# Research Papers — 2026-10 · bundle page 135 of 156

> Full-text markdown of papers #6701–6750 of 7799 published in 2026-10, newest first. AI-readable mirrors of the full 7799-paper 2026-10 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-10/page/135
[← Previous bundle](https://aipod.dev/papers/archive/2026-10/page/134.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-10/page/136.md)

### How Much Do LLM-as-a-Judge Design Choices Matter? A Systematic Comparison of Prompt Designs, Rating Scales, and Models

Authors: Laurène Vaugrante, Thilo Hagendorff
Published: 2026-10-04
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2610.05094v1

Researchers increasingly use Large Language Models as judges (LLM-as-a-judge) to evaluate model outputs. Yet there are no standards for how to design these judges. Typically, researchers choose the prompt, rating scale, and model intuitively. If these choices change the judge's verdicts, two studies can reach different conclusions about the same facts. To address this risk and to provide an empirical basis for judge designs, we evaluate 10 reasoning models across multiple designs on two tasks: a scalar rating of sentence sentiment and toxicity (over 500 items per category), as well as a binary accuracy classification of question-answer pairs (n=600). For the rating tasks, despite judges showing significant disagreements with the human ground truth, the practical size of differences is small enough to consider most judges reliable (mean absolute deviation of 0.11 points on a 1 - 7 scale); toxicity judges even outperform standard classifiers. Judges are also highly accurate on average (96.5%) for the accuracy classification task. However, design choices can produce shifts: changing the rating scale alone can shift measured bias by up to 0.93 points (rating task), and while accuracy levels are rarely impacted, design choices consistently impact judge leniency (classification task; leniency drop of 28.9 percentage points when using detailed prompts, and up to 56.1 percentage points when switching models). Counterintuitively, lower reasoning effort affects neither accuracy nor leniency. Across both tasks, model identity is the dominant source of variance. These findings suggest that while LLM judges are broadly trustworthy in aggregate, design choices can be meaningful sources of variance. Given the growing reliance on automated evaluation in LLM research, we intend this study as a methodological reference for designing more robust and replicable LLM-as-a-judge pipelines.

---

### Direction-Conditioned Policies for Online Goal-Conditioned Reinforcement Learning

Authors: S K Swaminathan, Damiya Gondha, Theyanesh Eswaramoorthy Rajahkrishnan, Aritra Hazra
Published: 2026-10-04
Categories: cs.LG, cs.AI, cs.RO
arXiv: https://arxiv.org/abs/2610.05087v1

Contrastive Reinforcement Learning (CRL) learns representations that estimate goal reachability, yet its policy remains conditioned on raw goals and therefore does not directly exploit the geometry encoded by its critic. We introduce Direction-Conditioned Policies (DCP), a method built around a small modification to CRL: DCP selects previously visited states as waypoints during online training and conditions the policy on their direction and distance in representation space. At deployment, DCP applies the same interface directly to the final goal, requiring neither waypoint selection nor planning. Across nine navigation and manipulation tasks, DCP attains higher final success rates than CRL on seven tasks and spends more time near the goal on seven. Controlled maze experiments further show that DCP captures shortest-path geometry more accurately and that the supplied direction causally influences the actor's behavior. We identify waypoint coverage and ranking as limits to exploration, and show that learned candidate generation improves goal reaching in two controlled mazes.

---

### Towards cross-cultural study of folksong lyrics with machine translation

Authors: Anna Dvořáková, Anna Aljanaki, Danbinaerin Han, Peter van Kranenburg, Matěj Kratochvíl, Inna Lisniak, Zdeněk Vejvoda, Jan Hajič
Published: 2026-10-04
Categories: cs.CL, cs.DL
arXiv: https://arxiv.org/abs/2610.05084v1

Music is universally present in human societies. Ethnomusicologists have long been documenting the diverse expressions of human musicality, and comparative musicology has recently brought several studies of folksong to a more global scale. Such cross-cultural research has not been conducted on lyrics: the language barrier has so far prevented work with multi-lingual data. However, Natural Language Processing (NLP) technologies have reached a stage where this language barrier may no longer be prohibitive. Combining folksong lyrics corpora across five languages, we machine-translate them to a pivot language with a pre-trained neural topic model, and we examine the relationship between content and social function within each language, and across languages for wedding songs. As expected, human evaluation of translation results shows that non-Indo-European languages suffer from overall worse translation quality. Experiments with topic models then indicate that the content of lyrics is at best partially related to the social function of folksongs across all languages. These experiments are just first steps into cross-cultural folk musics lyrics analysis; however, they do indicate that a previously unobserved web of cross-cultural relationships beyond ethnomusicological typologies may be uncovered through the study of what people sing across the world's diverse folk musics.

---

### Tracing a Sparse Emotion-Control Circuit in LLM-Based Text-to-Speech

Authors: Hongfei Du, Jiacheng Shi, Yanfu Zhang, Ye Gao
Published: 2026-10-04
Categories: cs.SD, cs.AI
arXiv: https://arxiv.org/abs/2610.05080v1

LLM-based text-to-speech (TTS) models can generate emotionally expressive speech, but how reference emotion is routed through the model and realized in decoded speech remains unclear. We introduce two emotion-sensitive metrics for matched neutral and emotional syntheses---a codec trajectory score and a late residual direction score---and use them to score activation-patching interventions. Under controlled matched-reference conditions, this analysis identifies a sparse source-to-readout component-level circuit: 23--27 attention heads and MLPs per emotion, roughly 5% of the components considered, recover or suppress 74--88% of the late emotion-readout shift on held-out cases. The circuit combines a shared component backbone with emotion-specific components; cross-emotion activation swaps reduce the target readout in 47 of 48 cases. In decoded speech, the same intervention produces consistent changes in pitch, energy, and spectral brightness over 24 matched pairs per emotion. A readout-matched residual-direction baseline produces only 17--27% of the intervention's pitch effect, showing that internal readout movement alone does not explain the decoded acoustic changes. These results trace a compact causal route from reference-derived prefix information to emotion-relevant properties of generated speech.

