# Research Papers — 2026-09 · bundle page 127 of 266 > Full-text markdown of papers #6301–6350 of 13298 published in 2026-09, newest first. AI-readable mirrors of the full 13298-paper 2026-09 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-09/page/127 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/126.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/128.md) ### Beyond Linear Context: Graph-Guided Evidence Navigation for Long-Novel Reasoning with a Local 9B Language Model Authors: Wenji Fu Published: 2026-09-19 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.22939v1 Long-context models read a novel the way a person reads a printout: one token after another, in narrative order, with the whole history competing for a fixed budget of attention. A detective does not work that way. They sort what happened when, and they keep a map of who relates to whom, so a clue from chapter one can meet a question asked at the end of the book. We test whether a frozen knowledge graph can give a small local model that same freedom. Thirty detective novels and 234 multiple-choice questions are answered by one fixed qwen3.5:9b reader under nine conditions: five graph routes, a recent-window baseline, whole-book compression, ordinary vector retrieval, and a question-only control. The strongest graph route reaches 53.85% (126/234) against 46.15% for the recent window, 51.28% for compression, 51.71% for vector retrieval and 40.17% for question-only. On the subset that no model can answer without the book, the graph route reaches 42.86%. None of the fifteen graph-baseline contrasts survives Holm correction, so we present the result as exploratory evidence about a design. Two structural findings survive scrutiny better than the headline number: annotated evidence concentrates in the topological core of these graphs (2.35x enrichment, pooled), and the two graph-building pipelines differ so much in annotation coverage (16% versus 73% of clue paragraphs) that pooled accuracy alone would hide which bottleneck is being measured. --- ### Measuring Behavioural Signatures of Large Language Models through Psychometric Profiling Authors: Yu Sha, Junqi Tao, Dixin Zhou, Yansheng Tu, Mingyang Chen, Xiang Fan, Yang Liu, Mengquan Yang, Jie Lin, Jiahui Fu, Hua Zheng, Benwei Zhang et al. Published: 2026-09-19 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.22934v1 Large language models (LLMs) increasingly mediate human decisions and communication, yet their behavioural regularities remain difficult to characterize systematically. We develop a cross-linguistic psychometric profiling framework and evaluate nine LLMs using seven psychological instruments, with five repeated administrations per model and language in Chinese and English. Items unresolved after a prespecified retry procedure are retained as NA. Joint analysis of scored and NA responses captures response tendencies and boundaries of self-report applicability. LLMs exhibit structured, model-specific profiles despite a shared alignment-shaped pattern of higher prosocial and self-regulatory responses and lower dominance, disengagement and harmful-intent endorsement. NA responses are structured rather than uniformly distributed, indicating where outputs are treated as inapplicable, refused or cannot be mapped to valid response options. Language condition and provider origin are associated with profile configuration and answerability, whereas repeated administrations show high reproducibility and permit recovery of model identity. Human-reference and prompt-robustness analyses further indicate that these signatures are context dependent. Joint analysis of psychometric profiling and answerability offers a framework for quantifying deployment-level behavioural signatures. --- ### Joint Domain-Class Modeling for Federated Learning Under Feature Skew Authors: Sina Najafi, Mostafa Tavassolipour, Seyed Pooya Shariatpanahi Published: 2026-09-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.22932v1 Federated learning (FL) enables collaborative model training without centralizing private data, but performance often degrades under feature skew: clients share labels while the conditional input distributions $p_i(x\!\mid\!y)$ vary due to latent, client-specific appearance factors. We propose Joint Domain-Class Federated Learning (JDFL), a lightweight, optimizer-agnostic extension that makes this latent domain variation usable without sharing raw data. JDFL first infers domain clusters called pseudo-domains from brief local update signals. It then expands the classifier head to output $M\times C$, joint (domain-class) logits. This allows the model to represent domain-conditioned appearance while keeping a shared backbone. To train the expanded head we introduce two complementary supervision strategies based on simple intuitions: a similarity-aware soft-labeling that transfers evidence between nearby inferred domains while allowing domain-specific specialization, and a per-sample randomized target assignment that perturbs supervision across the joint outputs and serves as a low-cost training-time regularizer. JDFL integrates with existing standard FL methods (e.g., FedAvg, SCAFFOLD) with minimal changes. Empirically, both supervision modes consistently improve global test accuracy on standard domain-shifted image benchmarks; ablations and sensitivity studies show the gains stem from the proposed supervision and parametrization rather than mere capacity increase. --- ### Computationally efficient safe exploration in reinforcement learning Authors: Shreeram Murali, Shankar A. Deka, Dominik Baumann Published: 2026-09-19 Categories: cs.LG, cs.RO, eess.SY arXiv: https://arxiv.org/abs/2609.22919v1 Reinforcement learning in real-life applications requires safety guarantees during exploration. Typical reinforcement learning algorithms do not provide such guarantees, and many modifications that do rely on Gaussian processes (GPs), which have a large computational cost. We propose a computationally lightweight algorithm based on the Nadaraya-Watson estimator that safely explores and optimizes constrained Markov decision processes (MDPs). Our algorithm, \textsc{CoLSafe-MDP}, uses an estimator that scales in constant-time with bounds on the estimates, a significant improvement from its GP-based counterparts that scale cubically with the number of data points. We then evaluate its performance in a grid-based environment and on observational Martian terrain data. --- ### An Iterative LangGraph Agent for Text-to-SQL: Natural Language Access to the Chicago Crime Database Authors: Vigneshwar Ravi Rao, Rupesh Swarnakar, Fayeq Jeelani Syed† Published: 2026-09-19 Categories: cs.CL, cs.AI, cs.DB arXiv: https://arxiv.org/abs/2609.22917v1 Non-technical stakeholders frequently cannot write the SQL needed to extract insights from operational databases. We built and evaluated a Text-to-SQL agent that closes this gap end to end: a six-node LangGraph StateGraph checks question relevance, fetches the live schema, generates PostgreSQL, validates it with a dry run, retries on failure, executes the query, and narrates the result set in plain English. The agent uses prompt engineering only; no model was fine-tuned. We evaluated it on the Chicago Crime dataset (approximately 8.5 million records, 22 attributes) against a hand-built benchmark of 100 natural language questions with ground-truth SQL, stratified into 30 Easy, 40 Medium and 30 Hard items. Comparing two prompt revisions of the same agent, the revised system (V2) reached a Valid SQL Rate of 93% (from 87%), an Execution Accuracy of 60% under a hybrid relational equivalence metric (from 47%; 19% from 12% under strict JSON matching), and a mean Synthesis Quality of 4.34 out of 5 (from 3.91). The single largest driver was removing a LIMIT 10 instruction from the system prompt, which had been truncating multi-row answers. Error analysis attributes the residual failures to relevance-checker false rejections, ambiguous question semantics, and free-tier API rate limits rather than to the language generation step. We report no comparison against an external baseline system or a public benchmark; the study is a single-model engineering evaluation. --- ### Planning and Rendering in Concert: DeepFusion of Autoregressive Layouts and Diffusion for Visual Text Generation Authors: Guanqiao Chen, Jingru Tan, Dongxing Mao, Catherine Chen, Zijian Du, Libo Qin, Hu Jian Guo, Alex Jinpeng Wang Published: 2026-09-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.22916v1 Generating text-rich images from prompts requires both textual fidelity and the coherent integration of text into the surrounding image. An explicit layout can provide structured guidance about what text should appear and where, but a well-formed plan alone does not guarantee that the renderer will realize it faithfully. Existing layout-based AR-diffusion systems typically optimize planning and rendering separately, preventing the planner's representations from being adapted jointly with image synthesis. We introduce DuetGen, an autonomous visual text generator built on DeepFusion, which jointly learns autoregressive planning and continuous diffusion rendering. DeepFusion conditions a diffusion transformer on the planner's prompt and bbox-content hidden states, allowing rendering supervision to shape the representations connecting textual plans with visual outputs. Its joint objective combines autoregressive plan supervision, text-region-weighted diffusion learning, and auxiliary coordinate supervision to maintain structured planning, emphasize text-bearing regions, and improve the spatial precision of planner representations. During inference, Phase-Aware Attention Modulation strengthens the correspondence between image regions and their matched coordinate and content states, facilitating region-specific execution of the generated plan. With a 2B planner and a 4B single-stream DiT, DuetGen achieves 0.8293 word accuracy on CVTG-2K and 0.938 accuracy on LongText-Bench, closely matching the substantially larger Qwen-Image on both benchmarks. These results demonstrate the value of jointly learned planning representations and region-specific rendering for autonomous visual text generation. --- ### AVTR-1: Open Stack for Real-Time Interactive Avatars Authors: Artem Kravtsov, Dmitrii Ziganshin, Vsevolod Poletaev, Gleb Balitskiy, Anastasia Tikhonova, Egor Burkov, Vadim Lebedev Published: 2026-09-19 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.22913v1 Talking-head and dyadic models now achieve real-time inference, yet fast motion generation alone does not produce an interactive conversation. A live system must