# Research Papers — 2026-09 · bundle page 92 of 266 > Full-text markdown of papers #4551–4600 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/92 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/91.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/93.md) ### BanglaKontho: Closing the Long-Form Gap in Bangla Text-to-Speech Authors: Mizbaul Haque Maruf Published: 2026-09-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.29146v1 Bangla, the seventh most spoken language in the world, remains under-resourced for neural text-to-speech. Public Bangla speech corpora are dominated by short read-prompt utterances collected for speech recognition, leaving long-form prosody and consistent single-speaker narration uncovered. We present BanglaKontho, a single-speaker Bangla TTS corpus of 20 hours derived from professional audiobook recordings: 7,050 segmented utterances with verified transcripts at 24 kHz. We also release a reusable Bangla text normalizer covering Bangladeshi-style digit grouping, currency and date expressions, Danda punctuation and Unicode normalization, together with the full preprocessing pipeline. An MB-iSTFT-VITS baseline trained from scratch reaches 9.5% WER and 4.46 naturalness MOS, against 16.0% and 3.16 for the same architecture retrained on the 12-hour IndicTTS-Bn corpus. The corpus is released openly under CC BY-NC 4.0. --- ### Claim-Gated Source-Risk Auditing for Generative Search Authors: Kainan Zhou, Chuhong Xu, Gangzhen Qian, Zhaoyi Li Published: 2026-09-24 Categories: cs.AI, cs.CC, cs.CE, cs.ET, cs.IR arXiv: https://arxiv.org/abs/2609.29145v1 A generative search answer can cite a supported passage yet omit a source relationship that changes its interpretation. We specify a claim-gated audit of the query-source-answer tuple. An omission is resolved only when relationship evidence, answer adoption, materiality, and disclosure are all observed; incomplete evidence remains unresolved rather than being treated as independence. The specification separates this endpoint from citation support and review priority, and binds decisions to versioned evidence spans. A reference checker makes the record contract executable. On an exhaustive synthetic suite, it reproduces all 81 three-state predicate combinations and rejects 192 deliberately malformed records. Common-guard baselines and predicate ablations isolate endpoint logic from missing-evidence handling, while controlled transitions check support separation and evidence removal. These are finite contract-conformance results, not detector accuracy or evidence of improved user outcomes. We define the independent annotation, held-out evaluation, and paired utility tests still required to establish semantic validity and deployment benefit. --- ### Scope Before You Persist: Preventing Cross-Family Interference in Agent Memory Authors: Yezhou Cheng, Runjia Du, Zeming Liu, Qibai Chen, Hang Lyu, Yankai Zeng, Yilan Wei, Bojun Lin Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.29144v1 Persistent memory lets language-model agents improve prompts and skills without updating model weights. We show that matching retrieval scope to certification scope enables these edits to support reliable repeated adaptation across recurring task families. We study frozen-model agents on ProcStream-RSI, a 12-round code-repair stream, using Orthogonal Regression Control (ORC), an execution-grounded gate for persistent skill edits. In an intervention that holds proposals and gate decisions fixed, retrieving each accepted skill only for its originating family raises mean hidden trajectory utility from 0.713 under global memory to 0.816 and changes harmful deployments from six of eight to none. In 27 paired randomized-order streams, Scoped-ORC improves mean trajectory utility by 0.063 (0.037, 0.094) over Global-ORC, accepts 63 rather than 12 updates, and produces multiple accepted updates in 19/27 streams, with 0/63 harmful acceptances. The global control reaches 0.713, below the static agent's 0.775, because locally valid edits can interfere with unrelated families. These results establish scope matching as a complementary control for persistent agent memory: certification determines whether an edit is supported, while retrieval scope determines where that evidence authorizes its use. --- ### AI-Moderated Interviews for Market Research and Digital Twins Calibration Authors: Yuting Deng, Jingxuan Liu, Olivier Toubia, Naman Jain Published: 2026-09-24 Categories: cs.CY, cs.AI, cs.HC, cs.MA arXiv: https://arxiv.org/abs/2609.29143v1 AI-moderated interviews are emerging as a scalable market-research method for generating consumer insights and building consumer "digital twins." Yet it remains unclear whether they match human-moderated interviews or improve on simpler, static data collection methods. In a pre-registered, between-subjects study (N = 317) with three industry partners, we compare AI-moderated (N = 139), human-moderated (N = 24), and static interviews (N = 154). AI moderation matches human moderation in depth, covers more themes, and, holding budget constant, recovers significantly more customer needs than human moderation or static interviews. However, participants sound more emotionally engaged when speaking to a live human. We then create digital twins using interview data and evaluate each twin against the participant's own held-out responses to six real-world marketing stimuli. We find that digital twins created from AI-moderated interviews predict consumer responses better than demographics-only personas. However, the additional richness from AI moderation does not translate into better quantitative predictions compared to static interviews. By analyzing open-ended thoughts generated from humans versus their twins, we find that prediction errors are connected both to differences in (self-reported) thinking styles between twins and humans, and to gaps between training and validation data (i.e., asking questions that are too far out of distribution). --- ### Not Every Token Is Worth Distilling: Selective Supervision for Direct-OPD Authors: Yibo Zhao, Zixuan Yang, Yunshi Lan, Xiang Li Published: 2026-09-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.29142v2 Direct On-Policy Distillation (Direct-OPD) transfers reinforcement-learning-induced policy improvements from a small model to a larger student by using the token-level log-ratio between post-RL and pre-RL checkpoints as dense supervision on the student's own rollouts. This transfer rewards the policy shift at every state, yet the log-ratio measures only relative change: it can stay fixed even as the probability mass that both checkpoints assign to the student's candidate tokens vanishes. Through an exact construction, we show that the Direct-OPD reward and its update can remain unchanged while the Jensen-Shannon divergence (JSD) and both KL directions between the checkpoints vanish with this mass, and we note that a small JSD bounds how much the teacher's behavior changed. Motivated by this analysis, we propose Selective Supervision for Direct-OPD (S$^2$D-OPD), which ranks student-sampled states by their teacher-reference JSD and masks Direct-OPD supervision at low-divergence states, retaining only the top 10% of states per response. Across two teacher pairs and four student models ranging from 1.7B to 8B parameters, S$^2$D-OPD improves held-out accuracy over dense Direct-OPD on AIME and HMMT benchmarks in seven of eight settings and matches it in the eighth, without extra forward passes. --- ### Sharp Limits for Honest Uncertainty in Hard-Budget Repeated Evaluation Authors: Yezhou Cheng, Runjia Du, Zeming Liu, Qibai Chen, Hang Lyu, Yilan Wei, Yankai Zeng, Bojun Lin Published: 2026-09-24 Categories: cs.AI, cs.IT arXiv: https://arxiv.org/abs/2609.29140v1 Repeated evaluation can estimate a benchmark score accurately while still requiring replication to certify narrow uncertainty. We characterize that requirement on a fixed grid of $M$ tasks with $L$ binary paths per task under the hard budget $(M+t)K$, where each path costs at most $K$ responses or episodes. For fixed $L \ge 3$ and $0 < α\le 1/12$, the optimal expected width on the worst pure cohort is $Θ_{α,L}((M(t+1))^{-1/2})$ when every task is observed and $Θ_{α,L}((M(t+\sqrt{M}))^{-1/2})$ when omission is allowed. The lower bounds cover adaptive hard-budget policies, and fixed random-subset designs attain both rates through disagreement certificates. A joint mean/disagreement interval turns the task-covering law into practical finite-budget inference. In an equal-budget LiveCodeBench replay with 16 models, 880 tasks, and five outputs per task, the task-covering design reduces median point-estimation MSE by 87.0\% relative to pooled uniform sampling, while the Joint certificate produces narrower confidence intervals in 15/16 panels and reduces median interval width by 30.6\%. Finite-regime analyses identify task coverage as the effective choice at the evaluated scale and characterize how cohort size and within-task agreement determine the useful operating region. Together, the sharp laws and fixed-budget evidence make replication and task coverage explicit design variables for information-efficient repeated evaluation. --- ### Tag-Aware Structured Text Translation: Towards a Systematic Understanding Authors: Zhanglin Wu, Hengchao Shang, Daimeng Wei, Jiaxin Guo, Zongyao Li, Tengfei Song, Ning Xie, Weidong Zhang Published: 2026-09-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.29131v1 Internet texts are replete with format tags that carry structural, semantic, and functional meaning. Current large language model (LLM)-based translation systems struggle to balance translation fluency with tag fidelity when processing tagged text. We argue that resolving this tension requires a systematic approach at three interconnected levels: data synthesis, capability building, and multi-objective alignment. At the data level, we identify and formalize a fundamental trade-off between structural tag diversity and translation naturalness in synthetic data generation; existing methods optimize for one at the expense of the other. We propose a hybrid synthesis strategy (Hy-LST) combining LLM-based synthesis tag method and Two-Stage LLM-based synthesis tag method to produce both diverse and natural tagged data. At the capability level, we decompose tag-aware translation into four sub-tasks of increasing difficulty in a multi-task supervised fine-tuning framework, enabling