---

### CGDD-Net: Context-Guided Dynamic Detail Modeling for Retinal Vessel Segmentation

Authors: Xincheng Li, Xiaoqi Sheng, Xinyu Zhang
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05078v1

Accurate retinal vessel segmentation requires features that capture vascular geometry while preserving information for fine-scale reconstruction. We propose CGDD-Net, a context-guided dynamic detail modeling network that connects adaptive feature extraction to a shared decoder pathway. Context-Guided Scale-Adaptive Deformable Encoding (CSDE) combines fixed-grid convolution with deformable local attention to capture vascular patterns at multiple spatial extents. Spatially Adaptive Multi-Kernel Gating (SAMG) selects receptive-field responses at each location. Dynamic Cross-Scale Detail Fusion (DCDF) aligns the gated intermediate features and compresses them into an eight-channel representation, which is reused at three decoder resolutions together with selected encoder skips. This design consolidates intermediate information before decoding instead of transferring each middle-stage feature through a separate direct skip. On DRIVE, CHASE\_DB1, STARE, and HRF, CGDD-Net achieves AUC values of 0.9824, 0.9938, 0.9895, and 0.9874, with F1 scores of 0.8323, 0.8102, 0.8510, and 0.8157, respectively. The complete model contains 1.96 million trainable parameters. In cumulative ablations, the full model improves F1 over the internal baseline by 2.48, 0.61, and 3.49 percentage points on DRIVE, CHASE\_DB1, and STARE. Twelve directed cross-dataset experiments further characterize transfer without target-domain adaptation. The results support shared intermediate detail delivery as an effective, parameter-compact architecture for retinal vessel segmentation. Code is available at \url{https://github.com/lixincheng-xcl/CGDD-Net}.

---

### How Execution Assumptions Change Short-Horizon Sharpe Rankings: Evidence from a Synthetic Trading Benchmark

Authors: Weicheng Xue
Published: 2026-10-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.05077v1

Backtests of LLM trading agents often assume that every order fills at the closing price. We ask whether this choice changes only reported returns or also the order of the agents. Five prompted LLM signal policies and seven classical baselines trade the same synthetic price paths under six execution settings, from near-ideal fills to latency, spread, participation, and impact stresses. The main experiment contains $2{,}462$ runs with matched decision frequencies and paired market paths. On the compressed two-asset board, agreement between the near-ideal and default-stress rankings falls to Kendall $τ_b=0.21$ in the high-volatility regime, compared with $0.82$ in the calm regime. The seed-bootstrap intervals, $(0.00,0.52)$ and $(0.48,0.94)$, are wide and overlap. On a fixed 11-policy board, agreement rises from 0.24 with two assets to 0.85 with ten; the two-asset point estimate differs substantially from the wider settings we tested. Rank changes are related to turnover, and comparisons with buy-and-hold also depend on how that anchor is initialized. The experiment does not compare LLM trading skill. It shows that, on a short horizon, an execution convention can become part of the benchmark's headline. Execution assumptions and rank stability should be reported alongside returns.

---

### Self-Generated Feedback Destabilizes Test-Time Training: A Causal Decomposition of Long-Horizon Adaptation

Authors: Cheng Luo, Bing Li, Bernard Ghanem
Published: 2026-10-04
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2610.05076v1

Test-time training (TTT) lets a model store information in its weights during inference. When the model learns from its own output, however, each update also changes the model that generates the next training example. Across 128K-token streams, retaining generated-text updates worsens prediction on independent human-written text with three TTT-E2E model configurations (labeled 125M, 760M, and 3B). The same failure occurs when Adam updates Qwen3-4B's existing weights. The same update mechanisms can improve on real text, so writing itself is not the failure. Three matched comparisons trace the causal pathway. Fixed Generation removes over 98% of the damage at 125M and 760M by using a frozen model to generate training chunks. Recorded Replay separates the loss caused by reading degraded text from the additional loss stored by updating on it. A paired one-update comparison then shows the local conflict: an update predicts its source better but new real text worse. This cost grows after Closed Loop adaptation, with a few trajectories accounting for most large failures. Finally, Settlement evaluates the candidate state on independent real text before commitment. It leaves mean endpoint gaps of 0.07 and -0.02 nats at 125M and 760M while retaining real-text adaptation. These results motivate checking prediction on independent evidence before retaining an update.

---

### ReDiffNet: Differential RGB-Infrared Learning for Low-Light UAV Oriented Vehicle Detection

Authors: Qifan Zhang, Ziran Zhou, Ruijie Li, Jincheng Tang, Hao Wang, Qihao Qiao, Chunliu Wang
Published: 2026-10-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.05074v1

Low-light UAV-based RGB-infrared oriented small-vehicle detection is important for nighttime traffic monitoring, emergency response, and urban inspection. Illumination variations, headlight glare, local shadows, and thermal-response degradation cause spatially varying modality reliability, while the small visual extent of vehicles further weakens boundaries, orientation cues, and thermal responses. Accordingly, selecting trustworthy observations based on local modality reliability while further exploiting complementary discriminative information in regions with ambiguous modality preference is key to constructing effective multimodal representations. Based on this insight, we propose ReDiffNet, a reliability-conditioned differential representation network in which modality reliability guides both evidence selection and complementary recovery. Specifically, degradation-aware reliability learning estimates relative spatial reliability, uncertainty-guided differential recovery exploits cross-modal differences to recover complementary cues in ambiguous regions, and reliability-conditioned reconstruction integrates retained and recovered evidence into a unified representation. ReDiffNet achieves 85.3% and 73.9% mAP50 on DroneVehicle and VEDAI, respectively, supporting its effectiveness.