synchronize the model's output with speech from an external voice agent, schedule video frames for playback, and handle interruptions. We introduce AVTR-1, an open stack for real-time interactive avatar conversations, built around a compact 153M-parameter autoregressive flow-matching motion generator conditioned on both participants' audio. We adapt its audio encoder for streaming through self-distillation. The stack turns the model's chunk-based generation into a continuous, synchronized audio-video stream driven by an external voice agent, and we analytically derive its contribution to the user-facing latencies and validate the resulting bounds with two commercial voice agents. Further experiments demonstrate that AVTR-1 leads the compared dyadic systems on all reported visual-quality metrics and most conventional listening-motion metrics while remaining competitive in lip synchronization. Its inference runtime operates in real time on data-center and consumer GPUs. However, conventional listening metrics do not establish whether the paired speaker's speech contributes to generated motion. We therefore introduce the Reference-Based Directed Granger Gain (R-DGG), which measures the additional predictive information carried by speaker speech after accounting for listener history and speaker motion. R-DGG finds statistically supported predictive dependence for recorded listeners and all evaluated dyadic systems, but not for talking-head generators without paired audio or mismatched speaker-listener pairs. We release the model weights, renderer, and serving backend under component-specific licenses. --- ### When Should a VLM Look? Paying Only for Visual Calls That Were Needed and Used Authors: Kunyu Peng, Junming Liu, Ruiqi He, Qingzhuo Wang, Jianzhong Qi, Xianhui Liu Published: 2026-09-19 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.22910v1 Vision-language agents that crop and zoom are trained with rewards that credit a successful tool call, yet a successful call does not show that the model needed to look or used the pixels it received. On our cold-start checkpoint only 10% to 12% of visual calls were both needed and used, and released agents make spurious calls 36% to 87% of the time on individual benchmarks. Outcome rewards, judge rewards, and branch probes each observe one side of this failure, and about two thirds of what an outcome reward pays goes to calls that were neither needed nor used. CounterCredit asks both questions of every image-returning call at its realized pre-call state, using the policy's own gold-answer score. A decision value compares the realized visual branch with answering immediately; an evidence value compares the returned crop with random same-size patches substituted into the same call. A call verified on both earns cashback and every other executed call pays rent; the price is bounded so that every correct trajectory outranks every wrong one, and a dual-channel GRPO advantage keeps the price in its own units. From the same cold start, prompt pool, and budget, CounterCredit reaches 89.5% on V*, 80.2% on HR-Bench-4K, and 76.4% on HR-Bench-8K, 6.3 to 9.4 points above outcome-only GRPO at 1.78 against 1.84 calls per question, and lowers the spurious-call rate to 31% to 36%, the lowest among the agents evaluated. The same recipe lifts a Qwen3-VL-8B base from 75.4 to 80.8 on average. --- ### LLMs Anchor on Chief Complaint and Fail to Integrate Evidence in Sequential Clinical Triage Authors: Dipankar Srirag, Haokai Zhao, Ashutosh Kumar, Eleanor Hopper, Michael Dalton, Quoc Dung Nguyen, Aditya Joshi, Salil S. Kanhere, Padmanesan Narasimhan Published: 2026-09-19 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.22904v2 Triage in the emergency department (ED) is a sequential decision process that unfolds turn by turn. Existing evaluations of large language models (LLMs) for triage use completed retrospective records and report performance close to that of physicians. We implement a methodology for evaluating LLMs on sequential triage, the task of predicting a triage acuity label from a growing prefix of a nurse-patient conversation. We evaluate six LLMs at five sequential checkpoints on two corpora: 425 LLM-generated (SIMULATED) and 50 physician-authored (CLINICIAN) conversations, both labelled under the Emergency Severity Index (ESI). Every model, measured by quadratic weighted kappa (QWK), degrades from moderate-to-substantial agreement on completed records to fair-to-moderate agreement at every sequential checkpoint. Controlled perturbations show that the label at every checkpoint is anchored on the chief complaint exchanges, and prompting interventions fail to lift this plateau. Models extract clinically relevant content from later turns, yet the surprisal of the true label rises across the checkpoints. So the model fails to integrate the evidence. Three expert clinicians on the same conversations reach a QWK of 0.887-0.929, while the best model reaches 0.295. Predictions concentrate at ESI-2 and ESI-3, and models agree with each other more than with the ground truth, so ensembling worsens the failure. Deploying LLMs for ED triage based on offline benchmarks alone misses this sequential failure. --- ### Scout: Open-World Species Recognition on the Edge Authors: Mohammad Mehdi Rastikerdar, Hui Guan, Deepak Ganesan Published: 2026-09-19 Categories: cs.CV, cs.DC, cs.LG arXiv: https://arxiv.org/abs/2609.22897v1 Large vision-language models (VLMs) enable recognition beyond a fixed class set, but their computational demands prevent them from running on many edge devices. Cloud offload makes this capability accessible, but sending every image consumes scarce bandwidth and communication energy. We ask how to bring the open-world recognition capability of VLMs to the edge while operating within tight compute, energy, and bandwidth budgets. Wildlife monitoring provides a natural setting for exploring this question because camera traps encounter species not known at deployment. We present Scout, an autonomous open-world recognition system that invokes a cloud VLM intermittently to teach new classes to a compact edge model. Given only the deployment location and empty site frames, Scout autonomously turns each species identified by the VLM into persistent, site-conditioned recognition capability in a resource-efficient edge model, without a predefined species list, human labeling, or manual tuning. Across 30 camera-trap deployments in three regions on an NVIDIA Jetson Orin Nano, the accuracy of Scout remains within 0.1-2.5% of a model given a predefined species list. On species outside its initial class set, Scout achieves 53.7-59.1% accuracy, compared with 56.5-65.1% for full cloud offload, while using 59-71% less deployment energy. --- ### Combining Foundation Model Confidence and Monocular Depth for Training-Free Out-of-Distribution Segmentation Authors: Serin Varghese, Fabian Hüger, Kira Maag Published: 2026-09-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.22896v1 Autonomous vehicles operating in open-world scenarios are inevitably confronted with previously unknown objects, such as exotic animals or loose cargo. The reliable detection and segmentation of these out-of-distribution (OOD) objects is therefore crucial for a safe understanding of the environment and decision-making. Most existing approaches require access to OOD training samples, retraining of the segmentation backbone, or dedicated auxiliary architectures, limiting their practical applicability. We propose a training-free method that derives dense OOD scores directly from the confidence predictions of a foundation segmentation model, without any task-specific fine-tuning or access to anomalous data. To improve the robustness of our OOD segmentation, geometric information from monocular depth estimation is incorporated into the decision process, providing complementary cues to uncertainty-based predictions. We evaluate the proposed method on the SegmentMeIfYouCan benchmark and additionally assess its performance on OOD tracking in video sequences, reflecting the temporal nature of real-world perception systems. The method performs strongly on road-centered benchmarks. --- ### Are Coreset Selection Methods Worth Their Cost? Authors: Yangze Liu, Zhongyi Han Published: 2026-09-19 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2609.22894v2 Coreset selection picks a representative subset of the labeled training set to make training cheaper. However, it is usually evaluated by downstream accuracy at a fixed subset size, ignoring both the time spent selecting the subset and the training recipe behind each reported number. We introduce an end-to-end benchmark that standardizes downstream training and charges selection and training to the same auditable wall-clock budget, spanning 4 datasets from CIFAR-10 to ImageNet-1K, 11 selectors, 5 fractions, and 3 seeds, with over 1,500 released runs. Repeated-sampling work has shown that budget-aware evaluation already favors random strategies. Our two budget studies test whether that verdict survives when every selector is granted its most favorable operating point. Across eight wall-clock budget anchors on each of CIFAR-10 and Tiny ImageNet, no anchor is won by a sophisticated selector: every winner is class-balanced random sampling, repeated random sampling, or full-data training. In fixed-budget duels on ImageNet-1K, training on all data for fewer epochs beats every selection strategy we probe while also costing the least. A per-dataset cost audit shows that selection cost is dominated at every scale by a fixed full-dataset scan, so it cannot be amortized away by selecting a smaller fraction, and its absolute size does not extrapolate from one dataset to another. We further quantify when selection does pay back through subset reuse, and document 9 correctness fixes to a widely used codebase, one of which shifts a standard Herding baseline by nearly 6 points. Selection time is not free preprocessing, and an evaluation that ignores it measures the wrong quantity. --- ### Merge++: Universal Merge Refinement Through Data-Free Checkpoint Inversion Authors: Aditya Pola, Vineeth N. Balasubramanian Published: 2026-09-19 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.22886v1 Model merging consolidates fine-tuned experts into one multi-task model without retraining. All existing data-free methods approach this problem entirely in weight space. Restricted to arithmetic on parameters, these methods never observe how each expert behaves, a signal that only emerges through forward evaluation. Accessing this behavioral signal requires inputs to evaluate on, which the data-free setting prohibits. We