrieval method consistently dominates the others in task success, while surprisingly, random retrieval achieves a non-trivial success rate. Second, standard retrieval-quality diagnostics do not reliably reflect downstream VLA performance. Third, demonstrations from different but related tasks can provide useful transferable information. Together, these results provide an initial step toward understanding how retrieval mechanisms shape the in-context learning capability of VLA models and their downstream task performance, while highlighting the need for more careful design and evaluation of retrieval mechanisms for reliable test-time adaptation. --- ### Robust Surgical Robotic Instrument Tracking via Sequential Multi-Cue Fusion and Sim-to-Real Self-Training Authors: Hanyang Hu, Zekai Liang, Florian Richter, Michael C. Yip Published: 2026-10-04 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2610.05491v1 Efficient and robust tracking of surgical robotic instruments is important for robot-assisted minimally invasive surgery, yet remains challenging due to the complexity of surgical scenes and the unconventional geometry of surgical instruments. Keypoint-based approaches are efficient, but their performance depends on reliable feature detection. Improving these detectors with real-world supervision is difficult because accurate real-world annotations are costly to obtain at scale. To address this limitation, we introduce a tracker-guided self-training framework that adapts a model pretrained on synthetic images to unlabeled real-world videos. Given measured robot joint states, an uncertainty-aware EKF recursively corrects the instrument pose and the observable joint angles by comparing projected model features with detected keypoints, shaft boundaries, and mask-derived cues. An RTS smoother subsequently refines the resulting trajectory, which is projected into pseudo-labels for fine-tuning the feature detector without laborious pose annotations. Experiments on real-world videos demonstrate consistent improvements from self-training across all evaluated keypoint metrics, and the resulting model outperforms prior approaches in both accuracy and runtime. The code and data will be released upon publication. --- ### Universality and Convergence of Generative Flows Authors: Leo Brunswic Published: 2026-10-04 Categories: cs.LG, math.PR arXiv: https://arxiv.org/abs/2610.05490v2 Generative flows sample from an unnormalized target by training a flow to be balanced, and the training loss is the signal a practitioner watches. We ask what that signal is worth: whether a small loss certifies an accurate sampler, whether the loss can be driven to zero, and how fast gradient descent does so. The loss decides the first. Losses that compare the two sides of the balance by their difference bound, in total variation, the error of the sampler the flow implies, with explicit constants that do not involve the policy; flow-matching losses that compare them through a ratio admit no such bound, already on a single cycle, whenever their generator is continuous at balance. On graphs, the backward policy decides the other two. Once it is frozen, balance becomes invariance under the backward chain, so that existence is free on finite graphs, and one constant --- the norm of that chain's Green operator, which plays the role of an inverse spectral gap --- fixes the order of the curvature of the loss around the balanced flow, from above and below, and sets a floor under the rate at which training converges near it. The mechanism is that gradient descent diffuses the flow along the backward policy. For the squared-logarithm generator of detailed and trajectory balance, training the balance loss on states converges globally on every finite path-connected graph, from every positive initialization. The constant can be infinite while backward trajectories are short on average, and exact flow matching can then fail. The bounds and rates are tested by exact computation on enumerable state spaces, and every theorem carries a certification status computed from a Lean~4 development. --- ### SkillGATE: Gate-Aware Monte Carlo Tree Search for Skill Retrieval Authors: Rongchen Zhao, Yu Chen, Yanming Yang, Shijia Xu, Juyuan Wang, Jin Xu, Zibin Zheng, Jingping Liu Published: 2026-10-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.05489v1 Skill Retrieval (SR) aims to identify the most relevant skills from external skill libraries, and becomes increasingly challenging as libraries grow in scale and diversity. Existing methods either rank skills independently or rely on predefined graph propagation and hierarchical routing, making them vulnerable to semantic distractors, local trapping, and early routing errors. We formulate SR as an adaptive information-foraging process that coordinates region-level navigation with skill-level selection according to the utility and uncertainty observed during search. Based on this formulation, we propose SkillGATE, a graph-guided hierarchical retrieval framework with Gate-Aware Monte Carlo Tree Search (MCTS). SkillGATE constructs a graph-preserving hierarchical index and performs adaptive retrieval through selection, expansion, simulation, and backpropagation. G-PUCT guides action selection, expansion explores new