---

### Discrete Action Matching: Learning Stochastic Dynamics from Samples via State Graphs

Authors: Mikhail Persiianov, Alexander Korotin
Published: 2026-10-04
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2610.05071v1

Learning population dynamics from unpaired temporal marginals is an ill-posed inverse problem that requires structural assumptions on the underlying dynamics. We introduce $\textit{Discrete Action Matching}$ (DAM), a finite-state counterpart of Action Matching based on discrete Wasserstein geometry. For a prescribed marginal path and transport geometry, we derive an action-minimization objective for its canonical minimum-kinetic-energy current. Our key observation is that the density dependence of the discrete action reduces to neighboring density ratios. Along an empirical interpolation of the snapshots, DAM first estimates these ratios and then learns an action potential. The learned fields also define a graph-supported Markov sampler. Experiments on controlled synthetic dynamics and real mouse gastrulation data evaluate marginal reconstruction and interpolation. Additional experiments approximate numerical surface-transport paths from paired samples.

---

### Outcome-Guided On-Policy Self-Distillation

Authors: ZheXu Wang, Mao-Lin Luo, Yankun Hong, Zi-Hao Zhou, Bo Ye, Jian Zhao, Xialiang Tong, Min-Ling Zhang, Tong Wei
Published: 2026-10-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.05070v1

On-policy self-distillation (OPSD) provides denser token-level supervision and better computational efficiency than Reinforcement Learning with Verifiable Rewards (RLVR). However, this denser supervision may introduce substantial noise and training instability. Existing improvements often rely on high-variance per-token statistics and introduce extra hyperparameters and trade-offs. Based on the advantage formulation in RLVR, we analyze the OPSD objective from the same perspective, incorporating outcome correctness signals. We find that vanilla OPSD imposes insufficient penalties and excessive rewards on incorrect trajectories because it applies a fixed divergence objective regardless of outcome correctness. Furthermore, the reliability of teacher supervision is associated with both trajectory outcome and the cumulative average teacher entropy along the rollout. Based on these observations, we propose Outcome-Guided On-Policy Self-Distillation (OG-OPSD), which dynamically adapts both the divergence objective and distillation position according to binary outcome rewards and the cumulative average teacher entropy. Extensive experiments show that OG-OPSD consistently improves the performance of vanilla OPSD and multiple strong baselines in mathematical reasoning, multimodal reasoning, and out-of-distribution tasks across Qwen3 models at 1.7B, 4B, and 8B scales, as well as Qwen3-VL-2B.

---

### Usage-Modulated Sentiment Representations in Large Language Models

Authors: Hongfei Du, Jiacheng Shi, Yanfu Zhang, Gang Zhou, Ye Gao
Published: 2026-10-04
Categories: cs.CL
arXiv: https://arxiv.org/abs/2610.05069v1

Prior work suggests that sentiment can often be captured by approximately linear directions in LLM activation spaces, but a single direction may not fully capture sentiment representations. In natural communication, sentiment is shaped not only by polarity but also by usage factors, such as tone and audience adaptation. We test whether these factors systematically modulate sentiment representations beyond a shared sentiment direction. We construct a controlled paired dataset that holds event content fixed while varying sentiment polarity and usage factors, and analyze Llama, Mistral, and Gemma. We identify a shared sentiment direction, remove it, and test the residual structure through erasure and generation-time tone steering. Across models, the shared direction is robust (median cosine 0.953-0.975), yet removing it leaves 0.833-0.909 of the original positive-negative representation-difference norm. The residuals contain compact, reproducible usage-conditioned structure. Targeted erasure weakens held-out usage metrics more than random and label-shuffled controls. On Llama, outputs steered along residualized tone components are preferred in 92.8% of blind target-tone comparisons while preserving the requested sentiment polarity in 98.7% of evaluated outputs.

---

### Salvation Lies Within: Eliciting Inherent Style Transfer in Step-Distilled Diffusion Models

Authors: Shengyin Sun, Yiming Li, Yingzhao Lian, Xing Li, Xingzhi Zhou, Anxin Tian, Zhili Wang, Haoyang Li, Ziqiang Cui, Chen Ma
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05066v2

Adapting step-distilled text-to-image (T2I) models through post-training incurs additional computational costs and affects native few-step generation behavior. This motivates a complementary route beyond style-specific adaptation: drawing on the visual knowledge already encoded in step-distilled T2I models to elicit stylistic capabilities through language. Pursuing this direction requires textual guidance that captures how visual attributes jointly define a style and remain applicable as the depicted content changes. To explore this approach, we introduce StyleForge, a fully automatic, training-free framework that expresses reference styles as reusable rendering instructions. By integrating overall rendering characteristics with local color and lighting behavior, StyleForge organizes visual evidence from reference images into a coherent specification of how the target style should be expressed. The specification is then compiled into textual guidance that can be reused across content prompts, enabling frozen step-distilled T2I models to render different subjects and scenes in the reference style while retaining native few-step generation. Extensive experiments show relative gains of up to 29.47\% in generation quality scores over the strongest baseline, while Pareto analysis indicates that improved stylization is accompanied by strong adherence to the requested content. Code is available at https://github.com/sunshy-1/StyleForge.