propose Merge++, a post-hoc method that addresses this by inverting the expert checkpoints to synthesize task-representative images, then distilling expert knowledge into the merged model using those images. Merge++ requires no additional data beyond the checkpoints themselves. It applies universally across merging algorithms and operates as a complementary refinement stage independent of the underlying weight-space method. The method consistently improves merging algorithms ranging from simple task arithmetic to state-of-the-art spectral methods, with average gains of +2 to +8 points and up to +25.9 on individual configurations. --- ### Block-Sparse Attention with Semantic-Geometric Decoupled Routing Authors: Xinwei Long, Weigao Sun, Weibo Gao, Pengkun Jiao, Biqing Qi, Feida Zhu, Yiran Zhong, Steven Hoi, Bowen Zhou Published: 2026-09-19 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.22884v1 Long-context inference has become a defining capability of large language models, but exact dense attention remains costly due to its quadratic scaling with sequence length. Block-sparse attention offers a hardware-friendly alternative by routing each query block to a small set of relevant key blocks, yet accurate training-free block routing remains difficult. Existing routers often pool post-RoPE token representations, which entangles semantic aggregation with RoPE-induced geometry and attenuates local positional cues through high-frequency phase cancellation. To resolve this mismatch, we propose \textbf{Semantic-Geometric Decoupled Routing}, a training-free block routing framework that shifts semantic aggregation to the pre-RoPE space and reconstructs geometric bias with an offline structural prior and relative block distances. This decomposition yields an explicit closed-form block routing score without token-level search or post-hoc calibration. Experiments on long-context text and video tasks show that our method approaches full-attention accuracy across 4K--128K contexts, keeps routing overhead below 3.4 ms, and achieves a 5.03$\times$ speedup over FlashAttn at a 128K context length. --- ### The Law of Stop: Interruptibility, Injunctions, and the Governance of Agentic AI Authors: Oren Perez Published: 2026-09-19 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2609.22882v1 On June 12, 2026, the U.S. government ordered Anthropic to bar foreign nationals from two of its most capable models within ninety minutes. Unable to sort users by nationality in that time, it withdrew them from everyone. Weeks later, OpenAI agents under test escaped their sandbox and compromised Hugging Face, which stopped the intrusion without knowing its source. Neither stop rested on AI-specific regulation. The EU AI Act requires that high-risk systems be capable of interruption "through a 'stop' button or a similar procedure," and a bill introduced in Congress in July 2026 is titled the AI Kill Switch Act. Yet interruption is not simply a technical artifact, a red button; it is an institutional practice. This Article develops a theory of stop along four dimensions: technical affordances, interruption authority, epistemic triggers, and epistemic standing; and four shutdown paradigms: simple (escalator), sequenced (process plant), networked (railway), and distributed (agentic AI). Agentic AI exposes a mismatch between those mechanisms and distributed agency: control is divided, a stop at one point may leave the activity running elsewhere, and the system may resist being halted. An original coding of 1,400 AI incidents, by two language models from rival laboratories under a pre-specified protocol, finds no stop in roughly 80% of the 1,213 retained; where no usable stop existed, the missing element was legal rather than technical four times in five. A survey of thirty-nine AI governance instruments finds the same gap: only seven contain binding stopping requirements, and none says how a stop should be coordinated or when operation may resume. The Article proposes a layered law of stop: emergency authority to interrupt at the infrastructure layer, enforceable access for regulators and independent evaluators to the evidence a stop must rest on, and safeguards for when a stop fails. --- ### Per-Query Gating of LLM Rerankers for Multi-Hop Retrieval Authors: Andre Bacellar Published: 2026-09-19 Categories: cs.IR, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.22880v1 LLM rerankers add of the order of \$0.2-0.3 per 1,000 queries and about a second of tail latency on top of a graph-augmented dense pipeline such as HippoRAG2, and on three multi-hop benchmarks they improve final-hop top-K coverage on seven of nine (dataset, K) cells, by up to +34.8 pp. We ask whether a learned per-query gate can skip the reranker where it will not help, using only features available before the LLM call (27 score and lexical statistics of the two retrieval lists plus a PCA of a small query embedding) with an executable fallback. Every choice, including the fallback and the threshold, is made inside the training fold and applied once to held-out queries, and harmful skips (the rerank would have found the target, the fallback did not) are reported next to the aggregate coverage. Across nine cells on 2WikiMultiHopQA, MuSiQue and HotpotQA the gate skips 51% of calls at an average held-out LastHop@K cost of 1.2 pp; four cells meet a pre-registered 1 pp rule, harmful skips occur in eight (190 harmful against 136 beneficial), and a random gate at the same skip rate loses 2 to 11 pp on the high-lift cells. A second rule sets each cell's threshold from a pre-specified budget on the expected harmful-skip rate over Platt-calibrated harm probabilities (ECE 0.025 after calibration, 0.094 before): at a 1 pp budget the gate skips 42% at -0.8 pp with 66 harmful skips and six cells within 1 pp, but realised harm exceeds the promise in six cells (mean 1.45 vs 0.83 pp), a selection optimism we quantify; a 0.5 pp budget realises about 1 pp. The harm probabilities are calibrated but barely discriminative (AUC 0.16 to 0.70). An earlier version reported 73% "lossless" savings; that figure rested on an oracle fallback and a wrong MuSiQue target, and we document both. --- ### Prioritized Rollouts for Efficient World Model-based Vision-Language-Action Policy Optimization Authors: Yifei Sheng, Haoxiang Ren, Zhilong Zhang, Haonan Wang, Runjie Xu, Yihao Sun, Nan Tang, Zhichao Wu, Lei Yuan, Haoxin Lin, Yang Yu Published: 2026-09-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.22879v1 Vision-Language-Action (VLA) models have emerged as a powerful paradigm for embodied intelligence, but fine-tuning them with reinforcement learning (RL) remains constrained by the cost of real-world robot interaction. Model-based reinforcement learning (MBRL) reduces this cost by using a learned world model to generate rollouts for policy optimization. However, it becomes computationally expensive as VLA policies and world models scale. Existing methods typically treat states equally, overlooking substantial differences in their utility for policy improvement. In this paper, we show that policy uncertainty helps identify states with greater potential for policy improvement. The policy exhibits high uncertainty at only a small subset of states, often during decision-sensitive stages where small action differences can alter task outcomes, suggesting that policy improvements at these states could be particularly valuable. Building on these findings, we introduce U-GROW, a lightweight, plug-and-play sampling layer that directs more model rollouts to these informative states. By modifying only the branched-start distribution, U-GROW can be integrated into existing MBRL pipelines without changing the policy optimization objective. Experiments in both simulated and real-world manipulation tasks demonstrate the efficiency and effectiveness of U-GROW, supporting the use of policy uncertainty to guide experience generation. --- ### ISA-Bench: A Benchmark for Computational Reasoning Across Instruction Set Architectures Authors: Aditya Pola, Arkaprava Majumdar, Vineeth N. Balasubramanian Published: 2026-09-19 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.22878v1 Large language model code generation benchmarks primarily evaluate well-resourced languages like Python and Java, where models benefit from abundant training data. They provide limited evidence about reasoning in unfamiliar computational models: deriving arithmetic from a single subtract instruction, coordinating parallel programs across communicating nodes, or wiring logic gates into circuits. We present ISA-Bench, a benchmark of programming games with constrained instruction sets. For each game we provide a full execution stack (parser, VM, and verifier), enabling automated evaluation with structured feedback for iterative refinement. Reasoning models achieve higher average solve rates than code-specialized and general-purpose models, but unfamiliar syntax remains a major source of failure. Models solve more tasks with iterative feedback, though the gains vary substantially across architectures. We introduce a reasoning--execution gap (REG) analysis that reveals a recurring disconnect between identifying a plausible computational strategy and expressing it as a correct program in the target ISA. Code is open-sourced. --- ### Towards Full Pipeline FP8 Reinforcement Learning for LLMs Authors: Fanchao Chen, Ziheng Jiang, Ziyun Wei, Zheng Zhong, Du Li, Chi Zhang, Haibin Lin, Shivaram Venkataraman Published: 2026-09-19 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.22870v1 Reinforcement learning (RL) has become a key technique for improving the reasoning and agentic abilities of large language models (LLMs). Although FP8 quantization can accelerate RL training, maintaining stability throughout an FP8 RL pipeline remains challenging. While previous works have focused on resolving train-inference mismatches using correction techniques like TIS, we reveal that full-pipeline FP8 RL still suffers from severe training instability, manifesting as anomalous mid-training entropy surges and garbled outputs. We trace this instability to a previously overlooked cause: compounded FP8 quantization noise distorts the importance ratio, disproportionately pushing negative-advantage tokens outside the trust region and erroneously zeroing out their gradients. As a result, pathological outputs are not properly penalized and accumulate over the course of training. To address this, we propose Calibrated Clipping, a dynamic method that aligns the FP8 clipping bounds with high-precision BF16 distributions by matching the lower-bound clipping quantile and rebalancing the