targeted capability acquisition and knowledge transfer. At the alignment level, we design three complementary reward functions under a group relative policy optimization framework, each targeting a distinct objective (fluency, tag fidelity, and tag-scoped translation quality), and show that joint optimization consistently outperforms single-reward alternatives. Experiments on six language directions (en2zh, en2ja, en2de, en2fr, en2ru, de2fr) demonstrate that each level contributes measurable improvements, and the complete system significantly outperforms existing methods. Qualitative analysis reveals specific error patterns and their mitigation after training with our method. --- ### FoCal: Frequency-Oriented Cross-Modal Interaction and Spectral Calibration for Aerial Visible-Infrared Object Detection Authors: Ben Liang, Chao Sui, Junqi Bai, Yuan Liu, Chunlai Li, Xiubao Sui, Qian Chen Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.29125v1 In aerial RGB--IR object detection, effectively exploiting complementary information across modalities is critical for robust perception under complex illumination and environmental conditions. Existing multimodal detectors mainly focus on spatial-domain interaction or frequency-specific feature enhancement, while the cross-modal interaction patterns of different frequency components remain insufficiently explored. Moreover, spectral discrepancy itself may contain both useful complementary cues and unreliable modality-specific responses, making indiscriminate frequency fusion suboptimal. To address these issues, we propose FoCal, a frequency-oriented framework for aerial RGB--IR object detection. First, a Frequency-Aware Dual-Domain Calibration (FADC) module is developed to explicitly model frequency-dependent cross-modal interaction. Low-frequency components are collaboratively consolidated into a shared structural consensus, whereas high-frequency components preserve modality-specific information through selective cross-modal exchange. The resulting frequency-aware cues are further transferred to the original feature domain to regulate cross-modal calibration. Second, we introduce a Discrepancy-Guided Spectral Modulation (DGSM) module, which characterizes cross-modal spectral imbalance using confidence-weighted relative amplitude discrepancy and transforms it into a bounded signed gate for adaptive enhancement, preservation, or attenuation of the joint multimodal spectrum. Extensive experiments on DroneVehicle, ESCVehicle, and ATR-UMOD demonstrate the effectiveness of FoCal, yielding $\mathrm{mAP}_{50}$ values of 83.5\%, 54.8\%, and 64.6\%, respectively. Meanwhile, with only 3.0M parameters, FoCal achieves 113.6 FPS while preserving leading detection accuracy, highlighting a favorable accuracy--efficiency trade-off. Code is available at {https://github.com/universeliang/FoCal. --- ### Accent Analogy Guidance: More Speaker Similarity at Equal Accent in Cross-Lingual Voice Cloning Authors: Yoomee Cho, Jisun Lee Published: 2026-09-24 Categories: cs.SD, cs.CL, eess.AS arXiv: https://arxiv.org/abs/2609.29123v1 In cross-lingual zero-shot text-to-speech, the accent of the reference leaks into the target speech. We propose accent analogy guidance (AAG), a training-free sampler term that subtracts an accent direction estimated from the model's own predictions for one synthetic voice rendered in both languages, so the voice cancels and only the accent remains. By a blind LLM accent judge on real dubbing data, reweighting classifier-free guidance between reference and text, and its variants, stay near one identity-accent trade-off curve; we score a method by its speaker similarity above that curve at equal accent ($Δ$SIM). Across four open TTS models AAG lies above the curve: on OmniVoice $Δ$SIM is +0.11 to +0.27 on three test sets (accent 3.51 to 4.28 on a 1-5 scale at speaker similarity 0.29, where reweighting keeps 0.02); MaskGCT and CosyVoice 2 also lie above their curves, and on F5-TTS it is more native than any reweighting setting. An LLM-free language-ID measure and a twelve-listener panel agree. A premise test and the reach of a model's own curve indicate in advance whether and roughly how much AAG can gain, predicting the one model where it gains nothing (X-Voice). --- ### Less is More: Encoder-only Audio-Visual Segmentation Authors: Ilpo Viertola, Vladimir Iashin, Sophie Tötterström, Esa Rahtu Published: 2026-09-24 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.29121v1 Audio-Visual Semantic Segmentation (AVSS) aims to identify, segment, and classify sound-emitting objects in video frames. Previous Transformer-based AVSS approaches largely inherit design principles from image segmentation models. Recent studies show that these image segmentation models contain redundant components that contribute little to the segmentation performance. Following this insight, we propose Encoder-only Audio-Visual Segmentation (EASE). EASE runs at up to 365 FPS, 3x faster than prior State-of-the-Art (SotA) AVS models at comparable accuracy, and trains in under 11 GPU-hours. Furthermore, we achieve SotA AVSS performance across different backbones and input resolutions. Our results demonstrate that AVSS can be both simpler and faster, providing a scalable foundation for future research and real-time applications. Code, model weights, and samples are available at https://ease-avs.notion.site --- ### UpDown-SC: Gravity-Canonicalized Dual-Envelope Scan Context for Indoor LiDAR Place Recognition Authors: Jie Xu, Yongxin Yang, Ziyi Jin, Kangjin Yu, Hongjun Huang, Chao Han, Zhongpu Xia Published: 2026-09-24 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2609.29118v1 LiDAR place recognition is a key front end for loop closure and global relocalization, yet indoor retrieval remains difficult when attitude or sensor mounting height changes between mapping and query sessions. Scan Context stores the maximum height in each polar cell; indoors, broad ceilings can suppress the lower and mid-level geometry that distinguishes adjacent rooms and corridors. We present UpDown-SC, a training-free polar descriptor that first canonicalizes gravity and then represents two complementary surfaces: the upper envelope of lower/middle structures and the lower envelope of overhead structures. Their physical split is estimated once from a cell-balanced map height distribution and reused by every query. A mask-aware, non-uniform two-channel distance retains discriminative lower-level evidence while limiting sensitivity to its cross-session variation, without treating unobserved cells as zero-height measurements. Conventional Scan Context shortlisting and circular yaw alignment are retained, so retrieved hypotheses directly initialize geometric verification. Experiments across repeated indoor sessions, mounting-height changes, mixed outdoor-to-indoor trajectories, and an outdoor transfer sequence show more reliable first-choice retrieval on the indoor and mounting-height-varied sessions. A paired test finds a significant gain over Scan Context on the in-house sessions. UpDown-SC also gives the best or second-best F1max and AUPR under threshold-based acceptance while retaining a lightweight CPU front end. Continuous replay confirms that the retrieved hypotheses support metric prior-map localization. Code and evaluation artifacts: https://github.com/jiejie567/updown-sc. --- ### A Concentration Bound for Two-Timescale Actor-Critic Algorithm Authors: Prashansa Panda, Shalabh Bhatnagar Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.29117v1 Significant research effort has been directed in recent years towards establishing both asymptotic and non-asymptotic convergence guarantees for two-timescale actor--critic algorithms, where the actor recursion is run on a slower timescale than the critic recursion. This work derives a uniform all-time concentration bound for the actor--critic algorithm with function approximation in the long-run average-reward setting. This bound helps us analyze the behavior of the actor parameter with high probability. We show that, after some finite time, the actor parameter enters a safe region and remains within it thereafter with high probability. Specifically, with probability at least $1-ε_1-ε_2$, the actor error $\Vert θ_k-θ^{*}\Vert$ is $O\left(\frac{n_0^{3/4}}{k}\frac{1}{\sqrt{ε_2}}+\left(\frac{1}{n_0}\right)^{1/4}\log^{1/4}\left(\frac{1}{ε_1}\right)+\left(\frac{1}{n_0}\right)^{1/4}\right)$ for all $k\geq n_0$ and sufficiently large $n_0$. We also present experimental results demonstrating that the aforementioned actor error diminishes with the number of actor-parameter updates. --- ### Spectral Amplitude Purification in Distribution Matching for Diffusion Distillation Authors: Zhenyu Zhou, Can Wang, Chun Chen, Zeyu Zheng, Defang Chen Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.29116v1 Distribution Matching Distillation (DMD) enables high-quality diffusion sampling in only a few steps, but its optimization dynamics remain dominated by coarse, low-frequency signals, delaying the recovery of fine-grained details. We identify a pronounced concentration of spectral amplitudes at low frequencies in the DMD directional error, where dominant low-frequency components overwhelm weaker mid- and high-frequency signals. To address this issue, we propose Spectral Amplitude Purification for Distribution Matching Distillation (SAP-DMD), a plug-and-play approach that adaptively modulates the amplitude spectrum of the DMD directional field. By suppressing the dominant tail of the amplitude spectrum, SAP-DMD reduces low-frequency dominance and promotes more effective recovery of fine structures and textures. Experiments on PixArt-$α$, SD3, and SD3.5 demonstrate that SAP-DMD accelerates training convergence and improves generation quality under both 2-step and 4-step sampling. --- ### CounterRoute: Self-Routed Reasoning via Hierarchical Counterfactual Credit Assignment Authors: Ruochen Jiao, Besnik Fetahu, Zhenyu Shi, Priyanka Nigam Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.29109v2 Reasoning-capable language models often produce long chains of thought when direct answers suffice, wasting inference compute. Dual-mode models offer both thinking and direct-answer modes, but typically leave mode selection to users. Automating this selection while improving responses under both modes is challenging because routing targets evolve with the policy, strong initial mode preferences can destabilize