---

### Revealing After Overwriting: An Exponential POMDP OPE Lower Bound under History-Dependent Logging

Authors: Youyu Luo, Pengzhan Zhou, Zhida Qin, Jia Wang, Zuotao Fu, Yu Liu, Chao Chen
Published: 2026-10-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.05063v1

Multi-step revealing can make off-policy evaluation tractable under memoryless logging. With history-dependent logging, state decodability and target-relevant evidence can separate. For every horizon $H\ge3$, we construct two exactly realizable POMDPs with four actions, at most four states per layer, a known logger, and a memoryless target. Action overlap, history coverage, and observation-only revealing remain bounded independently of $H$, yet the target values differ by $1/2$ and the KL divergence between the logged laws is $Θ(4^{-(H-1)})$, forcing exponential sample complexity. Logger memory makes states distinguishable, while reset erases the model-distinguishing evidence preserved by the target. A separate construction retains this barrier with common, known observation-only revealing operators. Under action and history coverage, we give a finite-class OPE guarantee using common observable value representations that remain valid at every history. The sample bound depends polynomially on their second-moment cost. In the common-operator construction, the same value direction has constant marginal decoding cost but exponential history-conditioned cost. Finally, on a fixed four-action continuum, we derive matching passive and budgeted readout rates. With one known channel and unit read cost, early reads are optimal. With unknown sensor bias, early reads alone remain exponentially costly. Combining them with post-reset calibration gives sample complexity independent of $H$ when both read types receive fixed positive expected budgets per trajectory.

---

### Why, Where, How: Taxonomy-guided Error Grounding for Code Repair in NL2SQL

Authors: Suchan Lee, Woomin Song, Hwanjo Yu, Sangwoo Mo
Published: 2026-10-04
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.05060v1

SQL queries that large language models write from natural language questions can execute successfully yet produce incorrect results, so execution alone does not reveal what to fix. An error taxonomy says why the query is wrong, but not where to look or how to change it. Existing methods can guide SQL correction through feedback, error reports, or generated plans alongside an unmasked query. We introduce TEG(Taxonomy-guided Error Grounding), which turns a supplied diagnosis into a structured correction input for natural language-to-SQL (NL2SQL) correction. Type-specific rules map each error type to construct classes to reconsider and an edit operation to request. TEG masks the selected constructs in the query when applicable and states that operation in an edit instruction. TEG generates candidate corrections from this input, uses execution feedback to guide candidate selection, and repeats the process one annotation at a time for queries with several errors. On NL2SQL-BUGs, TEG reaches 47.3 single-error execution accuracy and 37.0 overall with Qwen2.5-7B-Instruct. Across the model sizes and thinking modes evaluated in the main comparison, TEG outperforms all evaluated baselines on single-error queries, even when the baselines receive the same error-type annotations. With predicted types, TEG stays above direct LLM correction and ErrorLLM on single-error queries.

---

### A Statistical Inference Framework for PMI Estimation and SGNS Word Embeddings

Authors: Zhongqi Fan
Published: 2026-10-04
Categories: stat.ML, cs.LG, math.ST
arXiv: https://arxiv.org/abs/2610.05058v1

Pointwise Mutual Information (PMI) is a core measure of testing word association, and Skip-gram with Negative Sampling (SGNS) is essentially a method that implicitly factorizes a shifted PMI matrix. However, a systematic and well-rounded characterization of finite-sample uncertainty in PMI estimation remains absent and imperative to venture into. We provide a statistical framework for PMI estimation and its connection to SGNS. We prove consistency, asymptotic unbiasedness, and asymptotic normality of the empirical PMI estimator, derive its variance via the Delta method, and, applying stochastic approximation theory, obtain a variance decomposition for SGNS-based PMI estimation that separates data variance from optimization variance. Simulation experiments validate the Delta method approximation. Real-data experiments on the Brown Corpus (d = 100) reveal that SGNS systematically deviates from the theoretical relationship PMI + log K. The empirical relationship shows an attenuated PMI coefficient, an amplified log K effect, and a positive intercept, indicating systematic bias. Word analogy validation confirms the models are effective. The failure to validate the variance decomposition under low-dimensional conditions does not diminish its theoretical value; rather, it identifies the unbiasedness assumption as the key bottleneck and clarifies the gap between asymptotic theory and practice, providing implications for both practice and theory.

---

### CoDG-Net: Structure-Guided Style Diffusion and Collaborative Learning to Mitigate Catastrophic Forgetting in Medical Image Domain Generalization

Authors: Yucheng Song, Jincan Wang, Haokang Ding, Zhiqiang Tian, Kangxu Fan, Zhifang Liao
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05053v1

Domain Generalization (DG) for medical image segmentation is both highly challenging and critically important. However, existing medical DG methods largely overlook the issue of Catastrophic Forgetting (CF): \textbf{Models often sacrifice their ability to retain source-domain knowledge while pursuing cross-domain robustness.} This can directly threaten diagnostic safety in already-deployed clinical scenarios. To address this, we investigate data augmentation strategies and catastrophic forgetting for medical image DG segmentation. First, we propose a structure-guided style diffusion augmentation method. Constrained by anatomical structure consistency in the frequency domain, this method performs cross-domain diffusion on the amplitude spectrum, generating samples with more diverse and broader style coverage to better support domain generalization. Then, we design a collaborative learning network with a dual-branch interactive architecture (CoDG-Net), together with a novel learning bias-guided strategy that adaptively regulates knowledge transfer at both the layer level and the task level, thereby effectively mitigating catastrophic forgetting on the source domain. Experiments and ablation studies on single-source and multi-source medical DG benchmark datasets demonstrate that CoDG-Net not only outperforms existing state-of-the-art methods in target-domain segmentation performance, but also achieves a lower forgetting rate on the source-domain data. The code is available at: https://github.com/wangprocess/CoDG-Net.