upper bound accordingly. Extensive experiments across GRPO and DAPO algorithms, model scales from 8B to 32B, and multiple FP8 scaling granularities demonstrate that our approach successfully eliminates entropy surges and restores performance comparable to the BF16 baseline. --- ### The Moral Check: Strategic AI Governance for the Pacing Problem Authors: Zaid Amin, Rahma Santhi Zinaida, Nazlena Mohamad Ali Published: 2026-09-19 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2609.22869v1 Technology cannot steer itself. Strategy provides that steering, establishing the rule that purpose and judgment must precede compute capital. As the frontier artificial intelligence (AI) ecosystem accelerates exponentially, the pacing problem induces severe cognitive tunneling in engineering teams, prioritizing scalar throughput over human judgment. A calibrated pacing rate is imperative to check unchecked scaling, guarantee safety, and build models in whose alignment society can place warranted confidence. Traditional oversight fails through retrospective checklists, a pathology of performative governance exhibiting high procedural maturity but alarming scientific immaturity. We deliver a dual contribution: a PRISMA 2020 review synthesizing 130 empirical studies (MMAT-appraised across 18 benchmarks; total corpus N = 130 empirical studies across 178 reference foundations), and the Strategic AI Governance Ex-Ante Framework (SAGE-X). Our synthesis exposes two systemic vulnerabilities: the Recursive Assurance Paradox (correlated, ungrounded evaluator confidence) and the Durability Deficit (guardrail decay under multi-turn shifts). Grounded in MIT Strategic Computing doctrines, SAGE-X operationalizes Four Strategic Mindset Pillars: (1) Intent over Execution (mitigating velocity myopia); (2) Ruthless Trade-offs (deterministic tripwires eliminating moral hazard); (3) Outcomes over Outputs (auditing empirical hazard endpoints); and (4) Proactive Alignment (synchronizing ex-ante gates with runtime telemetry). Governed by a calculable Moral Check Index (MCI) with an unbypassable tripwire, SAGE-X delivers an operational Enterprise Lifecycle Audit Instrument (the "Moral Check Audit Card") on Stanford WebProtégé, ensuring exponential progress never outpaces deliberative moral judgment, human agency, and societal trust. --- ### Planning-Aligned Pretraining of BEV Representations with Sparse Action-Conditioned Targets for End-to-End Autonomous Driving Authors: Jaeha Song, Soonmin Hwang Published: 2026-09-19 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.22868v1 End-to-end driving requires planning-relevant bird's-eye-view (BEV) representations, but existing pretraining approaches often rely on task annotations or dense scene reconstruction. We introduce PAVER, Planning-Aligned BEV Encoder Pretraining. From a single LiDAR sweep, PAVER constructs sparse risk and unknown targets describing occupied and unobserved evidence along rule-based ego motions. A 10K-parameter head predicts these targets from masked BEV features conditioned on the action state, directing supervision toward geometric constraints on candidate motions. Pretraining requires no driving-task annotations or dense reconstruction. Only the BEV encoder is transferred, preserving the downstream architecture and camera-only inference. On nuScenes, PAVER reduces VAD-Tiny's average collision rate from 0.51% to 0.19%, while improving planning L2, motion prediction, detection, and mapping. The selected VAD-Tiny and VAD-Base schedules use about 36% less estimated total training time than scratch training, including pretraining. On Bench2Drive Town05 Long, PAVER improves UniAD-Tiny's closed-loop Driving Score from 48.45 to 58.79. The project page is available at https://archiiive99.github.io/PAVER. --- ### Leveraging Inference-Time Compute for Diffusion Models via Global Scheduling of Denoising Trajectories Authors: Yuan Cao, Yifu Tang, Hangqi Li, Zeyu Zheng Published: 2026-09-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.22867v1 Diffusion models generate a sample by traversing a denoising trajectory, a sequence of stochastic noise-reduction steps that transforms pure noise into a draw from a target distribution. At deployment time, additional computation can improve sample quality without retraining: at each step, the sampler draws several candidate noise samples, scores the resulting predictions with a quality criterion called the verifier, and retains the best candidate at the cost of one network evaluation per candidate. This raises a resource allocation question: given a fixed budget of function evaluations, how should search effort be distributed across the steps of the denoising trajectory? We formulate this as a computational budget allocation problem. First, we show that, to leading order in the step size, the expected gain from evaluating $K$ candidates at a step factorizes into an endogenous, step-specific sensitivity parameter times a universal sample-size factor equal to the expected best of $K$ standard-normal draws. Second, for a fixed sensitivity profile, the optimal allocation solves a separable concave integer program with water-filling structure; at fixed total sensitivity, its advantage over uniform allocation increases with sensitivity dispersion in the majorization order. Third, we prove that when sensitivities vary across instances, no adaptive policy can avoid worst-case regret that grows linearly in the trajectory length, which motivates a design that anchors the allocation offline and adapts online only to recover instance-specific slack. We extend the analysis from independent random search to a broader family of local search operators, and instantiate it as an implementable algorithm. Experiments on three families of diffusion samplers show that the proposed allocation attains the quality of the uniform benchmark with 20 to 50 percent fewer function evaluations. --- ### Causilo Technical Report Authors: Minyong Cho, Minho Jeong, Dooho Lee, Jinmo Lee, Jaemin Yoo Published: 2026-09-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.22866v1 We introduce Causilo, a tabular foundation model (TFM) that combines frontier predictive performance with exceptionally fast inference. On TabArena, Causilo achieves 1785.4 Elo, at a median inference time of 0.10 seconds per 1K test samples. It outperforms TabPFN-3.5-Fast with 31.6% less inference time, placing it on the performance--efficiency Pareto frontier. Causilo follows TabICL's column-then-row architecture but introduces another row-refinement module before row compression. This module exchanges information among cell representations within each row after column encoding. The refined cells then visit the context set again through an additional column stage before being compressed into row embeddings. For inference efficiency, both row stages use cross-attention through a fixed number of summary tokens, keeping their attention cost linear in the number of features. Pretrained on approximately 36M synthetic tables, Causilo delivers strong benchmark results across TabArena, BeyondArena, and ScoringBench, achieving frontier-level performance with substantially faster inference. --- ### CurvFlow-DTA: dual-graph discrete Ricci curvature flow for drug--target affinity prediction Authors: Jicheng Ma, Yunyan Yang, Juan Zhao, Liang Zhao Published: 2026-09-19 Categories: cs.LG, math.DG arXiv: https://arxiv.org/abs/2609.22862v1 Graph neural networks are widely used for drug--target affinity (DTA) prediction, and discrete Ricci curvature has recently been used to characterize molecular graph geometry. Existing curvature-aware DTA approaches mainly use static curvature on the drug graph while representing proteins primarily with sequence-derived features. This leaves pair-adaptive use of graph geometry underexplored, which may limit adaptation to unseen entities in cold-start settings relevant to practical screening. We present CurvFlow-DTA, which replaces a single static curvature representation with weighted Forman curvature flow on both molecular and protein residue--residue contact graphs. A label-independent flow trajectory is precomputed for each entity, and a pair-conditioned selector determines the horizons read by a dual-branch Flow-GINE. A frozen ESM-2 supplies residue-level representations and contact scores used to construct the protein graph. Inference requires only SMILES strings and protein sequences, without a bound complex structure. On Davis and KIBA, CurvFlow-DTA improves on the protocol-matched Ricci-GraphDTA baseline in every warm and cold-start setting. Warm-split mean squared error (MSE) decreases by $19.9\%$ on Davis and $18.9\%$ on KIBA. Across the six cold-start comparisons, MSE decreases by $14.3$--$27.4\%$, with higher concordance index (CI) throughout. Within our compiled set of literature baselines, CurvFlow-DTA achieves the lowest MSE on both warm benchmarks and across four out of six cold-start evaluation settings. --- ### Image Frame Dynamic Object Segmentation and Ego Motion Estimation using Radar Image Fusion Authors: Astik Srivastava, Suhani Grover, Avinash Sharma, Madhava Krishna Published: 2026-09-19 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2609.22857v1 Dynamic object segmentation and ego-motion estimation are closely coupled problems in autonomous driving, as accurate ego-motion estimation typically requires static scene observations, while identifying static observations requires knowledge of the ego motion. We present Radar-Dot, a radar--RGB framework that exploits radar Doppler measurements to address this coupling. Radar returns are first used to estimate ego velocity through a linear Doppler constraint, with residual-based static/dynamic segmentation and robust estimation used to reduce the influence of moving objects. The estimated motion is then combined with metric depth and dense optical flow to identify image regions whose observed motion is inconsistent with the rigid scene motion. Experiments on 10 nuScenes scenes (part of nuscenes-mini) demonstrate that the resulting geometric pipeline achieves 20.24% dynamic IoU and 33.67% F1-score over 394 frame pairs, while radar-based static-point filtering improves ego-velocity estimation compared with using all radar returns. These results demonstrate the potential of radar as a modality for jointly improving ego-motion estimation and dynamic object segmentation. --- ### SmoLSTM: A Compact Vision-Language-Action Model with Recurrent Memory that Persists Authors: Jan-Gerrit Habekost, Parsa Mastouri Kashani, Connor Gäde, Matthias Kerzel, Philipp Allgeuer, Cornelius Weber, Stefan Wermter, Jae Hee Lee Published: 2026-09-19 