regions, simulation evaluates candidate skills, and backpropagation updates search statistics. Experiments on six SR benchmarks show that SkillGATE consistently improves diverse retrieval and reranking backbones, achieving a 16.3\% improvement in overall R@1 over the strongest retriever-based baseline. Our code is available at https://github.com/Edwinbe/SkillGATE-v1/. --- ### Underscoring the Problem: Why Softpick Fails at Initialization Authors: Aryan Sood, Jaikaran Singh, Ishaan Bansal Published: 2026-10-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.05488v1 Softmax attention gives every token a nonzero weight, which in trained models concentrates into attention sinks and massive activations that widen the dynamic range low-precision inference must cover. Softpick removes this constraint by rectifying scores, eliminating sinks and lowering hidden-state kurtosis, but its advantage fades at scale. We reframe this failure as a normalization problem. Softpick's denominator splits into positive- and negative-shifted sums $D^+$ and $D^-$, used identically in the forward and backward pass, preventing their roles from being isolated. We separate them into a family of operators that independently choose each denominator. The failure originates at initialization: every layer contains rows where $D^+$ is exactly zero, while near-dead rows produce gradient norms above $10^{12}$ regardless of the backward denominator. Only Softpick and a stop-gradient variant, which keeps $D^+ + D^-$ forward but backpropagates through $D^+$ alone, train from scratch. At 230M parameters, the stop-gradient operator matches Softpick on quantization, has fewer dead heads, and retrieves passkeys more reliably, trailing only on peak attention-weight kurtosis. --- ### Universal Test-Time Training Authors: Zefan Cai, Qinzhe Hu, Ziqiao Ma, Hao Tan, Junjie Hu Published: 2026-10-04 Categories: cs.LG, cs.CL, cs.CV arXiv: https://arxiv.org/abs/2610.05484v1 Recent Test-Time Training (TTT) architectures compress context into fast weights that are updated online and queried as memory. Existing TTT designs keep this memory private to each layer: it recurs only over time, and depth merely indexes L separate memories. We argue that memory ownership need not be tied to depth, and introduce Universal Test-Time Training (uTTT), in which all layers read and write one shared memory while retaining layer-specific backbone parameters. The shared memory thus recurs over two dimensions, time and depth, with chunks and layers as their units: a write by a deep layer in one chunk can be read by a shallow layer in the next. We instantiate this idea as uTTT-MoE and uTTT-Dense. uTTT-MoE routes each token head to a few experts in a pool shared by all layers; uTTT-Dense applies the whole shared memory at every layer without routing. In language modeling, uTTT-MoE reaches 15.5 and 27.9 RULER accuracy at 124M and 760M, 2.6 and 2.1 points above its layer-private counterpart at equal state and active compute, the highest among tested bounded-state models, with per-token loss matching or beating full attention. In novel view synthesis, sharing at fixed per-layer compute gains 0.92 dB in view-23 object PSNR in routed models and 0.76 dB in dense models. --- ### FLEX-WAM: Flexible Block-Causal World-Action Models for Long-Horizon Imagination and Planning Authors: R. Khorrambakht, Joseph Amigo, Félix Lebel, Leon Seetoo, Jean Ponce, Zhenzhen Li, Ludovic Righetti Published: 2026-10-04 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2610.05483v1 World--action models (WAMs) promise a unified model that predicts action-conditioned futures, generates feasible actions, and supports planning in imagination. However, existing joint video--action models often use computationally heavy, fixed-horizon backbones ill-suited to streaming inference and stable long-horizon open-loop rollouts. We introduce FLEX-WAM, a Flexible and Efficient Block-Causal World--Action Model for unified simulation and policy inference. FLEX-WAM supports variable-length contexts and non-causal prediction horizons, as well as infinite autoregressive generation frame by frame or block by block. Its block-causal, KV-cacheable architecture combines axial attention and blockwise diffusion forcing to enable efficient real-time rollout and deployment-time latency--throughput tradeoffs without retraining. Joint training can nevertheless produce plausible futures that weakly respond to commanded actions. We address this failure mode by balancing state and action flow-matching gradient contributions across the state--action diffusion-noise grid and regulating world-model sampling using Forward-Dynamics (FD) elasticity, an efficient training-time proxy for action responsiveness. Across simulated and real-world datasets, FLEX-WAM achieves superior multi-step prediction quality and latency while producing stable joint state--action rollouts for thousands of steps. As a joint action proposer and simulator within MCTS, it solves long-horizon PushT and all five OGBench Puzzle-4x4 tasks entirely in imagination. On a bimanual OpenArm-based robot, a single checkpoint jointly serves as a play policy and expected-outcome predictor, enabling real-time identification and collection of model--reality mismatches for future