---

### LogSig-SSM: Time-Series Modelling with Multi-Scale Log-Signature Compression for State-Space Models

Authors: Felix Oury, Nicolas Calvo Peiro, Reiko J. Tanaka
Published: 2026-10-04
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2610.05051v1

Time-series data are often sampled irregularly at high frequencies and exhibit long-range dependencies, which makes long-horizon modelling difficult. Continuous-time models such as neural controlled differential equations (NCDEs) and neural rough differential equations (NRDEs) can handle irregular sampling, but they scale poorly to long sequences. Selective state-space models (SSMs) such as Mamba scale linearly with sequence length, but they provide limited recurrent mixing across hidden dimensions within a single block. We propose LogSig-SSM (Log-Signature Compression for State-Space Models), which first compresses long multivariate time series into a shorter sequence of tokens using multi-scale windowed log-signatures, and then processes these tokens with a selective SSM backbone. LogSig-SSM is scalable and robust to irregular sampling, combining log-signature tokens that capture higher-order cross-channel interactions with a selective SSM that models long-range dependencies. The model also admits a continuous-time interpretation as an NCDE/NRDE-style system driven by a log-signature-based input, in which selectivity induces an input-dependent rescaling of the latent dynamics. Across four benchmarks, namely long-sequence classification on UEA, high-frequency physiological regression on PPG-DaLiA, multivariate weather forecasting, and irregularly sampled clinical prediction on PhysioNet Sepsis, LogSig-SSM outperforms or matches strong SSM and continuous-time baselines while training up to $30\times$ faster and using up to $37\times$ less GPU memory than Mamba on the longest sequences.

---

### E$^2$-OPSD: Taming Entropy Overshoot in On-Policy Self-Distillation

Authors: Yifei Liu, Minghao Fang, Xinyu Gu, Chengkai Yao, Mengdi Liu, Tengfei Ma, Jiangbin Zheng, Chang Yu, Zhangyang Gao
Published: 2026-10-04
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2610.05048v5

On-policy self-distillation (OPSD) provides dense token-level supervision without a second model: one network acts as teacher with the reference solution and as student with only the problem. We identify a specific failure mode of this recipe. During training, student token entropy rises past the teacher's and remains elevated, a pattern we call entropy overshoot. We trace it to both sides of distillation. The reference-conditioned teacher is confident along its answer-directed reasoning path, but this confidence transfers poorly to student-generated prefixes, making its supervision overly tied to answer-specific cues rather than reusable reasoning patterns; meanwhile, the forward KL used by OPSD continually diffuses the student's predictive distribution without pulling it back. We introduce E$^2$-OPSD to address both causes. Exemplar-guided teaching replaces the current answer with a retrieved solved neighboring problem, providing transferable reasoning guidance without revealing the destination and better matching student-reachable states. Entropy-aware distillation uses the student-teacher entropy gap to determine the direction and strength of each token's correction. E$^2$-OPSD improves math reasoning by up to 4.3 points in mean@16 over OPSD, while out-of-domain evaluations show gains over the corresponding base models of up to 4.9 points in mean@16 and 5.5 points in pass@8. Despite these gains, E$^2$-OPSD remains simple, requiring no additional forward passes or networks. Code repo: https://github.com/orionfei/E2-OPSD.

---

### AraYoungVoices: A Diverse L1/L2 Corpus of Arabic Child and Adolescent Speech

Authors: Shammur Absar Chowdhury, Zien Sheikh Ali, Houssam Eddine-Othman Lachemat, Hamdy Mubarak
Published: 2026-10-04
Categories: cs.CL, cs.AI, cs.SD
arXiv: https://arxiv.org/abs/2610.05044v1

State-of-the-art ASR systems primarily target native adult speech, leading to substantial performance gaps for children, adolescents, and L2 speakers. We introduce AraYoungVoices, a 151.72-hour Arabic read-speech corpus from 286 speakers aged 7--18, comprising AraKids (7--12) and AraTeens (13--18). The corpus includes 146 native Arabic (L1) and 140 second-language (L2) speakers, with native speakers spanning Egyptian, Gulf, Levantine, and North African dialectal backgrounds and L2 speakers representing diverse linguistic backgrounds across the Americas, Asia, Africa, and Europe. We benchmark four pretrained ASR models under zero-shot and fine-tuned settings using unseen-speaker-$\&$-unseen-prompt (USUP) and unseen-speaker-$\&$-seen-prompt (USSP) evaluations. Results show that L2 speech remains substantially more challenging than L1 speech, with the largest errors observed mainly for younger L2 speakers. Age-specific fine-tuning improves the matched age group, while joint fine-tuning provides a stronger balance across populations. ASR hypotheses are also consistently closer to the standard reading prompt than to the verbatim transcription, particularly for L2 speech, suggesting partial normalization of reading deviations.

---

### CI-JEPA: A Counterfactual Analysis of Latent Representations in Joint-Embedding Predictive Architectures for Self-Supervised Learning

Authors: Mintu Dutta, Ritesh Vyas, Mohendra Roy *
Published: 2026-10-04
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.05043v1

Self-supervised visual representation learning learns useful features without manual annotations during representation training. The image-based joint-embedding predictive architecture (I-JEPA) predicts latent representations of masked image regions, but its objective does not explicitly model responses to specified visual interventions. We introduce CI-JEPA, a counterfactual intervention-aware extension that learns to predict the representation change $ΔZ = Z_{\mathrm{CF}} - Z$ between an original image and a modified counterpart. We assess representation robustness through selective sensitivity: stronger responses to task-relevant semantic changes than to nuisance changes. Experiments on Flowers102 use flower-center occlusion as a candidate semantic intervention and background blur and tint as candidate nuisance interventions. With frozen-encoder linear probing, CI-JEPA achieves a best validation accuracy of 78.14\%, compared with 77.55\% for both the pretrained ViT-B/16 and the I-JEPA baseline, a gain of 0.59 percentage points. The reported mean $L_2$ representation changes are 4.42 for center occlusion, 3.48 for background tint, and 2.83 for background blur. This ordering is consistent with relative semantic selectivity for the evaluated interventions, rather than complete nuisance invariance. The accuracy comparison is complementary and does not establish improved robustness over the baselines. These controlled image modifications provide a framework for studying intervention-induced changes in JEPA representations; they do not establish causal feature discovery or robustness to all visual changes.