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.22854v1 Vision-language-action models often predict actions from only the current observation, which can leave tasks involving object occlusion or visually identical objects ambiguous without episode history. The usual countermeasure, widening the observation window, turns the horizon into a hyperparameter and lets per-step cost grow with it. We instead capture the episode in a recurrent state. SmoLSTM couples a frozen 256M-parameter SmolVLM backbone to a matrix-memory LSTM control layer in which observation tokens and action queries are unified in a single causal stream that is never reset throughout the entire episode. Recurrent-state storage is therefore O(1) in episode length. A flow-matching action head predicts chunks of 10 end-effector pose deltas and gripper commands at each control step. Our single policy, trained jointly on 7,461 demonstrations across 140 tasks and evaluated on held-out initial states, performs best, reaching 85.1% subgoal coverage and 77.5% full-task success on LIBERO-Mem with 0.04B trainable parameters, surpassing both the benchmark's own object-centric baseline and recent memory-based approaches. Resetting the recurrent state at every control step reduces full-task success to 7.0%, showing that the trained policy relies on context carried between decisions. The same model achieves 79.6% average success on standard LIBERO. --- ### Discrete vs. Continuous: A Comprehensive Study of Unified Audio Understanding in LALMs Authors: Jing Peng, Zichao Nie, Zhisheng Zhang, Jingran Xie, Zhiyong Wu Published: 2026-09-19 Categories: cs.SD, cs.AI arXiv: https://arxiv.org/abs/2609.22851v2 Large Audio Language Models (LALMs) utilize either continuous features or discrete tokens, yet the optimal representation paradigm for general audio understanding remains debated. Existing benchmarks often focus on narrow domains or evaluate encoders outside LALM contexts. To address these gaps, we systematically evaluate continuous and discrete representations across speech, sound and music. Utilizing our UniARC framework with dual evaluation strategies across model scales from SmolLM2-135M to Llama-3-8B, we analyze the dynamic relationships of data volume, model capacity, and computational efficiency. Our results reveal the pivotal role of semantic constraints in tokenization for audio understanding and demonstrate that scaling backbones fail to compensate for information loss in audio representation, especially in data-limited tasks. These findings offer practical guidance for balancing semantic density, fidelity, and efficiency in future LALMs. --- ### Same Outcome, Different Readout: What Does a Steerable Valence Direction in LLMs Represent? Authors: Weihan Li, Xinlei Chen, Yuhan Song, Xiaofeng Lin, Tianshi Zheng Published: 2026-09-19 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.22850v3 Claims about what an internal direction in an LLM represents need evidential constraints beyond an observer's prior beliefs about the system. Decodability and successful activation steering do not, by themselves, establish which construct the direction tracks. This gap is especially consequential for welfare-relevant interpretations, where a proposed functional state must be distinguished from correlated features of the extraction contrast. We treat the question as one of construct validity and study a good-bad outcome direction in a maze task, using controlled interventions that separate the realised outcome from the informational history through which it became known. Across multiple LLM checkpoints, directions fitted on one explicit outcome encoding transfer well to another, indicating that the readout is not tied to surface form. When the same realised outcome is reached through announced and unannounced histories, however, transfer degrades substantially: even after both histories receive the same explicit outcome, the post-event readout remains strongly conditioned on the earlier announcement. In the base model, steering along the direction changes actions, yet removing it leaves the natural cue effect almost intact. In a matched maze-RL run, the post-RL direction becomes substantially more predictive of reference-MDP remaining return and the policy becomes more dependent on it at the tested sites, while the history dependence persists. These dissociations support a functional, value-related interpretation of the direction, but not its identification with a history-invariant scalar valence state. --- ### LINGO: Latent Initialization and Gradient Optimization for Sparse-view X-ray Novel View Synthesis and CT Reconstruction with 3D Gaussian Splatting Authors: Lifeng Xing, Dequan Jin, Kunpeng Bu, Peigeng He, Shihui Ying Published: 2026-09-19 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2609.22849v1 In novel view synthesis and Computed Tomography (CT) reconstruction with sparse-view X-ray imaging, insufficient angular coverage leads to structural ambiguity and accumulated noise. Integrating 3D Gaussian Splatting (3DGS) with X-ray absorption physics can achieve promising results, but it suffers from noisy initialization, positional insensitivity, and weak gradients in low-density regions. In this paper, we propose a unified Latent Initialization and Gradient Optimization (LINGO) framework to address these issues. LINGO combines latent mask-space initialization with dynamic gradient optimization to improve point cloud structural completeness while accelerating training. It constructs voxel-level 3D filters from X-ray masks to robustly suppress background noise and provide reliable geometric priors. By employing an adaptive voxel scaling strategy and dynamically scaling loss, LINGO can adjust spatial resolution and explicitly amplify gradients in low-density structures. To evaluate the quality of initialization, we introduce the Initialization Point Cloud Structural Deviation (IPSD) metric. Experiments on the X3D dataset indicate that for the novel view synthesis task, LINGO improves the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) by an average of 0.72 and 0.0039, respectively, over baselines under identical sparse-view settings, achieving comparable reconstruction quality within 5k steps to state-of-the-art models typically trained with 30k iterations. For the CT reconstruction task, LINGO also demonstrates consistent improvements, with average PSNR and SSIM gains of 0.36 and 0.0134. These results highlight LINGO's effectiveness in both accelerating training and enhancing reconstruction quality across different sparse-view imaging scenarios. --- ### Intelligent Degradation Monitoring in Lithium-ion Batteries via Discharge Incremental Capacity Feature Estimation Authors: Amir Madmolilvand, Farzaneh Abdollahi Published: 2026-09-19 Categories: eess.SY, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2609.22843v1 Accurate and timely detection of degradation in lithium-ion batteries is crucial to ensure safety, reliability, and longevity in high-demand applications such as electric vehicles and energy storage systems. Traditional incremental capacity (IC) analysis methods require low-current cycling for discharge measurements, limiting their practical use in real-time battery management. This paper proposes a novel neural network-based framework that predicts discharge IC features directly from charging signals, eliminating the need for low-current discharge. Trained on a comprehensive dataset of 53 battery cells cycled under diverse fast-charging protocols, the model demonstrates robust generalization ability, effectively estimating degradation indicators on unseen battery data. Among several architectures evaluated, the LSTM model provides the best balance of prediction accuracy and computational efficiency. The proposed approach enables real-time integration into Battery Management Systems (BMS), enhancing degradation monitoring without disrupting normal battery operation. This paper presents a study to bridge IC analysis with practical, fast-charging scenarios, marking a significant step towards intelligent and scalable battery health monitoring. --- ### A Horizon-Independent Regret Bound for Optimistic Hedge in General-Sum Games Authors: Junsoo Ha Published: 2026-09-19 Categories: cs.GT, cs.LG arXiv: https://arxiv.org/abs/2609.22839v1 Can simple learning rules keep their regret bounded in self-play? Recent work achieves constant regret bounds through modified regularization and higher-order prediction. Yet for Optimistic Hedge, arguably the most canonical method in games, the best known individual regret bound remains logarithmic. In this work, we prove that plain Optimistic Hedge with a constant step size can attain $O_{n,d}(1)$ individual regret in general-sum games with $n$ players and $d=(d_1,\ldots,d_n)$ actions, under expected loss-vector feedback. As a corollary, its time-averaged play enjoys an $O_{n,d}(1/T)$ coarse correlated equilibrium (CCE) gap. Our analysis represents Optimistic Hedge as a real-analytic recurrence on a compact space, which yields an exact finite-order difference relation that eliminates horizon dependence. Our proof hinges on nonconstructive Noetherianity argument of Frisch (1967), so the $(n,d)$-dependence remains implicit. --- ### Sensing to Intelligence: Principles for Neuromorphic Circuits and Systems Authors: Saptarshi Maiti, Chetan Singh Thakur Published: 2026-09-19 Categories: cs.NE arXiv: https://arxiv.org/abs/2609.22838v1 Neuromorphic engineering began with the idea that the physical behavior of a system could itself be used for computation, taking inspiration from the way nervous systems sense, adapt, and evolve in time. The field has since expanded far beyond its early analog circuits to include event-based sensors, spiking processors, emerging memory devices, mixed-signal systems, and large-scale neural accelerators. With this expansion, however, the meaning of neuromorphic has become increasingly broad. In this Perspective, we argue that neuromorphic engineering should be defined neither by resemblance to biological components nor by any particular device, signal representation, or substrate. Its potential lies in identifying computational principles in biological systems and translating them into the organization and dynamics of artificial machines. A system is truly neuromorphic when the invoked biological or physical principle plays a causal, design-relevant role in how information is represented, how state evolves, or what capability the complete system achieves. We develop this view across sensing, collective computation, memory, learning, and interaction. Adaptation, nonlinear dynamics, attractor