self-improvement. --- ### Reflecting on Creative-Boundaries with an AI Co-Doodler Authors: Samia Menon, Samyukta Jayaram, Chetan Goenka, Shm Garanganao Almeda Published: 2026-10-04 Categories: cs.HC, cs.AI, cs.GR arXiv: https://arxiv.org/abs/2610.05482v1 In this pictorial, we consider how the negotiation of creative boundaries with a co-creative AI system can create moments for personal creative reflection. We ground this in our experiences with Froggi-Draw, a single-initiative co-doodling system that gives users power to decide when and how much an AI "collaborator" (Froggi) contributes to their drawing. From a 1-week pilot study where (n=8) novice and experienced artists doodled daily with the system, we report on ways participants navigated creative risk and uncertainty, and how their usage and perceptions of Froggi shifted over time. In moments of disruption, participants described the system as encroaching on their creative territory. We consider how the design of a supportive, co-creative AI "collaborator" might look like the design of a supportive power dynamic---and finding ways to offer users control to find, reflect upon, and flexibly negotiate the boundaries of that dynamic. --- ### CodeForge-MA: Execution-Verified Multi-Agent Learning with Language-Conditioned LoRA for Multilingual Code Generation Authors: Zhizhou Gu, Xianting Wu, Siyu Gu, Tian Zhang, Kejian Tong Published: 2026-10-04 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.05481v1 Large language models for code generation often fail on execution, multilingual coverage, and contamination control, especially under frozen backbone constraints. We present CodeForge-MA, a unified framework that improves code synthesis through a multi-agent data forge, execution verified reinforced instruction tuning, and a language conditioned mixture of LoRA adapters. Four specialized agents, Composer, Reviewer, Executor, and Curator, iteratively refine instruction code pairs, validate them with tests, and filter duplicates and benchmark leakage. During training, we combine masked supervised fine tuning with a test driven reinforcement objective to align generations with executable correctness. For the larger model, we use sparse expert routing over low rank adapters to improve cross language transfer while keeping the base model unchanged at inference. Experiments show that joint data, objective, and adapter design yields robust gains across programming languages. --- ### Population Scaling or Data Dilution? Dynamics of Local Topology Evolution in Decentralized Learning Authors: Yin-Kuan Liang, Yan Gao, Yang Long Published: 2026-10-04 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.05476v2 Scaling decentralized learning changes not only the number of clients $N$, but also the dynamics of information propagation and consensus. We argue that the effect of increasing $N$ cannot be understood in isolation, because data allocation, topology-dependent mixing, and communication capacity may change simultaneously. We study these coupled effects on CIFAR-10 with $N\in\{10,50,100,200\}$, comparing a degree-two Ring, a Static Random graph, and Local-First Heuristic Evolution (LFHE), a locally adaptive topology process based on friend-of-friend discovery. The Ring provides an analytically transparent failure mode: its Metropolis spectral gap decays as $Θ(N^{-2})$, implying progressively slower contraction of model disagreement as the population grows. Experiments show that holding the nominal local dataset size fixed substantially reduces the apparent population penalty observed when a fixed total dataset is divided among more clients. The remaining degradation depends strongly on communication structure: Ring enters a high-disagreement regime, whereas Static Random and LFHE remain close to consensus. Increasing LFHE's degree threshold further improves accuracy and consensus, but at a substantially higher model-transmission cost. These results show that decentralized scaling is governed by coupled learning and communication dynamics, rather than by the number of clients alone. --- ### Hierarchical Reinforcement Learning with Stable Temporal Abstraction for Language Model Agents Authors: Shayan Mohajer Hamidi, Yize Cheng, Yuanda Xu, Zhengze Zhou, Alborz Geramifard Published: 2026-10-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.05473v1 Hierarchical reinforcement learning improves long-horizon control by organizing primitive actions around persistent subgoals and assigning credit at multiple temporal scales. Recent hierarchical language agents bring these benefits to interactive tasks by explicitly separating subgoal planning from action execution. We observe, however, that an explicit hierarchy does not by itself determine how stable the resulting temporal abstraction is: the learned boundary policy may replace the subgoal almost every turn, making it effectively transient, or retain a subgoal after it has stopped being appropriate. We call this temporal abstraction instability. We propose Stable Temporal Abstraction via Constrained Optimization (STAC), a constrained boundary-policy optimization method that represents premature replanning and stale persistence as constraint costs. STAC applies the resulting Lagrangian costs only to the sampled boundary decision, leaving