exploration, and sequence-level objectives entangle routing with response learning. We introduce CounterRoute, an online reinforcement-learning framework that jointly learns routing and mode-conditioned responses in one shared policy directly from a native dual-mode checkpoint, without method-specific SFT warm-up. Paired current-policy counterfactual rollouts estimate cross-mode routing credit, which is assigned only to the routing token, while within-mode GRPO trains response tokens. A paired-to-self-routed curriculum stabilizes early training with forced coverage of both modes before progressively transferring training to autonomous routing. Across nine benchmarks, CounterRoute better balances accuracy and efficiency than heuristic and learned adaptive-routing methods, improving macro-average accuracy at both Qwen3 scales while cutting mean generated tokens by 51% and 41%, respectively, versus always-thinking checkpoints. On instruction-following and commonsense benchmarks where direct answering is strong, think rates fall as low as 1% while response quality improves. Despite training only on math and synthetic instruction-following data, its routing and response quality generalize to held-out tasks in coding, science, knowledge, and commonsense. --- ### Functional Architecture of European Electricity Trading Markets: Requirements for AI Supported Trading Systems under Regulatory Constraints Authors: Walter Kurz, Wojtek Stricker Published: 2026-09-24 Categories: cs.AI, eess.SY, q-fin.GN, q-fin.TR arXiv: https://arxiv.org/abs/2609.29108v1 European electricity trading in the EU operates as a constrained multi-layer system in which legal design, exchange microstructure, and network physics are executed jointly across forward, day-ahead, intraday, and balancing horizons. This paper develops a functional architecture for AI-supported trading that is aligned with market-coupling mechanics, cross-zonal transfer constraints, and compliance obligations under REMIT, MiFID II, MiFIR, and EMIR. The contribution is a formal system specification composed of a decision-state vector, residual-exposure accounting, constrained optimization objective, executable-action permission gate, and fail-closed AI control logic with auditable records. The analysis maps major Nominated Electricity Market Operator (NEMO) venues and related exchange operators into an operational venue topology and identifies where cross-border coordination fails in practice: interface-level timing, permission heterogeneity, and balancing-layer coupling. The resulting framework proposes how AI can be deployed as a bounded decision component inside regulated market operation with explicit governance, rather than as an unconstrained prediction layer. --- ### WildHSR: Metric Feed-Forward 4D People-Scene Reconstruction from a 3D Foundation Model Authors: Jerrin Bright, John Zelek Published: 2026-09-24 Categories: cs.CV, cs.AI, cs.CG arXiv: https://arxiv.org/abs/2609.29106v1 3D foundation models recover video cameras and geometry in one forward pass, but some of the strongest are up to scale. Joint people-scene reconstruction then requires two missing outputs: metric scale and persistent person identity. We ask whether one up-to-scale foundation representation can support both through lightweight adaptation. Exact metric labels are scarce, but unlabeled in-the-wild video is abundant. We use people in curated web video to initialise the solution: a posed metric body and 2D keypoints give an approximate, closed-form scale pseudo-label. These pseudo-labels pretrain a Scale Readout, which is then fine-tuned together with a lightweight adapter using exact metric supervision from standard real-video training splits. At inference the head predicts metric scale from foundation-model tokens, without the ruler or its teachers. For person identity, we probe the pretrained foundation model alone and find evidence that its intermediate query-key features encode person correspondence across frames. In most evaluated moving-person clips, a mid-layer token prefers that person over the vacated location and other people. A tiny projection reads this correspondence; together with metric pelvis motion and proposal confidence, it drives dustbin-aware Sinkhorn association of per-frame bodies. WildHSR combines both readouts to reconstruct metric cameras, scene and people from monocular video. Each window is predicted feed-forward; analytic association and Sim(3) composition connect windows. On EMDB-2, WildHSR is the first feed-forward method in the published comparison to beat the best optimization-based WA-MPJPE and RTE while leading feed-forward methods on all three world-frame metrics. On RICH, it leads feed-forward people-and-scene methods on WA-MPJPE and W-MPJPE. The complete pipeline runs at 10.1 fps on one GPU. --- ### ELF-REG: Scaling Continuous Diffusion Language Models to Reasoning Tasks Authors: Zeyu Michael Li, William Xingxu Chen, Bingshuo Qian, Jiayin Liu, Xiang Cheng Published: 2026-09-24 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.29102v1 Fully continuous diffusion language models (dLMs) denoise continuous representations without intermediate discretization, then decode all response tokens in parallel at the final step. Their performance on challenging reasoning tasks remains less established than that of autoregressive (AR) LLMs and masked dLMs. We scale Embedded Language Flows (ELF) to mathematical reasoning and code generation on GSM8K, MATH-500, HumanEval, and MBPP. We introduce ELF-REG, which improves learning with representation alignment and entanglement (REPA+REG), where a frozen AR teacher supervises intermediate denoiser features and supplies a global representation that is jointly denoised with the response. ELF-REG-L achieves 55.96% pass@1 on GSM8K at 64 network function evaluations (NFE), and 13.39% on MATH-500 and 22.56% on HumanEval at 128 NFE. It outperforms the evaluated comparable-scale dLMs in pass@1 on GSM8K and code, and improves MATH-500 pass@1 from 10.55% for the ELF-L baseline to 13.39% with ELF-REG-L. Without few-step training, the same task-specific checkpoints support strong low-NFE performance through early-stop, which decodes an intermediate clean prediction without completing the denoising trajectory. At 16 NFE, ELF-REG-L reaches 41.21% HumanEval pass@10, outperforming recent continuous dLMs of comparable scale. --- ### Language Specificity vs. Domain Diversity: Benchmarking Transformers for Bangla Medical NER Authors: Rakib Abdullah, Md. Maruful Islam Maruf Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.29101v1 Medical Named Entity Recognition (NER) for low-resource languages remains a challenging task due to high linguistic variability and a scarcity of domain-specific annotated corpora. This work presents a comprehensive empirical benchmark evaluating three fine-tuned transformer encoders-BanglaBERT, multilingual BERT (mBERT), and XLM-RoBERTa-against GPT-4o mini under zero-shot and few-shot prompting configurations for Bangla medical NER. In contrast to prior studies that evaluated large language models on limited subsets of only 50 samples, we conduct a large-scale evaluation across the full test set of 3,179 samples, providing statistically robust and reproducible baselines. Our fine-tuned XLM-RoBERTa model achieves an F1- score of 0.5959, establishing a new state-of-the-art and surpassing the previously reported best result of 0.5848. Crucially, we demonstrate that the language-specific BanglaBERT model consistently underperforms its multilingual counterparts with an F1-score of 0.4937, indicating that pretraining domain diversity can outweigh language specificity in highly specialized clinical settings. Furthermore, we present a detailed per-entity-type analysis for this task, revealing that Medicine and Specialist categories are recognized with high reliability, achieving F1- scores above 0.83, while the Symptom category remains the most challenging with an F1-score of 0.4367 despite being the most frequent training class. Finally, fine-tuned transformer models outperform the optimal prompting configuration by a factor of 3.76, confirming that prompt-only pipelines remain inadequate for structured clinical entity extraction in low-resource language environments. --- ### TraceGuard: Adaptive Multimodal Poison Filtering through Cross-Feature Rank Agreement Authors: Haoyang Li, Yaxin Xiao, Linyan Dai, Jiawen Fu, Zi Liang, Jason Xue, Qingqing Ye, Haibo Hu Published: 2026-09-24 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2609.29099v1 Multimodal training relies on image-text corpora collected from external sources, creating opportunities for attackers to poison the data. Stealthy attacks can preserve plausible image-text pairs while concealing the differences used by detectors, so apparently clean data can still redirect the trained model. We therefore ask which properties a poison set must preserve for the attack to remain effective. A small poison set must still exert enough collective influence during training to induce the attacker's target behavior. We analyze this influence in terms of how often an attack pattern occurs and how strongly the examples carrying it jointly affect the model. This analysis motivates six corpus-level features that examine cross-modal neighborhoods, recurring text, and changes after text-span erasure without training the victim model. We introduce TraceGuard, an adaptive rank-based filtering method that uses agreement among complementary feature rankings to identify suspicious examples. It refines the selected set through shared patterns and adapts the removal threshold to each corpus without knowing the attack or poison rate. Across 19 attack configurations spanning image-text learning, generative vision-language model fine-tuning, and encoder-transfer tests, TraceGuard removes an average of 98.4% of poisoned examples and 5.4% of clean examples. After training on the filtered corpora, the residual attack metric is at most 1% in 13 configurations. Matched-removal controls and ablations support the contributions of sample selection and adaptive removal. Stress tests also identify detection failures under adaptive attacks and unnecessary removal on poison-free corpora. --- ### Downside-Controlled Online Forecast Combination under Delayed and Revised Outcomes Authors: Minkyoung Kim, Hyunjung Byun, Yohan Lee, Beakcheol Jang Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.29096v1 Post-hoc correction