---

### Communication Shapes Collective Inference in Self-Adapting LLM Societies: Evidence from Mafia

Authors: Haonan Huang, Joey Xiao
Published: 2026-10-04
Categories: cs.MA, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2610.05041v1

When does communication help a group identify hidden adversaries, and how does its value change as the group adapts? In Mafia, an informed minority hides inside an uninformed majority whose only evidence is open play. The zero-information game, where each day's vote eliminates a random player, is exactly solved and scores every society; matched-casting comparisons between protocols identify the effect of communication. Societies of 8-100 claude-haiku-4-5 agents (7,416 analyzed games, 1.9M model calls) adapt by rewriting and inheriting private strategy notes. Simultaneous broadcast improves adversary identification over silence in all nine compositions tested (8-46 players). Turn-taking removes most of this advantage; its voting landslides are as frequent as broadcast's but land on mafia near chance (1.08x versus 2.53x). At 70 players, agents reading eight statements per day identify adversaries worse than silent ones, and limited talk is worth less than at 46 players. Adaptation is fast but need not help. In their first broadcast games, citizens announce their role far more often than mafia (91% vs. 30%) and first-day votes find mafia at three times chance; within two generations citizens stop announcing and the cue fades, a change the inherited notes carry. In controlled redeployments at 16 players, societies carrying sixty generations of their own notes score below societies with none. Communication shapes both collective inference and the signals it depends on, so a protocol's value must be measured together with the adaptation that changes those signals.

---

### Calibrated Weak Supervision for Post-Harvest Burned-Cropland Mapping Under Label Scarcity

Authors: Raunak Bhagate, Maitri Polisetty, R I Minu
Published: 2026-10-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.05040v1

Mapping post-harvest burned cropland is difficult when fires are small and fragmented and reliable labels are scarce. We developed a calibrated weak-supervision framework for Punjab, India, using Sentinel-2 spectral change, VIIRS active-fire context, and MODIS MCD64A1 as a coarse external calibration and agreement reference. Three pseudo-label recipes, five feature representations, and linear, tree-based, boosted, and neural classifiers were evaluated using nested district-held-out cross-validation over three seeds and five folds. NBR and dNBR were excluded from classifier inputs. The best configuration used the very-strict recipe, a multilayer perceptron, and the full optical feature set (mean Cohen's kappa 0.395, AUROC 0.753, F1 0.747, balanced accuracy 0.703); Random Forest, XGBoost, and LightGBM were practically tied. Higher agreement with held-out pseudo-labels did not establish improved label correctness or independent burned-area accuracy. For deployment, a Random Forest with the very-strict recipe and full optical features was retained. An externally calibrated threshold of 0.60 yielded district-level MODIS agreement of R-squared 0.636, a mapped-to-MODIS burned-area ratio of 1.005, and Spearman correlation of 0.779 with district fire counts. Pixel-level MODIS agreement remained modest (F1 0.205, kappa 0.093). Zero-shot transfer to Haryana was promising (mean kappa 0.641), but Punjab cross-year stability was weak, and Sentinel-1/Sentinel-2 feature concatenation did not improve the optical baseline. Optical observations ended before the seasonal fire context, limiting coverage of late burns. The framework supports district-scale burden assessment and hotspot screening, with limited support for exact scar boundaries or temporally stable annual mapping.

---

### Causal Improvement Graph for Agentic Harness Optimization

Authors: Junjie Zhang, Shunyu Liu, Haoyu Wang, Ting-En Lin, Yongbin Li, Dacheng Tao
Published: 2026-10-04
Categories: cs.CL
arXiv: https://arxiv.org/abs/2610.05039v1

Agentic Harness is the runtime that constructs task context and controls execution flow, thereby shaping overall agent performance. Given a fixed model and external evaluation, automated Harness optimization seeks to improve this runtime through an iterative proposal--evaluation loop to better solve target tasks. Existing meta-harness methods mainly adopt proposer-centric discovery, in which an LLM-based proposer integrates accumulated experimental findings to determine subsequent Harness revisions. This places the burden of maintaining the evolving improvement state on the proposer as history expands and its underlying experimental logic becomes harder to discern. In this paper, we introduce the Causal Improvement Graph (CIG), a graph-governed meta-harness framework that externalizes the evolving improvement state in a persistent graph, allowing prior findings to directly govern subsequent Harness optimization through local proposer operations. CIG grows and links Evidence, Hypothesis, Intervention, and Outcome nodes to represent what was observed, how it may be explained, how to test that explanation, and what the evaluation reveals. Their structural relations preserve how the improvement state changes across iterations, allowing local proposers to build directly on relations among prior findings rather than recover them from raw history. Across various agent tasks, CIG discovers stronger Harnesses than previous meta-harness baselines and remains robust to the choice of task solver and proposer. Structural ablations further support the design of an explicit improvement state with graph-governed evolution.

---

### Optimal Oracle Complexity for Finite-Sum Monotone Inclusions

Authors: Qihao Zhou
Published: 2026-10-04
Categories: math.OC, cs.LG
arXiv: https://arxiv.org/abs/2610.05038v1

We present an oracle-optimal method for finite-sum monotone inclusions under mean-square Lipschitz continuity. Our switching regularization method finds a point $y$ and a certificate $g\in G(y)$ with $(\mathbb{E}\|F(y)+g\|^2)^{1/2}\le\varepsilon$ using $\mathcal{O}(n+\sqrt{n}LR/\varepsilon)$ expected component evaluations and resolvent evaluations. It removes the additive $n\log n$ cost of restarting a variance-reduced solver at every regularization stage by switching to a centered stochastic proximal iteration at regularization strength $L/\sqrt{n}$. Carrying an operator estimate between the remaining stages limits their total cost to $\mathcal{O}(n)$. A matching $Ω(n+\sqrt{n}LR/\varepsilon)$ lower bound holds for randomized linear-span component-oracle algorithms with adaptive stopping and expected query budgets. Thus, for $0<\varepsilon\le LR/2$, our method attains the optimal worst-case expected component complexity in this oracle model, up to universal constants.