the underlying algorithm's rewards, critic targets, subgoal advantages, and primitive-action advantages unchanged. Across two backbones and two benchmarks, STAC improves success over a strong hierarchical baseline by $8.1$ and $7.9$ points on ALFWorld and WebShop with Qwen3-0.6B, and by $23.5$ and $15.8$ points with Llama-3.2-1B-Instruct. --- ### Hallucination Across the Reasoning Lifecycle: Interface Visibility, Causal Evidence, and Release Control in Large Reasoning Models Authors: Zhe Yu, Mohan Li, Lei Yu, Ka-Ho Chow, Chengwei Qin, Xingyu Wu, Wenpeng Xing, Shuguang Xiong, Meng Han Published: 2026-10-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.05472v1 Reasoning errors can propagate into later decisions and memory. This survey synthesizes 312 papers and first-party reports on text-based reasoning hallucinations around three questions: what evidence is observable, what study designs establish, and which corrective actions the evidence supports. UIPCA records unsupported premises (U), invalid inferences (I), dependent reuse (P), visible answer-trace consistency (C), and action-policy failures (A). Across 58 reviewed sources, no comparison establishes that a specified intervention improves reasoning while reducing factual reliability under matched conditions. The synthesis connects diagnosis to verification, repair, selective release, and persistent-state control across memory, tools, and training feedback. --- ### Writing as a Self-Organized Critical Process Authors: Nikolay Mikhaylovskiy Published: 2026-10-04 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.05466v1 We explain autocorrelation decay power laws omnipresent in texts by self-organized criticality. Specifically, we analyze the recently released KLiCKe keystroke dataset and show that not only the final texts' autocorrelations form a manifold that adheres to a power law with a finite-size scaling, but also the text revisions generate revision-size-dependent restoring dynamics toward that manifold. Thus, human writing appears to dynamically regulate semantic correlations in a text toward a critical state. --- ### Human-Like Attention? A Psychophysical Comparison of Visual Search in Humans and MLLMs Authors: Renchi Zhang, Joost C. F. de Winter, Dimitra Dodou, Harleigh C. Seyffert, Yke Bauke Eisma Published: 2026-10-04 Categories: cs.HC, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2610.05463v1 Visual search is a fundamental cognitive ability. This study investigates whether Multimodal Large Language Models (MLLMs) exhibit human-like difficulty signatures in visual search tasks. We compared search performance of humans (n = 1,250) and MLLMs using identical 2D and 3D stimuli across different set sizes. Both groups showed efficient performance in feature searches, most clearly when the target had a unique color, but performance degradation in conjunction searches as set sizes increased. Additionally, we found strong correlations between human and MLLM error rates ($ρ= 0.82$), which suggests that MLLMs are sensitive to similar objective complexities, such as stimulus heterogeneity. However, differences were found as well: whereas humans invested extra search time to respond accurately on target-absent trials, MLLMs exhibited extreme present/absent response biases in complex searches. We conclude that MLLMs replicate high-level human performance signatures, yet their underlying computations differ significantly. --- ### AI Safety via Debate is Compromised by Cognitive Biases Authors: Gefei Liu, Sonya Rashkovan, Sophia Lloyd George, Isaac Sheidlower, Serena Booth Published: 2026-10-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.05461v1 Reinforcement learning from human feedback (RLHF) has played a central role in making large language models responsive to human instructions. However, human evaluators often favor flattering or persuasive responses over truthful ones, creating incentives for models to appeal to evaluators at the expense of accuracy. AI safety via debate has been proposed as a way to improve the supervision of language models: in this paradigm, two agents argue opposing positions and challenge each other's claims, potentially exposing falsehoods to the adjudicator. A central premise of AI safety via debate is that truthful arguments are easier to defend than false ones under adversarial scrutiny. In this work, we investigate whether this advantage persists when debaters use rhetorical strategies that exploit biases in human judgment. Inspired by competitive debate, we construct 68 LLM-generated dialogues about detective mysteries with known culprits, spanning four interventions: anchoring, fallacy oversight, pro-jargon, and verbosity. We apply each intervention to either the side advocating for the true culprit or the side advocating for an innocent suspect, allowing us to distinguish influence on adjudication from correctness. In a study with 369 participants, we find that, pooled across bias types, these interventions significantly shift judgments toward the manipulated side. These findings expose a vulnerability in debate-based supervision: human adjudication is sensitive to manipulative rhetorical strategies. --- ### Measuring and Reducing Cross-Vendor Mismatch in Language Models Authors: Erland Hilman