adjusts a forecaster that cannot be retrained, such as a foundation model, but a correction fitted where errors are stable can hurt where they shift. We aim for downside control: not much worse than the starting forecast. We combine the frozen forecaster, a static corrector and an online corrector on the simplex, using only losses that mature after the horizon. Across seven benchmarks and four base models, two of them foundation models, the worst deterioration over 28 pairs at the main horizon is 0.15% and gains reach 11.5%. On day-ahead load for seven European bidding zones it lowers mean MSE in all seven zones, while single correctors raise mean MSE by up to 102% where the published forecast is most accurate. Three empirical conditions on expert speed, stream length and outcome alignment, each fixed by a documented failure, delimit its scope. Learning from the provisional outcome improves four zones on the settled one; learning on the settled outcome restores all seven. --- ### Where Does Exactly-Once Live? Model, Harness, and Tool-Contract Effects on Duplicate Side Effects in LLM Agents Authors: Jiapeng Li Published: 2026-09-24 Categories: cs.LG, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2609.29095v1 When a tool-using agent's write times out or returns a server error, the action may already have taken effect. Retrying blindly duplicates it -- a second charge, a second announcement, a second deployment -- while giving up skips required work. We ask where exactly-once behaviour should be enforced: in the model, in the agent harness, or in the tool contract. We introduce LIMBO, a deterministic sandbox of six services with realistic contracts (optional idempotency keys, eventually consistent and missing read paths) and twelve fault modes injected at the service boundary, including late commits, redelivery and partial batches; every episode is graded against a ledger of committed effects. Across 25,930 episodes spanning nine recent models, three production agent harnesses, two contract variants and fifteen recovery conditions, the answer depends on the fault. When an immediate read-back can reveal what happened, the model decides: frontier models instructed to act exactly once almost never duplicate a write whose acknowledgement was lost (0.5%), weaker models often do, and the model explains 53% of the explained variance. When it cannot -- the request is still in flight, or the transport delivered it twice -- the same frontier models duplicate in 56% and 74% of episodes, and the contract explains 81%. We prove that no verification-only policy is exactly-once under late commits without a bound on in-flight time. Waiting works when such a bound is short and known, but with heavy-tailed in-flight delays even an hour of waiting per episode falls short of offering an idempotency key on every write, which lowers the duplicate rate from 28% to 4% because agents use keys when they exist. The harness barely matters, a guard that attaches keys transfers across harnesses unchanged, and agents reported success in 90% of the episodes in which they had duplicated an effect. --- ### DAWN: Noise-Robust Quadruped Parkour via Depth-Denoising World Models Authors: Yohan Choi, Min-Jun Kim, Jin-Sung Kim, Yong-Jae Kim, Youn-Hee Han Published: 2026-09-24 Categories: cs.RO, cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.29092v1 Vision-based legged locomotion methods assume clean depth at training time and rely on hand-tuned post-processing filters at deployment. However, filter parameters are rarely disclosed, hindering reproducibility, and performance degrades substantially when depth noise is left unaddressed. Building noise robustness directly into the learning pipeline would eliminate this dependency. While such robustness has been explored for proprioceptive inputs, analogous approaches for depth perception remain largely absent in legged locomotion. We propose DAWN (Denoising and Alignment in World models for Noise-robustness), a noise-robust perception framework for legged locomotion, which builds noise robustness directly into a world model via two modifications: (1) feeding noisy depth to the encoder while keeping clean depth as the reconstruction target, forcing the model to implicitly denoise its input; and (2) applying contrastive learning to align the latent states of noisy and clean depth. Importantly, DAWN is not tied to a specific noise model, requiring no manual tuning to the noise distribution at deployment. Furthermore, it incurs no additional inference cost over existing world model-based methods. Without any manual filter calibration -- relying solely on the learned noise-robust representation -- DAWN achieves zero-shot quadruped parkour on a Unitree Go1: traversing stairs up to 18 cm, clearing gaps up to 70 cm, and mounting steps up to 45 cm from raw depth observations. Ablation studies show that denoising and contrastive alignment contribute at complementary levels -- reconstruction and representation, respectively -- and yield additive gains when combined. Videos and code are available at: https://dawn-parkour.github.io/ --- ### Can Classical Semantic-Extractive Summarization Be Evaluated in Hindi? A Replication Study Authors: Showket Ahmad Khan, Mudasir Mohd, Nasrullah Sheikh, Mohsin Altaf Wani, Abid Hussain Wani, Hilal Ahmad Khanday, Niyaz Ahmad Wani Published: 2026-09-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.29090v1 We replicate the distributional-semantics extractive summarisation method of Mohd, Jan and Shah (2020) and adapt it to Hindi, substituting a Devanagari-appropriate component at every language-specific step. The system is evaluated on two independent corpora --- the Hindi portion of XL-Sum and FIRE ILSUM 2.0 Hindi --- under a Devanagari-aware ROUGE implementation validated against the XL-Sum authors' own multilingual scorer, with all comparisons drawn as 1000-resample paired bootstraps. In its published equal-weight configuration the replicated system is significantly worse than a three-sentence lead baseline on both corpora, trailing Lead-3 by 0.042 ROUGE-1 Fon XL-Sum and by 0.265 on ILSUM. A feature ablation shows that sentenceposition is the only feature that contributes: position alone reproduces the lead baseline exactly, removing position gives the weakest configuration,and a validation-tuned weighting can at best equal Lead-3 and never exceed it. TextRank fails identically, making this a class-level rather than an implementation-level result. A selection analysis shows the remaining features steer extraction towards long, entity-dense body sentences while the references reuse the article lead.Current Hindi benchmarks therefore cannot reward non-lead content selection, motivating purpose-built evaluation resources. --- ### Physics and Data Driven Transformer-Mamba Framework for Flow Field Authors: Zhuo Zhang, Shun Zou, Canqun Yang, Xi Yang Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.29087v2 While deep learning accelerates expensive partial differential equation solving in computational fluid dynamics (CFD), existing methods like PINNs and FNOs often struggle with generalization, noise robustness, and physical consistency. We introduce the Transformer-Mamba for Flow Field (TM4FF) framework, a physics-constrained operator learning model with three key innovations: a Residual Wavelet Mamba (RWM) layer for feature denoising, a Transformer-based attention mechanism for enhanced feature fusion, and a physics-informed loss using Fourier derivatives to enforce the Navier-Stokes equations. Experiments on four CFD datasets show TM4FF achieves high accuracy and robust generalization across varying flow conditions. --- ### A Rapid Pipeline for Training and Deploying ML Models on WeBe Band Authors: Ehsan Kourkchi, Asmita Asmita, Houman Homayoun, Mahdi Eslamimehr Published: 2026-09-24 Categories: cs.AI, cs.LG, cs.PF arXiv: https://arxiv.org/abs/2609.29084v1 Developing optimized machine-learning algorithms for edge devices with limited computational and memory resources is challenging, time-consuming, and highly dependent on device-specific constraints. In this work, we streamline an edge ML workflow to enable rapid development, optimization, and deployment of machine-learning (ML) models directly on the WeBe Band, a wrist-worn wearable device designed for multimodal physiological data monitoring. The proposed system automatically generates hardware-efficient ML models that can be easily integrated into the WeBe core firmware, supporting AutoML, hardware-aware quantization, and performance profiling to build models that meet desired latency targets while remaining compatible with device memory and power limitations. The proposed framework tightly integrates the open-source Piccolo AI ecosystem with an automated pipeline that generates deployable firmware artifacts, performs hardware-aware model compilation, and supports over-the-air (OTA) deployment. The system supports multiple lightweight model classes, including classical machine-learning algorithms and neural networks, and provides built-in on-device profiling tools to evaluate inference latency and memory footprint under realistic execution conditions. Experimental results demonstrate clear trade-offs between model complexity and deployability on a microcontroller, showing that classical models offer strong real-time performance while lightweight neural networks require careful resource management. Rather than proposing new learning architectures, the current work mainly focuses on system-level automation, deployability, and enabling researchers and developers to rapidly iterate on models and evaluate them directly on target hardware. Although demonstrated on the WeBe Band platform, the workflow is designed to be extensible to other ML-powered edge devices. --- ### Feature Space Selection and Heterogeneous Effect Estimation for Blood-Brain Barrier Permeability: A Random Forest to the Generalized Random Forest Pipeline Authors: Tshemollo Rapolai, Seite Makgai, Mohammad Arashi Published: 2026-09-24 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.29076v1 Predicting blood-brain barrier (BBB) permeability is critical for central nervous system drug discovery. Using the MoleculeNet BBBP dataset (n = 2039), this study systematically ablates molecular feature spaces to isolate featurisation from model architecture. We evaluate three feature families (Morgan fingerprints, RDKit physicochemical descriptors, SMILES bigrams) across four