---

### SparseCraft: Agentic Hardware-Software Co-Optimization for Sparse Computing

Authors: Rajatabha Chakraborty, M P Samartha, Vedant Pahariya, Priyesh Shukla
Published: 2026-10-04
Categories: cs.AR, cs.AI
arXiv: https://arxiv.org/abs/2610.05037v1

Sparse-accelerator design spaces are usually searched against analytical models, so a design point is admitted on what a model predicts rather than on what the hardware does. SparseCraft closes that gap with a language model inside a closed CHIA loop. In each of 15 iterations the model reads the measured outcome of the previous one and edits the Chisel RTL, the memory configuration and the sparse-kernel schedule of a Gemmini accelerator through MCP tool servers, and no candidate counts until it has been checked for legality, elaborated, simulated cycle-accurately, checked bit-for-bit on every output against a golden reference, and synthesised. The harness turns each measurement into the next work order, a diagnosed bottleneck with matching strategy guidance, the history of tried designs and a score of the model's own prediction, and a second model repairs changes that fail a gate. On a $512 \times 512$ GraphChallenge sparse-DNN layer the loop reaches 2.1x fewer cycles, 9.8x less off-chip traffic and 22.8% less area than the block-sparse Gemmini baseline, with 5.61x higher modelled perf/W and 11.8x lower EDP. The levers span three layers: a schedule that keeps the dense operand resident removes 9.8x of the traffic, a zero-gated MAC and a zero-row skip unit that the model wrote in Chisel cut energy, and resizing the memories cuts area.

---

### Code2Games: Enabling Coding Agents for Gaming World Generation

Authors: Wei Wu, Ziyang Xu, Zeyu Zhang, Yang Zhao, Hao Tang
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05033v1

Generating a high-quality gaming world from a natural-language game intent requires joint reasoning about scene structure, spatial layout, gameplay objectives, interactive entities, and executable gameplay logic. Existing coding agents can generate individual assets, scenes, or scripts, but often struggle to maintain consistency across these components. We propose Code2Games, an agentic framework that builds a structured gaming world upon a base Blender world generated from the same game intent. Code2Games coordinates scene analysis, gameplay planning, constrained gaming-world generation, and gaming-engine customization through a shared scene-gameplay representation with persistent element correspondence. After world generation, Code2Games adapts the generated world to Unreal Engine 5 and employs an execution-guided reconstruction process that uses compilation diagnostics, runtime feedback, and gameplay test results to resolve inconsistencies arising during engine adaptation. To systematically evaluate gaming-world generation, we introduce the GameCode4D benchmark, which comprises ten fixed game prompts spanning different levels of scene and gameplay complexity. We evaluate the generated results across four dimensions: visual quality, interactive fidelity, multimodal artifact quality, and playable-game quality. Experiments demonstrate that, compared with direct gaming-world generation by coding agents and existing baseline methods, Code2Games consistently improves the visual quality and interactive fidelity of generated gaming worlds, as well as the quality of the resulting games after engine adaptation.

---

### When LLMs Sit Above Diagnostic Tools: Unrealized Complementarity in Industrial Fault Diagnosis

Authors: Donghwan Kim
Published: 2026-10-04
Categories: cs.CL, eess.SY
arXiv: https://arxiv.org/abs/2610.05031v2

Large language models are increasingly used as integration layers above specialized tools, but a stronger component does not necessarily produce a stronger combined system. Across five diagnostic datasets (bearing vibration, process monitoring, semiconductor equipment), we study whether an LLM can reliably use external diagnostic information; paired repeat calls separate advice effects from output instability. In all five, conflicting external information overturned initially correct LLM judgments. Among the four datasets with direct integration comparisons, none showed a consistent advantage for implicit LLM integration over the stronger standalone source. On a Tennessee Eastman confirmation set whose protocol was fixed before evaluation, unaided accuracy was 64.67%, implicit LLM-specialist integration 77.43%, and the specialist alone 83.33%. Specialist information improved the LLM by 12.8 points (95% interval 9.7 to 15.9), yet the integrated output stayed 5.9 points below the specialist (95% interval -12.0 to -0.7). A two-source selector oracle reached 92.76%, indicating complementarity that the integrated output did not fully realize. The integrated output missed 140 of 295 specialist corrections (47.5%) but lost 15 of 99 initially correct LLM judgments (15.2%). The deficit remained under prompt and specialist sensitivity analyses. Among CWRU cases solved under both evidence presentations, task-aligned physical evidence yielded lower estimates of susceptibility to incorrect advice in six of seven models (five intervals excluding zero); higher reasoning effort gave no reliable reduction in five models, and a separate four-model TEP analysis gave no clear evidence that it resolves the integration problem. Source quality and integration quality should be evaluated separately: an integration layer should be compared with its stronger standalone component, not only with the unaided LLM.

---

### EVISKILL: Grounding Skill Evolution in Replayable Evidence

Authors: Yan Zhou, Yili Wang, Yiwei Dai, Qinggang Zhang, Xin Wang
Published: 2026-10-04
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.05030v1

Continual skill evolution enables LLM agents to accumulate and refine reusable procedural knowledge from interaction experience without updating model parameters. Its effectiveness depends on determining not only what to change, but also why a change is justified and when it should become persistent guidance. However, existing experience-driven methods can lose the behavioral evidence and task contexts supporting edits. Moreover, a global validation outcome provides an incomplete judgment of its constituent changes: locally supported corrections may be discarded with a rejected revision, while evidence may require further experience to inform useful updates. To this end, we introduce EVISKILL, an evidence-driven framework that organizes execution observations into Replayable Evidence Cards and synthesizes edits with explicit links to their supporting contexts. Targeted replay verifies these edits through re-execution and provides feedback for correction. Across epochs, EVISKILL preserves evidence and provisionally retains supported edits for further refinement, while global validation governs their incorporation into the final skill. Experiments on three interactive benchmarks across six LLM backbones demonstrate the effectiveness of this approach.