Fuadi, Chong Tian, Xiaosong Ma, Qirong Ho Published: 2026-10-04 Categories: cs.LG, cs.DC arXiv: https://arxiv.org/abs/2610.05458v1 Running the same language model on different graphics processing unit (GPU) vendors can produce different logits, even when the model weights and inputs are the same. We analyze cross-vendor mismatch in two dense and two mixture-of-experts (MoE) models with five metric families, namely bitwise equality, logit differences, top-K consistency, token agreement, and task accuracy. We trace one source of the mismatch to accumulation order inside vendors' matrix instructions. Upcasting to FP32 reduces the dense model's logit error by 43% at three times the runtime, yet keeping only the MLPs in BF16 retains 94% of this gain at 1.3 times the runtime, so most of the cost of full upcasting buys little. In the MoE models, FP32 and FP16 both lower the probability error but raise the logit error and change expert selection, and FP16 fails in the dense model. An output-head low-rank adapter (LoRA) does not help either, since the final hidden state does not predict the mismatch. The mismatch also carries into training. With every seed fixed, a student distilled from a teacher running on AMD answers 431 MMLU questions differently from one distilled from the same teacher on NVIDIA. Under FP32 upcasting, bitwise equality barely changes while the output distributions move most of the way to the reference, so judging cross-vendor agreement by a single measure misreads both its cost and its gains. Code is available at https://github.com/crova-project/crova. --- ### Taylor Representations for Model-Free RL in Networked MDPs Authors: Salah Chikhi, Abdelhaq Chaoui, Asuman Ozdaglar, Saurabh Amin Published: 2026-10-04 Categories: stat.ML, cs.LG, math.OC arXiv: https://arxiv.org/abs/2610.05456v1 In Networked Markov Decision Processes, transition dynamics are often unknown and the state--action space grows rapidly with the number of agents. In this setting, Taylor representations naturally approximate $Q$-functions, but a naive order-$n$ expansion over $N$ agents requires $Θ(N^n)$ coefficients. We justify these expansions under smooth expected future local rewards with controlled derivatives. Under this condition, finite-speed information propagation and discounting imply that local-critic Taylor coefficients decay exponentially with the graph distance to the farthest agent involved. Discarding distant-agent coefficients and marginalizing then yield scalable local Taylor representations with a bound controlled by graph locality. Building on these representations, we propose a scalable model-free actor--critic algorithm, establishing finite-sample critic and near-stationarity guarantees for a linear LSTD critic. We then introduce a more expressive neural TD parameterization. Unlike prior constructive spectral methods, our approach covers settings without access to a known local dynamics map, such as hidden switched linear--quadratic regulation. Across three control benchmarks, our method matches or outperforms spectral baselines while scaling efficiently to large graphs. --- ### The Poisoned Conversation: Privacy-Leaking Watermarks in Unified Multimodal Models Authors: Tobias Braun, Jonas Henry Grebe, Emil Sivic, Patrick Mohr Gordillo, Hossein Shakibania, Marcus Rohrbach, Anna Rohrbach Published: 2026-10-04 Categories: cs.CR, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2610.05453v1 Multimodal models are increasingly shifting toward unified architectures that understand and generate text, images, and other modalities within a shared conversational context. This design enables fluid interaction across modalities, but it also changes the privacy threat model: Information revealed in one part of a conversation may remain accessible when the model later generates content in another modality. This risk is particularly concerning in settings where users rely on locally deployed models for privacy, assuming that sensitive interactions remain confined to their device. We introduce Privacy-Leaking Watermarks (PLWs): invisible, trigger-dependent watermarks that a malicious model provider can condition on prior chat history. With this adversarial intervention, the usual separation breaks: a sensitive keyword or semantic cue mentioned earlier in the conversation can cause a later, unrelated image to carry a hidden yet detectable watermark. PLWs pose a novel threat to users of unified multimodal models: A poisoned model can retain utility while covertly turning image generation into a channel for privacy leakage, even when deployed locally. Across 13 sensitive-attribute triggers and two model families, PLWs reach up to 100.0% TPR at 1% FPR. For example, across all tested conversational separations, OmniGen2 detects every prior disclosure of depression while falsely flagging only 1% of images generated without such a disclosure. --- ### VERA: Verdict-Conditioned Reliability for Adaptive LLM Judges Authors: Qiushui Xu, Syamil Mohd Razak, Tao Yuan, Piotr Habas Published: 2026-10-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.05452v1 Accurately estimating judgment reliability is a central challenge in adapting LLM judges to newly verified feedback while preserving previously learned behavior. However, existing