learning algorithms. Results demonstrate that predictive performance depends jointly on feature representation and algorithm. Dynamic Random Forest using combined features achieved the highest mean AUC (0.970, 95% CI: 0.963-0.977). Second, this optimal representation enables exploratory estimation of heterogeneous associations between molecular structure and BBB permeability using Generalized Random Forests. Constructing a pseudo-treatment from a LogP median split, we applied double/debiased machine learning to account for confounding. Orthogonalization substantially attenuates the heterogeneity detected by naive causal forests; no conditional effects remained significant after false discovery rate correction (smallest adjusted p = 0.082). Furthermore, orthogonalized feature importance shifted toward residual structural information in SMILES bigrams. Ultimately, once observed confounding is properly accounted for, evidence that LogP-BBB associations vary systematically across chemical space is insufficient. This underscores that feature representation and model architecture are coupled design choices, and that unorthogonalized causal forests risk overstating genuine treatment effect heterogeneity. --- ### CRISS: A Retrieval-Augmented AI Chatbot for Assisting Cancer Registrars Authors: Vani Seth, Mohammad Beheshti, Anirudh Kambhampati, Vishwa Bhayani, Lucinda Ham, Prasad Calyam, Iris Zachary Published: 2026-09-24 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.29075v1 Cancer registrars, including Oncology Data Specialists (ODSs), must interpret complex and frequently updated coding and staging standards. We developed CRISS (Cancer Registry Intelligent Support System), a retrieval-augmented generation (RAG) conversational assistant that provides rapid, citation-supported access to registry guidance. This study evaluated whether CRISS could (1) support accurate and citation-supported responses, (2) improve access to and interpretation of relevant guidance, and (3) support training/helpdesk use while preserving human oversight of final abstraction decisions. We built a domain-specific knowledge base from national cancer registry standards, segmented into metadata-tagged passages and indexed as dense embeddings. Retrieved passages were used to generate citation-grounded responses through a large language model (LLM). Open-weight, proprietary, and non-RAG baseline models across Gemini and GPT families were evaluated on easy, medium, and hard registry questions using an LLM-as-a-Judge protocols. RAG configurations consistently outperformed non-RAG approaches, especially as question difficulty increased. Mean grounding scores for RAG were 0.62/0.56/0.59 across easy/medium/hard tiers versus 0.29/0.26/0.29 for non-RAG. RAG models also achieved higher semantic-similarity scores overall. Proprietary RAG models performed strongest on easy and medium questions, while local RAG models ranked highest on hard questions and proprietary models were generally more cautious. Domain-specific RAG improved evidence grounding and response quality for cancer registry questions while enabling citation-supported assistance across complexity levels. CRISS demonstrates the potential of human-centered, citation-grounded AI to support cancer registrars while preserving human oversight for final coding decisions. --- ### Seeing Is Not Measuring: Tool-Augmented Metric Spatial Reasoning for Vision-Language Models Authors: Kai Glantz, Clemens Grange Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.29073v1 Vision-Language Models (VLMs) describe scenes well but reason poorly about metric 3D structure such as absolute distances, physical sizes, or egocentric directions. We present a modular, predictor agnostic, tool-augmented framework that equips a small VLM (Qwen3.5-4B) with geometric tools: 3D object detection, metric depth estimation, and deterministic solvers for distance, size and bearing. Each object is detected in the camera frame of its own best view, and the tools use that frame's pose to lift every detection into one shared world frame. Moving metric computation out of the model's weights and into explicit solvers yields large gains on three of four ReVSI-Bench tasks: with a strong monocular detector (WildDet3D), absolute distance rises from 0.46 to 0.74 Mean Relative Accuracy (MRA), relative distance from 39.1% to 67.4%, and relative direction from a below-chance 25.9% to 73.4%. Because any detector can be swapped in behind the tool interface, comparing real detectors against ground-truth boxes separates perception error from reasoning error: orchestration costs only 0.03 MRA. Object size is bounded by the detector: the tools are near-exact on groundtruth boxes (0.97) yet the best real detector barely beats the no-tool baseline (0.61 vs. 0.58), because size reads straight off a box extent monocular detectors get wrong. Without a predefined recipe, the model already sequences the tools correctly on its own, matching a scripted pipeline on three of four tasks. --- ### BranchShine-CR: Compact Multilingual IPA Transcription with Self-Conditioned CTC and Consistency Regularization Authors: Nikhil Navas, Sergio Chevtchenko, Talisson Damiao, Saeed Afshar Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.29069v1 We introduce BranchShine-CR, a 25M-parameter model for multilingual transcription into the International Phonetic Alphabet (IPA). It combines log-mel features, a rotary-position E-Branchformer encoder, intermediate self-conditioned connectionist temporal classification (CTC), and consistency regularization across augmented views. On 16,646 shared IPApack++ test utterances, it achieves 4.47% IPA character error rate, a 22.3% relative reduction from ZIPA-CTC-NS, with approximately one-twelfth as many parameters while being trained from scratch. BranchShine-CR also outperforms a similarly sized NeMo Conformer baseline across all 41 dataset language labels. Ablation studies indicate the individual components synergetically acting in model performance contribution. These findings support compact IPA recognition capabilities under limited compute budget, for applications in low-resource on-device pronunciation assessment. --- ### EIB-Net: Entropy-Guided Information Bottleneck for Generalizable AI-Generated Image Detection Authors: Zhida Zhang, Xinlei Ma, Jie Cao Published: 2026-09-24 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.29064v1 The proliferation of photorealistic AI-generated images demands robust detection methods that generalize across diverse generative models. While existing approaches target manipulation-based forgeries with local artifacts, generation-based images (e.g., from diffusion models) lack such traces, posing a fundamental challenge. We observe that generative models prioritize global semantics at the expense of local texture fidelity, making low-texture regions key indicators of synthetic origin. To exploit this, we propose EIB-Net, an Entropy-guided Information Bottleneck Network. EIB-Net introduces a novel Image Entropy (IE) metric to automatically select the most informative (lowest-entropy) patch, then processes it with a Variational Information Bottleneck (VIB) to learn compact, generalizable features. Extensive experiments on DIFF, DiffusionForensics, and GenImage benchmarks demonstrate state-of-the-art performance: EIB-Net achieves 85.7\% accuracy using only 2\% of training data, outperforming full-image baselines by over 15\%, and maintains robust cross-generator generalization (83.5\% average accuracy on GenImage). Furthermore, our entropy-guided patch selection (EGPL) consistently enhances diverse backbones (CNNs and Transformers), proving its practical value for data-efficient detection. --- ### Transformers as Cross-Task Learners: Shared Structure Drives Sample Efficiency in In-Context Learning Authors: Zhongjie Shi, Rongjie Lai, Alexander Cloninger, Wenjing Liao Published: 2026-09-24 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.29060v1 Transformers achieve remarkable performance by jointly learning broad families of tasks during pretraining and adapting to unseen tasks from only a short prompt. Yet a rigorous mathematical and statistical understanding of this phenomenon remains limited. This paper aims to study how Transformers exploit shared cross-task structure and how this structure affects the sample complexity of in-context learning (ICL). Specifically, we characterize task-space complexity through covering numbers under a prescribed metric, thereby quantifying the low-dimensional cross-task structure without requiring an explicit parametric representation. The resulting cover provides a set of anchor functions, which we use to introduce a task-identification-and-evaluation procedure: context observations localize an unseen task among the anchor functions, and the response at a query is predicted by aggregating the corresponding anchor function query evaluations. For approximation, we explicitly construct a Transformer with Softmax attention to approximate this procedure. For generalization, we derive an error bound that separates the effects of the number of pretraining tasks and the prompt length. The scaling with respect to the number of pretraining tasks is governed by the intrinsic dimensions of the task space and input domain; once sufficiently many tasks are available, the dependence on the prompt context length becomes dimension-free. To the best of our knowledge, this is the first work to quantify cross-task complexity for general nonlinear task families and explicitly construct a Transformer that exploits their low-dimensional structure to perform ICL. Our theory provides a quantitative explanation of how joint pretraining across related tasks improves in-context generalization. --- ### Empath: Tracing Multi-Level Emotion Dynamics in Crisis Counseling Dialogues Authors: Ziwei Gong, Yuchen Huang, Wen Liang, Nicholas Deas, Melanie Subbiah, Kathleen McKeown, Julia Hirschberg Published: 2026-09-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.29056v1 Emotion dynamics are critical for understanding crisis-support conversations, yet most computational work treats emotion as static utterance-level labels. We introduce EMPATH, a framework for understanding affective dynamics in mental health dialogues across three granularities: turn-level labels, transition probabilities, and global conversation archetypes. Applying EMPATH to text-based crisis conversations with self-identified Black