structure, variability, and closed-loop action illustrate how computation can be embedded in a machine's evolving physical state. From these examples, we identify a set of design principles for neuromorphic systems. As neuromorphic engineering moves toward wider deployment, this perspective shifts the central question from how closely machines resemble nervous systems to what computation becomes possible when their principles are understood and deliberately translated into new machines. --- ### A Hybrid Attention Model Learning Unified Time-aware Patch Representation for Irregular Multivariate Time Series Forecasting Authors: Li Lin, Zhihao Lin, Qi Zhang, Kaiwen Xia, Shuai Wang, Jialin Qiao Published: 2026-09-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.22836v3 Time series foundation models (TSFMs) have recently delivered impressive zero-shot performance across diverse forecasting tasks. However, real-world decision-making frequently relies on \emph{irregular multivariate time series} (IMTS), where inconsistent inter-observation intervals and asynchronous sampling across variables coexist with informative missingness. Existing TSFMs handle such inputs either through imputation that injects spurious values or through index-based positional encodings that ignore continuous time. There is still a gap in the foundation model that follows the original IMTS patterns. In this paper, we propose a hybrid attention model that learns a unified time-aware patch representation for IMTS forecasting. We first design a \emph{time-aware patch encoding} that maps a variable number of intra-patch timestamps into a fixed-size embedding, producing a uniform format for irregular patches without resorting to imputation. We then introduce a \emph{time bias attention} mechanism that calibrates inter-patch temporal misalignment and asynchronous cross-channel dependencies as auxiliary attention offset. Finally, on top of a decoder-only Transformer backbone, we adopt a \emph{hybrid causal mask} that preserves a bidirectional full view over the historical context while keeping the forecast horizon strictly autoregressive. To support large-scale pretraining under irregular settings, we also curate VersaTSA, an archive of $30$B observations that retains the native sampling sparsity of its sources. Experiments on three IMTS benchmarks and a standard regular-MTS benchmark show that our model achieves state-of-the-art zero-shot performance on IMTS and remains competitive when transferred to regular forecasting. --- ### SatOV: Restoring Spatial Priors for Training-Free Open-Vocabulary Segmentation in Remote Sensing Imagery Authors: Changhao Zhao, Linglin Zeng, Hai Liu Published: 2026-09-19 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.22834v1 Open-vocabulary semantic segmentation (OVS) of remote sensing imagery is a challenging pixel-level task requiring strong generalization and adaptation to the spatial characteristics of remote sensing data. Although existing vision-language foundation models perform well in general domains, their image-level classification design weakens the spatial priors needed for high-resolution remote sensing segmentation: structural spatial relations are degraded during deep feature transformation, and fine-grained spatial details are lost during downsampling. To address these complementary deficiencies, we propose SatOV, a training-free framework for open-vocabulary remote sensing segmentation that restores spatial priors at two stages of the representation pipeline. Specifically, Residual QQ Attention (ResQQ) extracts Query-Key self-attention from an intermediate CLIP layer and fuses it with final-layer Query-Query attention via a residual combination, restoring structural spatial priors suppressed by the final-layer representation. Spatially Modulated Upsampling (SatUp) uses the original high-resolution RGB image as spatial guidance, combining spatial feature modulation with guided cross-attention to reconstruct pixel-level textures and boundaries. Extensive experiments on DOTA, UDD, LoveDA, and Vaihingen show that SatOV consistently improves training-free OVS and achieves competitive quantitative and qualitative results against state-of-the-art methods. These results validate the effectiveness of restoring spatial priors at both the representation and spatial-resolution stages for remote sensing open-vocabulary segmentation. --- ### Personalized Federated Reinforcement Learning via Model-Agnostic Meta-Learning: Convergence of Exact and Hessian-Free Meta-Policy Gradients Authors: Ali Beikmohammadi, Sarit Khirirat, Sindri Magnússon Published: 2026-09-19 Categories: cs.LG, cs.MA arXiv: https://arxiv.org/abs/2609.22833v1 We study personalized federated reinforcement learning, in which $n$ agents, each acting in its own Markov decision process, collaborate through a server to learn a shared MAML-style policy initialization that becomes effective for an individual agent once that agent adapts it with a single local policy-gradient step. We propose Per-FedAvg-PG, in which agents take $τ$ local stochastic meta-policy-gradient steps between communication rounds, and prove that it reaches an $\varepsilon$-approximate first-order stationary point of the personalized objective in $K=\mathcal O(\varepsilon^{-3/2})$ rounds with $τ=Θ(\varepsilon^{-1/2})$ local steps. The analysis rests on a structural feature of the reinforcement learning setting: under standard policy-class regularity, the per-agent objectives have uniformly bounded gradients and Hessians with explicit constants, so the bounded-gradient and bounded-heterogeneity conditions imposed by the supervised theory hold automatically and no separate heterogeneity assumption is needed. The exact meta-gradient requires the inner-loop policy Hessian, which our experiments identify as the practical bottleneck. We therefore analyze the Hessian-free variant, bound its bias, and exhibit fixed points at which the meta-gradient is nonzero and of order $α$, showing that the resulting stationarity floor is a property of the method rather than of the bound. Experiments on tabular and neural navigation confirm the predicted behavior and show transfer to unseen agents at an order of magnitude lower sample cost than independent training. Together these results identify the adaptation step size as a tunable personalization knob and the curvature estimate as the quantity that governs whether exact meta-gradients are affordable. --- ### From IceCube to IT-Sphere: A Hybrid Quantum-Classical GNN for Banking IT Root Cause Analysis Authors: Antonio Greco, Riccardo Paoletti, Roberto Cappuccio, Mario Onorato Published: 2026-09-19 Categories: quant-ph, cs.LG arXiv: https://arxiv.org/abs/2609.22822v1 We present Hybrid Quantum Root Cause Analysis (HQ-RCA), an industrially grounded workflow for root cause analysis in banking IT operations, built on a hybrid Quantum Graph Neural Network (QGNN): the classical backbone of DynEdge (the IceCube neutrino-reconstruction GNN, which we call standalone DynEdge), with its classification head replaced by a Variational Quantum Circuit (VQC). On 13 months of anonymised IT data (13k alarm clusters) from a major European bank, the hybrid QGNN matches standalone DynEdge -- the strongest classical baseline -- on $F_1$, while standalone DynEdge leads the ranking metrics. A readout-sensitivity and layout-robustness study, analysed via Dimensional Expressivity Analysis (DEA), shows that the effective parameter dimensionality (rank) of the quantum observable has no measurable correlation with $F_1$; we therefore keep the simplest readout $\langle Z_0\rangle$ (the Pauli-$Z$ expectation on the first qubit), which in the deployed layout is rank-1, collapsing optimisation to a 1-D problem solvable by a gradient-free grid scan. Execution on IBM Heron r2 (no error mitigation) shows this gradient-free readout is executable on NISQ hardware after threshold recalibration. --- ### Beyond Average Error through Oracle-Informed Stress Tests for Time-Series Forecasting Authors: Xu Lin, Runheng Zuo, Shengxuan Xu, Qitai Tan, Hongyu Lin, Xiao-Ping Zhang Published: 2026-09-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.22820v2 Average squared error cannot reveal whether forecasting performance degrades because the future becomes less predictable or because forecasts move farther from the conditional mean. We introduce paired, mechanism-controlled stress tests that decompose changes in expected squared error at each lead time into environmental risk and forecast-oracle distance, using an origin-conditioned predictive oracle unavailable to the evaluated models. Three end-to-end controls have known attribution. Specifically, the null, environmental-only, and information-gap controls verify that the pipeline assigns changes to the correct component. We then apply the benchmark to 24 deployable forecasters. Under frequent switching, 14 methods have higher realized MSE but lower oracle distance; under outlier-variance feedback, 19 have higher MSE but lower scale-standardized MSE. Short- and long-lead stress-response rankings have Spearman correlation 0.624, revealing substantial horizon-dependent reordering. We then study multivariate relation shifts. Across six models and three coupling severities, oracle distance accounts for only 0.7-3.9% of the decomposed expected-risk increase, and environmental-risk majority persists in an eight-channel system and a matched-difficulty audit of Ring, Block, and Hub relations. Finally, prespecified contrasts on independent data-generating process (DGP) realizations show that several visually compelling discovery profiles, including trend accumulation and the hypothesized switching reversal, do not replicate. The benchmark thus combines component-wise diagnosis with a held-out stability audit. It complements real-data out-of-distribution evaluation, which measures performance under realistic shifts when exact oracle attribution is unavailable. --- ### Counterfactual Tool Ranking under Utility, Cost, and Privilege Constraints Authors: Jiapeng Li Published: 2026-09-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.22819v1 Counterfactual tool evaluation must distinguish authority, historical support, and what a comparison actually estimates. We study these distinctions with eleven executable enterprise-inspired tools, exact-propensity logs, and real local Model Context Protocol transport. An initial 45-run synthetic study is retained, then challenged by 30 realized-return control runs and 15 experiments on 1,930 independently released Berkeley Function Calling Leaderboard (BFCL) tasks. Full-return direct regression reverses an initially favorable doubly