approaches often rely on output-level confidence, which can be overconfident and poorly aligned with judgment correctness. We propose VERA, a VErdict-conditioned Reliability Axis that estimates reliability from hidden activations by distinguishing correct from incorrect judgments within each predicted-verdict group. Using VERA as a control signal, we develop a VERA-guided periodic adaptation framework that integrates reliability-ranked corrective updates, reliability-residual replay, and periodic refresh of the reliability directions. After VERA-guided adaptation on Chatbot Arena, 8B- and 14B-parameter judges outperform the strongest baseline on each of four held-out public benchmarks, with relative gains of up to 23.01%. The framework also improves focal-class recall by up to 16.1% relative to the strongest adaptive baselines on a separate proprietary temporal auditing task. --- ### Groupwise Distortion Guarantees for Preference-Based Alignment Authors: Jacob Brodkey, Roberto Tamez, Aaron Roth Published: 2026-10-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.05450v1 Preference-based alignment methods such as reinforcement learning from human feedback (RLHF) and Nash learning from human feedback (NLHF) aggregate pairwise preferences to learn an LLM policy, but a natural goal is maximizing social welfare (average cardinal utility), which comparisons alone need not identify. Gölz, Haghtalab, and Yang (GHY) measure the gap by distortion: the worst-case ratio between the welfare of the best fixed lottery (distribution over responses) and of the learned lottery. They show NLHF is optimal when every user receives the same lottery. Account-based LLMs, however, have information about their users and can serve different lotteries to different people. We give an efficient algorithm, GLHF, that learns a single group-conditioned policy from one comparison per user. Under individual Bradley--Terry comparisons, GLHF asymptotically matches GHY's optimal population distortion bound simultaneously on every group in a prespecified, possibly overlapping collection, with sample complexity growing logarithmically in the number of groups and inversely with the smallest group mass. A sharper guarantee for groups with similar preferences approaches distortion of one when members share a feasible favorite response. In experiments using human coffee ratings and synthetic LLM-generated ratings, GLHF lowers distortion in every evaluated group and substantially reduces worst-group distortion relative to NLHF and other group-agnostic baselines. --- ### Hierarchical Time-aware Bootstrapping for Off-Policy Subgoal Value Learning Authors: Bingyun Liu, Yuheng Jing Published: 2026-10-04 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.05446v1 Off-policy hierarchical reinforcement learning must estimate the values of high-level decisions while the low-level policy changes. HIRO adapts replay data through subgoal relabeling, but after a label change, the value update targets the relabeled subgoal instead of the subgoal the high-level policy originally needed to update. We propose Hierarchical Time-aware Bootstrapping (HTB), which evaluates specified subgoals under the current low-level policy while retaining accumulated task rewards. Remaining execution time distinguishes subgoal continuation from a new high-level decision. Together with primitive-action conditioning, it enables off-policy Bellman updates based on the stationary environment transition law. HTB combines these one-step updates with multi-step suffix returns and truncated relabeling, reducing dependence on intermediate value estimates. A shared value component supports learning across actions, while nonnegative residuals constrain upward corrections relative to that component. At a fixed mixture weight of 0.95, HTB achieves 32.8% AntFall success versus 9.6% for matched local HIRO over five paired seeds at 10M environment steps. Ablations identify contributions from recursive continuation and mixed supervision; fixed-policy tests show more accurate predictions for actions whose returns were excluded from fitting. --- ### The sublevel Flood bifiltration: towards scalable 2-parameter persistent homology Authors: Mattéo Clémot, Julie Digne, Julien Tierny Published: 2026-10-04 Categories: math.AT, cs.CG, cs.LG arXiv: https://arxiv.org/abs/2610.05441v2 Multiparameter persistent homology is a rapidly developing branch of topological data analysis that improves the robustness of single-parameter persistent homology to outliers, while still capturing the metric characteristics of the data. However, a notable limitation is its lack of scalability. In this paper, we introduce a novel approach for efficiently computing 2-parameter persistent homology on large point sets. Our work extends the Flood filtration, originally developed for single-parameter persistence. Our construction, called the sublevel Flood bifiltration, offers a scalable approximation of the sublevel offset bifiltration. We show that it benefits from theoretical stability properties and describe how to compute it efficiently. We demonstrate the performance of our approach in classification tasks on low-dimensional synthetic datasets, where density awareness is critical, as well as on real-world time series datasets. ---