texters discussing grief, we find persistent negative affect, gradual hope-ward transitions, distinct texter-volunteer emotional roles, and heterogeneous recovery trajectories. These results highlight the informative patterns that emerge from computationally understanding crisis support and expressions of grief as dynamic processes within conversations, as well as the overall value of emotion-dynamic analysis for analyzing and comparing affect in dialogues. --- ### From Self-Distillation to Self-Practice: Privileged Information for Multi-Turn Agents Authors: Xingyu Su, Abhishek Kumar, Qing Ping, Youzhi Luo, Jonathan Buck, Zach Zhang, Subramanian Chidambaram, Vinayak Arannil Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.29051v1 On-policy self-distillation (OPSD) has become a popular recipe for post-training LLM agents. It supervises the agent model at the token level with a stronger teacher view of the same model, obtained by conditioning on privileged information (PI). In this work, we show that in multi-turn agents, this paradigm teaches the student to act with confidence but without the information behind it. The trained agent behaves as if it had privileged information it never observed, and its performance falls well short of plain RL, in the worst case below the untrained base model. Therefore, we propose Privileged Self-Practice (PSP), which keeps the PI and moves it from the loss to the sampler. When the student's rollouts on a task mostly fail, we inject a short per-task instruction written by an analyzer model, sample the task again with the instruction in context, and train on the result with an unchanged GRPO objective. The privileged information stays in the prompt and never enters the loss. Across AppWorld and SWE-bench Verified, with three different student models, PSP obtains the best average score in every setting and is the only method that consistently outperforms plain GRPO, improving task-goal completion by up to 65% on AppWorld and the resolved rate by up to 61% on SWE-bench Verified. --- ### SLCA-GRPO: Resolving Cross-Segment Credit Misattribution in Tool-Calling RL Authors: Yan Zhan, Shaobo Liu, Qiunan Liu, Yuanjun Shi, Siqi Xu, WeiYi Hou, Xiang Xu, Zekang Li, Weizhou Pan, Jiahong Yan Published: 2026-09-24 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.29050v1 Tool-calling agents produce heterogeneous outputs, interleaving structured tool invocations with user-facing natural language summaries. This output heterogeneity presents a structural failure mode in standard on-policy Reinforcement Learning (RL): algorithms like GRPO indiscriminately broadcast a homogeneous trajectory-level scalar advantage to all tokens. Consequently, gradient noise from summary generation leaks into tool-decision tokens, causing cross-segment credit misattribution and brittle optimization. In this work, we propose SLCA-GRPO, a framework incorporating Segment-Locked Credit Assignment (SLCA). To enable scalable exploration without costly real APIs and stable training, we first construct the Schema-Guided LLM Simulator (SGLS) as foundational training infrastructure. Building on this, SLCA decouples advantage estimation at the structural segment level within a single group of rollouts, without requiring additional rollouts from intermediate states. Supported by Hierarchical Rewards (HierR), SLCA routes execution advantages to tool tokens and preference advantages to summary tokens, eliminating advantage contamination (the dominant cross-segment credit misattribution channel) within each policy update. On a 7B backbone, SLCA-GRPO accelerates convergence and outperforms standard GRPO, ToolPO, and RLTR by +2.53 pp on in-domain evaluation, +1.36 pp on the Berkeley Function-Calling Leaderboard (BFCL), and +9.15 pp on $τ^2$-Bench under the same training budgets, achieving higher accuracy with reduced tool redundancy and costs. --- ### Where Hallucinations Live: A Cross-Architecture Circuit in VQ-Tokenized Vision-Language Models Authors: Shamanthak Hegde, Xiangrui Liu, Maitreya Patel, Yezhou Yang Published: 2026-09-24 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.29048v1 Unified vision-language models (VLMs) that tokenize images through a vector-quantized (VQ) codebook routinely hallucinate objects on grounded yes/no benchmarks, yet existing decoding-time fixes treat this as generic miscalibration without an architectural account. Using activation patching across twenty-five models spanning eight LLM families, we identify an early-layer ($L_0$) attention routing circuit shared across VQ-tokenized VLMs and propose a three-gate diagnostic that distinguishes the models carrying it from those that do not. The diagnostic isolates ten positive models (five natural unified-VQ VLMs across three LLM families and five induced variants) and rejects the remaining fifteen. A single-variable architectural swap (LLaVA-1.6 CLIP+MLP $\rightarrow$ VQ+Linear) installs the circuit, while a matched-compute MLP control on identical data does not, isolating vector quantization as the source of the pathological signal; the routing pathway that carries it is one that the backbone already provides. Against tuned VCD and DoLA baselines, tuned DoLA wins on binary calibration, but \textbf{only $L_0$ ablation reduces object hallucination in open-ended generation} (CHAIR$_i$ reduces by $31\,\%$ relatively, whereas tuned DoLA and VCD leave it unchanged or worsen it). These results recast object hallucination in unified VQ VLMs as a property of architecture and pretraining, and yield a targeted intervention that mechanism-agnostic decoding cannot replicate. --- ### The Tokens Remember: When Tokenization Bypasses Knowledge Editing and Unlearning Authors: Manit Baser, Aditya Nawal, Dinil Mon Divakaran, Mohan Gurusamy Published: 2026-09-24 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.29045v1 Open-weight LLMs give downstream users control over the inference stack, but this flexibility can undermine post-release guarantees that sensitive knowledge has been modified or removed. Model editing and machine unlearning are used to modify or remove targeted knowledge without retraining models from scratch. However, existing security evaluations of these techniques face two critical limitations. First, they typically require access to either the original pre-edit/unlearning model or auxiliary classifiers to detect modifications or reconstruct pre-edit behavior. Second, they evaluate modifications under the canonical tokenization of an input, implicitly treating tokenization as a benign preprocessing step. We show that this assumption creates a security gap: the same input string can be represented by alternative valid tokenizations that induce different computational trajectories, allowing an adversary to bypass localized modifications and recover information intended to be suppressed. We introduce Toketive, a simple yet powerful reference-free attack that exploits the tokenization-based side channel to (i) detect modified knowledge and (ii) reconstruct the corresponding pre-edit response. It operates solely on the released model and requires neither the pre-edit model, training data, shadow models, nor auxiliary classifiers. Across five LLMs, six datasets, and six editing and unlearning techniques, we find that 38.6% of alternative tokenizations bypass the modification and recover the pre-edit response. Toketive detects modified facts with an F1 score of 84.2%, a 26.2% relative gain over the strongest baseline, and reconstructs pre-edit responses with 74.5% top-5 accuracy, 21.7% higher than the best baseline. Our results show that localized modifications should not be treated as robust knowledge-control boundaries without adversarial evaluation over alternative representations. --- ### Multi-Agent Orchestration of 3GPP Channel Estimators Authors: I. Zakir Ahmed, Hamid Sadjadpour Published: 2026-09-24 Categories: cs.IT, cs.AI, eess.SP arXiv: https://arxiv.org/abs/2609.29044v1 Pilot-aided channel estimation is a decisive block in orthogonal frequency-division multiplexing (OFDM) receivers for both 5G New Radio (5G-NR) and Long-Term Evolution (LTE). A large body of estimators exists, from simple least-squares (LS) interpolation to statistically optimal linear minimum-mean-square-error (LMMSE) variants and, more recently, deep convolutional denoisers, yet no single estimator is uniformly best: the winner depends on the propagation scenario, the numerology, the operating signal-to-noise ratio (SNR), the mobility (Doppler), and the antenna configuration. In this paper, we quantify this fact through a unified study of eight literature estimators evaluated over the 3GPP TR~38.901 Urban-Macro (UMa), Urban-Micro (UMi), and Rural-Macro (RMa) channels generated with NVIDIA Sionna, for both 5G-NR and LTE numerologies, in single-input single-output (SISO) and $8\times2$ multiple-input multiple-output (MIMO) settings. We then propose a \emph{condition-adaptive multi-agent orchestrator} that treats each estimator as an independent agent and dispatches, per operating condition, to the agent that is best on a validation split without any genie knowledge. The orchestrator tracks the per-realization oracle to within $1.07$~dB and improves the normalized mean-square error (NMSE) over the best \emph{fixed} strategy by up to $3.6$~dB at high SNR, where the low-SNR champion is no longer optimal. Because the agents are independent, running them concurrently delivers this best-of-eight accuracy at essentially single-estimator latency: a data-parallel partition scales the wall-clock nearly as $1/K$ with $K$ workers (up to $6.9\times$), whereas naive by-algorithm partitioning is Amdahl-limited by the heaviest agent. The results substantiate multi-agent orchestration as a practical route to robust channel estimation across heterogeneous 5G-NR/LTE deployments. --- ### Design and Evaluation of LLM Chaining-Based Task Planning for General Purpose Service Robots Authors: Lucas Da Mota Bruno, Jiahao Sim, Yoshinobu Hagiwara Published: 2026-09-24 Categories: cs.RO, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.29043v1 General Purpose Service Robot (GPSR) tasks, as defined in the RoboCup@Home benchmark, require robots to interpret diverse natural language commands and generate multi-step action sequences in real home environments. Conventional Single Prompt (SP) approaches suffer from context bloat and the "Lost in the Middle" phenomenon, leading to unreliable task planning. We