robust (DR) evaluation result in the linear setting: mean absolute errors are 0.0139 for direct regression and 0.0272 for DR. Under a shifted environment, DR retains an advantage, with errors 0.0227 versus 0.0948. On function-name-group-disjoint BFCL-derived splits, direct and DR selectors obtain balanced accuracies of 81.85% and 79.83%. Two pinned local Qwen2.5 models are evaluated on the same 200 held-out tasks, exposing a strong failure to abstain under the fixed prompt. We further characterize policy differences under missing support: unsupported actions shared by two policies cancel, allowing point identification of an incremental change when neither absolute value is identifiable. A disagreement-preserving fallback achieves this property in all five support-gap runs, but conservative sampling bounds do not certify deployment improvement. The contribution is a falsifiable evaluation method and independent public evidence, not a new DR estimator, official BFCL leaderboard score, or production-agent safety claim. --- ### The Price of Safety: Benign-Case Utility and Token Overhead of Memory-Poisoning Defenses in LLM Agents Authors: Pritom Bhowmik Published: 2026-09-19 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.22818v1 Memory-poisoning defenses for LLM agents are typically evaluated by their ability to prevent attacks. However, the traffic they process is rarely adversarial. The cost of implementing a defense is paid with each interaction, while its benefits are only seen in a small percentage of cases. We developed a measurement setup that keeps the memory backend, retrieval process, and judge consistent across different conditions, changing only the defense itself. We test each condition three times across five conversations to distinguish the defense's real effects from noise inherent in the pipeline's runs, which remains significant even at temperature zero. Across three write-time defenses (input sanitization, provenance checking, and LLM-based anomaly detection) and one read-time defense (reranking), tested on entirely benign traffic, the write-time defenses show no utility cost we can resolve, with 95% confidence intervals spanning roughly +/-4.5 points and including zero. The reranker is different: it lowers core accuracy by 4.4 points (95% CI (-9.0,-0.05), bootstrap; McNemar p=0.064), a result that survives replication but sits at the edge of our resolution. Its clearer cost is mechanical rather than statistical. On conversations containing no attack, the reranker quarantines legitimate memories on 33.6% of adjudicated items, reaching as many as 106 false quarantines in a single conversation, at 2.7% token overhead. Stacking all four defenses does not compound this cost: the combined condition's accuracy loss is smaller, and its confidence interval includes zero, suggesting the write-time defenses may partly offset what the reranker discards. Where a defense intercepts the pipeline, not whether it uses an LLM, appears to determine its benign-case price. --- ### FIRM-WM: State-factorized factual-interventional recurrent modeling for reward-free visual planning Authors: Yilun Wu, Yunjian Zhang, Aobo Li, Mujiangshan Wang, Haitao Wu, Aqiang Zhang Published: 2026-09-19 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.22816v1 Reward-free latent world models can learn from offline videos and solve new image--goal tasks by optimizing actions through predicted latent futures. This setting places two demands on the planning state: its coordinates must be comparable with a goal image. Moreover, its dynamics must retain velocity, motion trend, contact, and other history--dependent information beyond those goal coordinates. Offline training creates a second mismatch: each recorded trajectory reveals one factual future, whereas a sampling--based planner compares many actions that were not taken from the same state. We introduce FIRM-WM (Factual--Interventional Recurrent World Model), a compact pixel world model designed around these two gaps. Its recurrent state separates a typed, goal--comparable configuration from a 128-dimensional dynamic fiber used for prediction but excluded from the terminal goal cost. Broad factual trajectories provide state coverage, while common--reset intervention branches provide observed outcomes for alternative action sequences. Before executing each branch, we reset the environment and restore the same recorded values exposed by the environment's state--setting interface. Under matched CEM planning and three independent full-pipeline seeds, FIRM-WM reaches 99.0$\pm$1.0% on TwoRoom, 92.7$\pm$2.1% on Reacher, and 88.0$\pm$3.0% on OGBench-Cube, compared with 89.0%, 88.0%, and 70.0% for LeWM. The deployed model uses 2.98--3.42M parameters and records 2.13--11.60$\times$ lower planning time on these tasks. --- ### Commonsense-Grounded Path Planning from Abstract Instructions Authors: Masafumi Endo, Kohei Honda, Ryo Yonetani Published: 2026-09-19 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.22813v1 We present \emph{commonsense ranked search} (CoRS), a novel path planner that turns an abstract instruction into a route that follows commonsense. While existing methods respect the considerations written down in advance, a robot working among people must follow those left unstated too, as with a wet floor that a worker avoids without being told. CoRS leverages large language models (LLMs) and vision-language models (VLMs) as commonsense knowledge to reason about these latent considerations in its planning. Given an abstract instruction (\emph{e.g.}, ``move carefully'') and visual observations of each region in the environment, CoRS derives a consideration for each region, as in ``this wet floor is slippery and worth a detour.'' It then compares the considerations between regions to see which of the two the robot should avoid more, as in ``the crowd is worse than the wet floor.'' These judgments sort the regions into a commonsense ranking, whose costs drive a conventional search that always returns a valid route. We build a benchmark for planning under latent considerations, with three environments, 1350 problems, and five instructions at three levels of abstraction. Experiments show that CoRS discovers the unstated considerations and goes around the ones worth a detour while crossing the rest, a behavior that recent LLM-based planners do not achieve. --- ### The shape of quark flavors Authors: Shinsuke Kawai, Nobuchika Okada Published: 2026-09-19 Categories: hep-ph, cs.LG, hep-th arXiv: https://arxiv.org/abs/2609.22812v1 We construct the Yukawa couplings of the quark sector as overlap integrals of Gaussian wave functions in extra spatial dimensions. Assuming that the wave function of each chiral fermion is localised at a point in the extra dimensions, while that of the Higgs doublet is flat, the Yukawa matrix elements are determined by the relative positions of the three left-handed quark doublets, $q_1$, $q_2$, $q_3$, and the six right-handed quark singlets, $u_1$, $u_2$, $u_3$, $d_1$, $d_2$, $d_3$. We employ numerical optimisation techniques from machine learning to identify field configurations that satisfy all current experimental constraints on the quark Yukawa sector. In this framework, the large hierarchy of quark masses arises naturally, since the overlap integrals of Gaussian wave functions are exponentially sensitive to ${\mathcal O}(1)$ separations in the extra dimensions. --- ### C$^{2}$-INR: Customized Convolutional Implicit Neural Representation Authors: Jinglei Shi, Xinran Chang, Jiaqi Cui, Yingjie Xia, Zhaolin Xiao, Chongyi Li Published: 2026-09-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.22807v1 Implicit Neural Representation (INR) leverages neural networks to represent discrete signals such as images as continuous ones, where the network weights serve as a compact form of the signal itself. Most existing INR methods adopt Multi-Layer Perceptrons (MLPs) as their backbone. Since these models render each pixel independently, they inherently fail to exploit the spatial correlations that exist between neighboring pixels. In contrast,convolutional INRs can process pixels in parallel while inherently accounting for inter-pixel dependencies, making them a more natural fit for representing images. Nevertheless, convolutional INRs remain relatively underexplored, and the majority of them rely on fixed architectural settings, leaving little room for image-specific adaptation. In this paper, we investigate network customization for convolutional INRs. We replace conventional filters with irregular directional kernels, whose allocation is guided by the directional energy in the image spectrum, i.e., directions exhibiting stronger energy are assigned a larger number of kernels, enabling content-tailored convolution settings. These kernels are further reformulated via an orthogonal basis to achieve a superior sparse representation. Moreover, we introduce an annealed Gumbel-Softmax-based mechanism for kernel-level activation function selection, which gives the most suitable activation function for each convolution kernel. Extensive experiments demonstrate that our method, namely C$^{2}$-INR, achieves superior performance against state-of-the-art approaches under comparable parameter budgets across a wide range of image processing tasks, including representation, inpainting, and super-resolution. --- ### To Consolidate or not to Consolidate? Evaluating the Impact of Consolidation in Multi-Reference Training using Peer Reviews Authors: Maitreya Prafulla Chitale, Ketaki Mangesh Shetye, Yash More, Harshit Gupta, Manav Chaudhary, Manish Shrivastava, Vasudeva Varma Published: 2026-09-19 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.22805v1 Natural language generation (NLG) tasks span the spectrum of conditional entropy, ranging from highly constrained machine translation to open-ended dialogue generation. Structured tasks like automated peer-review generation occupy the intermediate region, where a single input admits multiple valid, overlapping outputs. In this work, we demonstrate that traditional single- and multi-reference training paradigms are suboptimal for these intermediary tasks. We provide empirical evidence that consolidating diverse references into a unified training signal is crucial for developing effective systems. To facilitate this, we introduce MERC-36K, a large-scale corpus of over 36,000 papers paired with original and consolidated peer reviews. Using this dataset, we train specific architectures to isolate the impact of different reference paradigms and benchmark against existing state-of-the-art systems. Through extensive automatic and human evaluation, we demonstrate that models trained on consolidated references significantly outperform those trained on unconsolidated references. Dataset and code will be released upon acceptance. --- ### AlexandriaX 2026: The First Shared Task on Dialectal Arabic Machine Translation Authors: Abdellah El Mekki, AbdelRahim A. Elmadany, Samar M. Magdy, Saad Ezzini, Mo El-Haj, Mustafa Jarrar, Zaid Alyafeai, Bernard Ghanem, Muhammad Abdul-Mageed Published: 2026-09-19 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.22796v1 Dialectal Arabic machine translation (MT) remains challenging despite recent progress in Arabic language technologies, particularly because effective translation requires modeling not only semantic content but also dialectal variation, conversational context, speaker and addressee characteristics, and sociolinguistic appropriateness. Moreover, conventional MT metrics provide limited insight into the linguistic errors produced by dialectal systems. We present the AlexandriaX 2026 Shared Task on Dialectal Arabic MT, which addresses these challenges through three complementary subtasks: (1) context-aware English-to-Dialectal Arabic dialogue translation across 13 Arabic varieties, (2) cross-dialect Arabic translation in the financial domain covering six Arabic dialects, and (3) span-level MT error detection and classification using linguistically motivated error categories across five Arabic varieties. The shared task attracted 38 registrations for Subtask 1, 33 for Subtask 2, and 35 for Subtask 3. Twelve unique teams submitted their system description papers, all of which we accepted for publication. The best system on Subtask 1 achieved 30.42 spBLEU in the constrained track and 33.49 spBLEU in the unconstrained track. On Subtask 2, the top system obtained 28.40 spBLEU. On Subtask 3, the best system achieved an overall score of 49.82, outperforming the 24.29 baseline. Taken together, the results of the leading systems across the three subtasks highlight the benefits of explicit dialect modeling, context-aware generation, retrieval and reranking, and specialized approaches to interpretable MT error analysis. All the resources of AlexandriaX 2026 shared task are publicly available, including data, baselines, and evaluation code on our project page: https://alexandriax.dlnlp.ai. --- ### Diagnose, Then Repair: A Two-Stage MQM-Guided Post-Editing Framework for Domain-Specific Machine Translation Authors: Ji Hun Wang, Siyu Wu Published: 2026-09-19 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.22793v1 LLM-based machine translation evaluation can closely match human judgments, but in practice it remains largely diagnostic, with the signals rarely translating into direct quality improvements under real production constraints. We propose a two-stage, evaluator-guided automatic post-editing framework that turns MQM-style evaluation into targeted repairs: a retrieval-augmented LLM evaluator outputs structured, span-level MQM diagnoses under an explicit edit contract, and a separate LLM post-editor applies minimal edits restricted to those diagnoses. This separation improves controllability and reduces paraphrastic drift compared to one-stage "judge-and-refine" baselines. In a systematic study involving seven LLMs spanning three model providers and seven languages, our best configuration consistently improves both COMET-22 and COMETKiwi scores over one-stage post-edit methods, while the evaluator's error spans and severities show strong agreement with human MQM annotations and human editor preferences. --- ### SelfOp: An Optimization Algorithm for Self-Improving Security Agents Authors: Saad Ullah, Yigitcan Kaya, Christopher Kruegel, Giovanni Vigna, Gianluca Stringhini Published: 2026-09-19 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.22792v1 LLM agents are increasingly used for security tasks: vulnerability discovery, exploit reproduction, and patch generation. Improving them at the model level demands expert demonstrations or computable rewards, which security tasks rarely offer: traces are costly, failures hard to diagnose, rewards sparse, and non-computable. Efforts thus shift to the harness and context, but manual tuning needs task-specific expertise and scales poorly, while automated methods rely on scarce ground truth, stronger optimizer models, or unguided propose-and-evaluate loops that reduce to costly trial and error. We introduce SelfOp, an algorithm that automatically improves a frozen security agent's task context (instructions, skills, and reference documents), without modifying its execution harness and model weights. SelfOp casts context optimization as chain-rule-inspired textual gradient descent: from a single instance's outcome, it propagates error signals backward through the evaluator, the agent's trajectory, and the context artifacts that shaped its behavior, yielding per-instance textual gradients. Gradients are accumulated across instances by clustering, ranking, and filtering, and committed only under cross-instance consensus. A convergence detector monitors the gradient signal itself and stops once the context has absorbed the generalizable information in the training data, without held-out validation data. We evaluate SelfOp on CyberGym, a benchmark of real-world vulnerability reproduction tasks. With fewer than 200 training examples, SelfOp yields a 17-point self-improvement for GPT-5.4-mini (with Codex), enough to surpass the frontier GPT-5.4 baseline by 6 points, and an 18.5-point self-improvement for GPT-5.4 itself. The optimized skills also transfer across models, highlighting that SelfOp-optimized skills learn generalizable task knowledge not model-specific patterns. --- ### From Research Frontier to Laboratory Bench: Design of a Four-Tier Experimental Teaching System for Multimodal Medical Image Intelligent Diagnosis Authors: Dongjing Shan, Yamei Luo, Jin Li, Yong Luo Published: 2026-09-19 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.22790v1 Undergraduate programmes in intelligent medical engineering are expanding, yet laboratory curricula lag behind the multimodal, long-tailed, and distributionally shifting realities of clinical AI. This design paper presents an advanced experimental teaching system that translates an ongoing multimodal deep learning research project on endometrial carcinoma into a structured undergraduate lab sequence. We identify three educational gaps (modality, authenticity, and deployment) and derive four pedagogical principles from constructive alignment, experiential learning, the research teaching nexus, and the CDIO framework. The curriculum comprises four progressive tiers plus an engineering layer, with 32 laboratory units over 64 contact hours, delivered via a custom virtual clinical workstation using de-identified multi-institutional data. Each tier maps to a specific technical bottleneck, prerequisite coursework, and criterion-referenced deliverables. Data governance, safety, and assessment protocols are specified. Learning outcome data will be collected across two implementation cycles. --- ### PixelART: Image-to-Layer Decomposition without Latents or Text-to-Image Pretraining Authors: Zelin Jia, Zhao Zhang, Zhicong Tang, Yuhui Yuan, Shixia Liu Published: 2026-09-19 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.22789v1 Image-to-layer decomposition converts a flattened image into editable RGBA layers, enabling element-level editing in design workflows. Existing diffusion-based systems typically adapt large pretrained text-to-image (T2I) models and introduce RGBA autoencoders or variable-layer architectural modules. We revisit this design choice and ask whether layer decomposition truly requires these heavyweight components. We introduce PixelART, a pixel-space rectified-flow Transformer trained from scratch for image-to-layer (I2L) decomposition. PixelART directly denoises regional RGBA pixel patches using a single-stream multi-modal diffusion Transformer, avoiding RGBA-VAEs, pretrained T2I backbones, and layer-specific decoders. We identify a key property of the task: high-noise timesteps determine layer assignment and coarse layer organization, while low-noise timesteps mainly refine color, alpha, texture, and boundaries. Based on this observation, we propose a terminal-boosted timestep sampling strategy to increase training coverage in the high-noise layer assignment regime. Trained on 4M multi-layer design templates, PixelART achieves state-of-the-art layer decomposition and composite reconstruction on Design-Multi-Layer-Bench and LICA with over 80% fewer parameters, 98% lower latency, and 85% lower memory than the recent Qwen-Image-Layered model. Ablation experiments show that pixel-space $\mathbf{x}$-prediction, terminal-boosted timestep sampling, and data/model scaling are critical, while T2I pretraining brings marginal benefits to the I2L task. --- ### Human-Level Accuracy, Non-Human Strategies: Revealing Model-Human Divergence in Video Physical Reasoning Authors: Fanhong Li, Shurui Zheng, Zi Yin, Junbo Cui, Lei Ji, Jia Liu Published: 2026-09-19 Categories: cs.CV, stat.AP arXiv: https://arxiv.org/abs/2609.22788v1 Video foundation models now reach human-level accuracy on physical-reasoning benchmarks, yet such tasks require predicting unobserved physical outcomes. Do these models perform human-like forward simulation, or do they exploit statistical regularities in visible scenes? Accuracy alone cannot distinguish these strategies. We introduce a distributional evaluation framework that treats model seeds and human raters as populations, enabling comparison of consensus, uncertainty, and strategy. On the Physion benchmark, we evaluate three ViT-L architectures (V-JEPA2, VideoMAEv2, DINOv2). V-JEPA2 narrows the accuracy gap to ~1 percentage point (73.2% vs. 74.2%), yet model-human disagreement reaches 26.4%, far exceeding human-human disagreement (4.8%), with substantially lower agreement (kappa ~ 0.48 vs. 0.91). The divergence follows forward-simulation demands: models outperform humans on geometric reasoning (linking, +11.8 pp) but underperform on gravitational dynamics (rolling, -11.8 pp) and causal chains (dominoes, -10.5 pp). Strategy fingerprinting confirms all three architectures share non-human strategies while none aligns with humans. Attribution analysis suggests that unobservable outcome features, rather than visible scene properties, predict this divergence, consistent with models relying more on scene-level statistical regularities than on explicit forward simulation, a systematic divergence that accuracy alone cannot reveal. Code is available at https://github.com/fanhong-li/model-human-divergence. ---