propose an LLM chaining architecture that separates instruction classification and action generation into two specialized stages, reducing per-inference prompt length by approximately 45% while improving planning consistency. We evaluate our method using 100 randomly generated GPSR commands across three language models spanning local open-source and frontier cloud deployment contexts. Results show consistent planning improvements over SP across all models, with gains of up to +37 percentage points on local models. Further, real-robot execution experiments on the Toyota Human Support Robot (HSR) reveal that planning success alone does not guarantee task completion, with 6 of 10 tasks completing successfully and execution-layer failures identified as the primary remaining bottleneck. --- ### Personalised federated learning for Riemannian and Euclidean EEG decoding Authors: Thibault Pautrel, Florent Bouchard, Ammar Mian, Guillaume Ginolhac Published: 2026-09-24 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.29037v1 Federated learning (FL) lets EEG decoders learn from recordings of several subjects without pooling them. We consider two light EEG decoders, the Riemannian SPDNet and the Euclidean EEGNet. Both split into a trunk, which builds a latent representation, and a head, which classifies it. Inter-subject variability, however, makes a single shared FL model a poor fit for each subject. Personalised FL addresses this: all subjects learn a common trunk, and each subject keeps its own head. We adapt it for SPDNet and study its effects against standard FL and centralised training, with EEGNet as a Euclidean baseline. Experiments cover three motor-imagery datasets that span diverse regimes in channels, subjects and classes. We observe that personalised SPDNet reaches higher accuracy than both standard FL and centralised training, while converging in fewer rounds and communicating fewer parameters than standard FL. It also outperforms every EEGNet configuration on two of the three datasets, although centralised EEGNet outperforms centralised SPDNet. --- ### Paging the Experts: A Reproducible Characterization of Flash-Backed MoE Inference on iPhone Authors: Musa Shams Published: 2026-09-24 Categories: cs.PF, cs.LG arXiv: https://arxiv.org/abs/2609.29032v1 Sparse activation reduces mixture-of-experts computation without eliminating the need to store all experts. We present Routide, a Swift/MLX runtime that executes the text path of a pinned public Qwen3.6-35B-A3B quantized checkpoint while keeping expert weights in iPhone storage and a byte-budgeted subset in memory. We characterize cache-policy sensitivity, numerical comparison boundaries, and measurement limits. Across five recorded 128-token workloads, fixed-route replay gives 0.00% demand hits with a 512 MiB LRU cache, 18.80% with seeded random eviction at the same budget, and 38.58% with 576 MiB LRU. The apparent capacity cliff is therefore a policy/workload interaction, not a universal memory requirement. Same-runtime Mac controls preserve generated sequences across eviction and asynchronous prefetch, including 2,560 exact token comparisons and 10,334 speculative loads. In contrast, complete resident-Python versus recorded-phone sequences disagree on all five tested cases, precluding a general numerical equivalence claim. Two separately scoped iOS 27 memory protocols observe sampled process-footprint peaks of 1.87-2.32 GiB on short prompts and 2.39-2.73 GiB on one longer prompt. We retain a thermal stopping event, negative timing comparisons, and a single qualified whole-device power estimate. These results establish bounded feasibility and identify limitations that a deployment claim must not hide. --- ### Exploiting answer-invariant redundancies in satellite imagery for efficient VLM inference on edge Authors: Ishani Janveja, Davis Zhang, Seoyul Oh, Deepak Vasisht Published: 2026-09-24 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.29029v1 Onboard vision-language models could enable satellites to answer queries directly, but exhaustive tiled inference over high-resolution imagery is slow and energy-intensive. We identify answer-invariant token redundancy (AITR): image tiles and vision tokens that can be removed without changing the final answer. We present Rift, a two-stage system that performs query-conditioned tile pruning followed by elastic prefill to reduce token budget. We evaluate it on LLaVA-1.5 7B running on Jetson AGX Orin. Compared with exhaustive tiled inference, Rift reduces energy by 78% and latency by 69%, while increasing accuracy from 45% to 73%. --- ### RGBD20K: A Large-Scale Benchmark for RGB-D Semantic Segmentation Authors: Shaohua Dong, Zexuan Meng, Haiyan Sun, Bing Fan, Cuicui Zhang, Dylan Joseph, Kewei Sha, Yunhe Feng, Heng Fan Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.29028v1 In this paper, we propose RGBD20K, a novel dataset for facilitating the development of more robust and general RGB-D semantic segmentation by encompassing abundant categories and high-quality annotations. RGBD20K possesses several attractive properties: (1) Expanded Semantic Space. In particular, it covers 160 fine-grained categories, largely surpassing the category diversity of existing popular RGB-D benchmarks (e.g., NYUv2 with 40 classes and SUN RGB-D with 37 classes). With such enriched semantic coverage, we expect to promote the learning of more generalizable segmentation models. (2) Larger Scale. Compared with current benchmarks, RGBD20K offers 20,000 RGB-D image pairs, providing a substantially larger training resource that benefits the development of more powerful deep models. (3) High-Fidelity Annotation. We perform rigorous re-evaluation and correction of existing labels to resolve long-standing annotation noise, resulting in a clean and reliable ground-truth foundation. Furthermore, we propose a novel score-purified fusion (SPF) method, which achieves state-of-the-art performance across all evaluated benchmarks, demonstrating the effectiveness of our approach in leveraging high-quality multimodal information for RGB-D semantic segmentation. The dataset is here: https://github.com/ShaohuaDong2021/RGBD20K/. --- ### Generative Atmospheric Super-Resolution from Heterogeneous In Situ Observations through Composable Interfaces Authors: Yang Xu, Dibyajyoti Chakraborty, Haiwen Guan, Sen Wang, Romit Maulik Published: 2026-09-24 Categories: cs.LG, physics.ao-ph arXiv: https://arxiv.org/abs/2609.29027v2 Atmospheric observations are sparse, heterogeneous, and unevenly distributed, whereas many generative atmospheric models learn distributions over regularly gridded multivariate states. Once pretrained, diffusion models can supply atmospheric priors that can be combined with observation-derived likelihood factors in a Bayesian formulation. However, these observation sources differ substantially in geometry and sampling density, complicating the consistent use of their observations within a common inference framework. Here, we formulate this reconstruction problem as generative atmospheric super-resolution and introduce composable observation interfaces for conditioning a single pretrained 13-variable atmospheric diffusion model. The interfaces convert sparse radiosonde (R), clustered aircraft (A), and dense irregular surface-station (S) observations into source-specific likelihood factors that specify where observations constrain the gridded state, how residuals are counted under uneven sampling, and how strongly each source guides posterior sampling. We developed the aircraft and surface observation interfaces using 2019 observations and evaluated the selected interfaces throughout 2020 without further tuning. Compared with reconstructions conditioned only on radiosonde observations, the composed R+A+S interface reduces RMSE evaluated against ERA5 by $9.24\%$ across all 13 state variables over the CONUS domain. The aircraft and surface factors provide complementary improvements in upper-air and surface variables. The R+A+S combination also lowers the Continuous Ranked Probability Score (CRPS), while evaluations at held-out aircraft and surface-station observations show reduced prediction errors. Together, these results demonstrate a modular route for conditioning a pretrained atmospheric generative prior on heterogeneous in situ observations without retraining the underlying model. --- ### Growth-Inspired Graph Generation and Inverse Design of Mechanical Lattices via Dot Matrices Database Augmentation and GCNN Authors: Weiyun Xu, Jiamu Liu Published: 2026-09-24 Categories: cs.LG, cond-mat.mtrl-sci, physics.app-ph arXiv: https://arxiv.org/abs/2609.29024v1 Natural load-bearing and transport networks are not assembled in a single step; they emerge through a temporally ordered process of growth, branching, reinforcement, and loop formation. Inspired by this developmental logic, this work introduces a morphogenetic graph-generation framework for mechanical lattices in which a discrete dot matrix provides potential nodes and the final architecture is created by sequential cross-layer and intra-layer growth. The same rule is visualized in two dimensions as a leaf-vein-like developmental sequence and implemented in three dimensions on a 3x3x3 nodal matrix containing 27 candidate nodes. A dataset of distinct three-dimensional lattices was evaluated by beam-based finite element analysis and represented directly as graphs. A graph convolutional neural network (GCNN) with three graph-convolution layers and dual global pooling learns the topology-property mapping and predicts effective compressive stiffness. Coupling the GCNN surrogate with rapid structural sampling enables inverse design: for a target stiffness of 1000 MPa, the selected design was predicted at 1042.43 MPa and validated by finite element analysis at 1027.49 MPa. Beyond straight members, the framework has also been extended to parameterized horseshoe-shaped curved beams made of nonlinear materials, enabling topology-geometry design toward prescribed deformation shapes. Our work provides a paradigm for augmenting the database of mechanical metamaterials, and the resulting perspective links biological morphogenesis, graph learning, and nonlinear shape programming in a unified generative design framework for architected materials. --- ### EvoTreeNAD: Genealogy-Guided Evolution for LLM-Driven Neural Architecture Discovery Authors: Lishan Yu, Derek Jiu, Qizhen Lan, Xiaoqian Jiang Published: 2026-09-24 Categories: cs.NE, cs.LG arXiv: https://arxiv.org/abs/2609.29016v1 AI-driven scientific discovery accelerates research by autonomously developing solutions and designs. Large language model (LLM) agents support this process through iterative generation and evaluation. Yet these iterations alone do not ensure cumulative progress or establish which directions to pursue next. Costly evaluation further constrains the scope of exploration. Neural architecture discovery brings these challenges together, coupling open-ended design with resource-intensive experimentation. We introduce EvoTreeNAD, a genealogy-guided evolutionary algorithm that constructs trainable architectures without a supplied seed or a hand-specified search space. Starting from an empty root, it grows a persistent genealogy in which each new node represents a complete architecture. Top-percentile values computed from each node and its descendants guide lineage selection. Using the selected design history, an Idea Agent proposes a variant and a Code Agent implements it. Each evaluated variant becomes a child node, expanding the genealogy while providing evidence for subsequent lineage selection. Our theoretical analysis establishes the existence of stationary variation regimes as the genealogy grows. Under specified variation assumptions, sustained top-percentile family values quantify the probability of generating high-reward architectures in these regimes. EvoTreeNAD discovers architectures that outperform the compared NAS and NAD baselines, achieving CIFAR-10/100 test errors of $2.05{\pm}0.06\%$ and $15.09{\pm}0.22\%$. On all six MedMNIST-v2 tasks, the discovered architectures surpass the strongest listed baselines. A controlled CIFAR-10 study further shows that EvoTreeNAD outperforms direct generation, best-of-$N$ greedy continuation, and full-family-mean routing. --- ### MeshHeal: Two-Timescale Self-Healing for Gray Failures in Decentralized LLM Agent Networks Authors: Keru Chen, Sen Lin, Yingbin Liang, Nathaniel D. Bastian, Shaofeng Zou Published: 2026-09-24 Categories: cs.AI, cs.CL, cs.DC arXiv: https://arxiv.org/abs/2609.29015v1 Decentralized LLM-based multi-agent systems coordinate through local interactions, but an agent can remain responsive while its task-solving quality persistently degrades. Such gray failures require protecting current tasks before sufficient evidence exists to alter future routing, while still allowing recovered agents to rejoin. We introduce MeshHeal, a fully decentralized self-healing framework that couples ability-matched peer review across two timescales. At the fast timescale, an adaptive hierarchy escalates uncertain or low-scoring outputs from repeated single-reviewer evaluation to committee deliberation and, when needed, correction before use. At the slow timescale, a task- and ability-conditioned peer-relative detector aggregates scores to distinguish persistent degradation from ordinary output variation, trigger mandatory committee review, and eventually exclude degraded agents from ordinary routing; recovery probes provide fresh evidence for reintegration. To faithfully evaluate routing, we introduce Model-Backed MAS Evaluation, which ties ability assignments to execution models, since prompt-based ability assignments alone can leave routing errors hidden. Across BBH, MATH, and MMLU-Pro, MeshHeal achieves 0.839 degraded-phase accuracy using 51k total model tokens per task, versus the strongest baseline Symphony's 0.807 accuracy using 115k per task. Under staggered degradation and recovery, MeshHeal isolates degraded agents, keeps them excluded from ordinary task execution until recovery, and returns them to normal routing. --- ### AlphaDiverse: Post-Training Local Quantitative Research Agents for Diverse Exploration in Alpha Factor Mining Authors: Qingzhuo Wang, Zikun Wei, Zhihua Wei, Wen Shen Published: 2026-09-24 Categories: cs.AI, cs.CE, cs.MA arXiv: https://arxiv.org/abs/2609.29014v1 Large language model (LLM)-based multi-agent systems can automate alpha factor mining, but their reliance on external APIs limits control over cost, availability, and confidentiality. Long research loops also tend to revisit a few successful economic mechanisms that lead to research path collapse. To address these limitations, we propose AlphaDiverse, a framework that integrates a multi-agent alpha research system, diverse research path collection, and post-training for local agents. We let the research system generate complementary plan portfolios and vary research environments across loops to collect diverse research paths. Using these diverse traces, we warm-start local Planner and Realizer agents with supervised fine-tuning. Then, we propose a joint GRPO method to optimize both of them using predictive quality and diversity of contributions. Research feedback is confined to inner period data, while a frozen final model is evaluated on a later outer period data, thereby avoiding test-set tuning. Experiments across four Chinese stock universes show that AlphaDiverse can combine competitive prediction with broader exploration. --- ### When Does Action Credit Need Updating? Authors: Hongye Yang, Boxiao Huang Published: 2026-09-24 Categories: cs.AI, stat.ML arXiv: https://arxiv.org/abs/2609.29007v1 Tool-using agents are continually updated with new interaction data. After each policy update, however, previously estimated action credits may become stale. Recomputing them from scratch can require many additional tool calls and environment interactions, making repeated updates increasingly expensive. We ask a simple question: when does historical action credit actually need to be updated? Our key observation is that a change in action value does not necessarily imply a change in the decision. Historical credit can still be useful as long as policy-induced drift is too small to overturn the existing action ranking. Building on this idea, we introduce pairwise branch sensitivity to capture how strongly a policy update affects the downstream regions that distinguish two candidate actions. We then derive a first-order anchored credit-transport estimator that updates historical credit using old interventional trajectories, and propose a Decision-Sufficient Credit Gate (DSC-Gate) that chooses whether to reuse, transport, or resample credit. Experiments show that branch sensitivity explains credit drift substantially better than global policy distance. With sufficient historical data, credit transport reduces estimation error, while its benefit to decision making is concentrated on updates that affect action-distinguishing branches. On a fully independent test set, DSC-Gate changes mean regret by only +0.00004 relative to a gap-based gate while reducing mean new tool steps from 472 to 286, a 39.4% reduction. We observe the same pattern after a real tool-agent parameter update. Overall, our results show that agents do not need to recompute action credit after every policy update: much of the historical evidence can be reused or cheaply corrected, reducing the additional interaction required to keep action decisions up to date. --- ### FluidRain: Incompressible Rain Flow as an Attention Bias for Loop-in-Loop Video Deraining Authors: Pu Wang, Yongcong Wang, Wenhao Li, Xiang Chen, Guangwei Gao, Jinshan Pan, Siyuan Yao, Shujun Fu, Zhuoran Zheng Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.29006v1 Existing video deraining methods typically exploit neighboring frames through either explicit alignment or implicit spatiotemporal aggregation. Explicit alignment relies on accurate motion estimation, which can become unreliable under dense rain, while implicit aggregation avoids alignment but lacks explicit guidance on the directional and temporally coherent structure of rain. This leaves a gap between reliable temporal aggregation and explicit modeling of rain motion. To address these limitations, we propose FluidRain, a lightweight video derainer that uses divergence-free rain flow to guide Loop-in-Loop attention across scales and neighboring frames. Motivated by fluid mechanics, we model rain motion as a divergence-free image-space flow and use it to organize multi-scale and temporal aggregation. Specifically, FluidRain first estimates a rain-flow field for each frame and projects it onto the divergence-free subspace. The resulting flow steers window attention along rain streaks, enabling neighboring frames to be aggregated without explicit alignment. Since rain-flow structure is preserved across scales and nearby frames, Loop-in-Loop reuses the same attention operator across both dimensions, resulting in a three-frame model with only 0.80M parameters. Experiments on four benchmarks show that FluidRain remains competitive with substantially larger restoration models. We further examine how temporal evidence scales with different input views. To evaluate whether the model remains reliable when rain motion changes across frames, we introduce RainSyn-Gust, which injects controlled changes in rain-streak direction into existing benchmarks. We also develop a physics-based no-reference metric that evaluates real-rain removal without requiring clean targets. --- ### Polite but Misaligned: Evaluating LLM Politeness Judgments Against Human Pragmatic Norms Authors: Rong Wang, Kun Sun, Yadong Guo Published: 2026-09-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.29001v1 Despite strong performance on standard benchmarks, it remains unclear whether large language models (LLMs) evaluate social pragmatics in ways that align with human judgments. We evaluate LLM politeness judgments using two English-language datasets with complementary annotation formats: continuous human ratings and three-way categorical labels. Across the seven evaluated models, we find that inter-model agreement is stronger than model--human agreement. Strategy-level analyses suggest that model--human alignment is associated with explicit linguistic cues, while some rapport-building strategies occur more frequently in misaligned cases. In the categorical task, model predictions exhibit systematic neutral compression, characterized by the overproduction of Neutral labels and the underprediction of Impolite labels. This pattern persists when expert consensus is used as the reference on a diagnostic subset. Our findings highlight the need for pragmatic evaluations that go beyond aggregate agreement metrics by examining directional patterns of model--human disagreement across different human references. ---