---

### SPACE-CLIPv2: Decoding Local Geometry from Frozen CLIP for Monocular Depth Estimation

Authors: Hyun Song, Taewan Cho, Kangmin Kim, Andrew Jaeyong Choi
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05029v1

Vision-language foundation models such as CLIP provide strong semantic representations, but their patch tokens are not directly optimized for dense metric geometry. SPACE-CLIP showed that frozen CLIP features can support monocular depth estimation through layer-group feature fusion, yet it leaves open how neighboring CLIP tokens should be combined to recover fine local structure. We present SPACE-CLIPv2, a frozen-backbone depth decoder that aggregates fixed local neighborhoods in CLIP token space. At selected decoder stages, the model samples a fixed token stencil, predicts aggregation weights, and injects the resulting response through a gated residual update. A token-space high-pass branch further preserves shallow local contrast. On NYU Depth V2, SPACE-CLIPv2 improves over a matched SPACE-CLIP baseline, while five-seed experiments consistently favor fixed over learned-offset sampling. Zero-shot iBims-1 evaluation further improves boundary and planar-geometry measures. These results support constrained local token aggregation as a practical mechanism for decoding geometry from frozen CLIP representations.

---

### GeoBridge-VLA: Geometry-Aware Residual Adaptation for Vision-Language-Action Models

Authors: Hyun Song, Kangmin Kim, Loren Jinsoo Um, Minhui Han, Jaehyeok Park, Taewan Cho, Andrew Jaeyong Choi
Published: 2026-10-04
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2610.05026v1

Vision-language-action (VLA) models encode semantic information from vision-language pretraining, but manipulation also requires precise spatial reasoning. We present GeoBridge-VLA, a two-stage method for learning geometric features from a pretrained VLA's frozen visual encoder and using them for action prediction. Stage I trains a feature bridge and geometry decoder with depth supervision. Stage II freezes these modules and trains a gated residual interface together with the action-side projections and action expert. The residual augments the existing visual tokens without adding a second image encoder or increasing the token count. Deployment requires RGB, robot state, and language, but no depth observations. Under matched evaluation conditions, GeoBridge-VLA achieves 70.9% success on LIBERO, compared with 60.0% for SmolVLA. Disabling the residual in the same trained checkpoint reduces success from 70.90% to 69.85%, with mixed effects across suites. On a physical ROBOTIS OMY robot, GeoBridge-VLA succeeds in 148 of 200 trials (74.0%) across four tasks, compared with 108 of 200 (54.0%) for SmolVLA.

---

### Triggering Generalist Reasoning via Predictive Uncertainty for Dual-System VLA

Authors: Hyemin Yang, Wooseong Jeong, Giwon Lee, Kuk-Jin Yoon
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05025v2

Dual-system Vision-Language-Action (VLA) models improve real-time robotic control by pairing a slow, reasoning-capable generalist with a fast specialist action expert. However, existing methods invoke the generalist at a fixed frequency, ignoring the fact that decision-making complexity varies throughout a rollout. This static strategy wastes computation in easy phases and can delay renewed reasoning when the scene changes unexpectedly. We propose TUD (Triggering generalist reasoning via predictive Uncertainty for Dual-system VLA), an adaptive inference framework that selectively skips unnecessary generalist calls. TUD measures the cross-step dispersion of action re-predictions at the upcoming chunk slot under the cached generalist context, as a predictive uncertainty signal. This signal captures how much the future action plan shifts as new observations arrive and is computed from forwards the architecture already runs, requiring neither manual phase labels nor an auxiliary uncertainty model. On VLA-Arena, it achieves a higher success rate at matched call budgets than alternative uncertainty baselines while maintaining low wall-clock overhead, and more consistently separates successful from failed rollouts. Also, TUD finds a more favorable cost-success trade-off than non-adaptive baselines, tracing an entire operating curve as a single threshold is varied, and substantially reduces VLM calls at matched success rate. The same trade-off appears in our real-robot experiments, where TUD cuts generalist calls by 75% relative to the strongest fixed-interval baseline while achieving an even higher success rate. Our results suggest that predictive uncertainty provides a practical criterion for adaptive reasoning in efficient VLA control.

---

### LightVLN: Efficient Aerial Vision-and-Language Navigation with Compact Memory and History-Guided Local Aggregation

Authors: Yiming Zhao, Tianshun Li, Jingle He, Ruonan Chai, Xinhu Zheng
Published: 2026-10-04
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2610.05024v1

Aerial vision-and-language navigation (VLN) enables unmanned aerial vehicles to execute long-horizon natural-language instructions from visual observations in complex three-dimensional environments. However, recent aerial VLN models often rely on large-scale vision-language backbones and dense visual histories, imposing substantial computation and memory costs that hinder onboard deployment. We propose LightVLN, a lightweight history-aware aerial VLN framework that combines a compact 0.5B language backbone with compact representations of both historical and current observations. LightVLN compresses each historical frame into a single token using visual features already computed by the policy. It further introduces history- and instruction-conditioned local aggregation to reduce the current observation from 256 to 32 visual tokens while preserving navigation-relevant spatial information. With up to 16 historical frames, the policy uses at most 48 observation-derived tokens. On the public OpenFly dataset, LightVLN achieves 50.93% Test-Seen and 36.14% Test-Unseen success rates (SR), outperforming the evaluated 7B language-backbone baselines on most reported metrics. It also achieves 25.83% SR on AerialVLN-S Val-Seen. In a reconstructed unseen campus, we deploy LightVLN on a DJI M350 RTK with an external Jetson Orin NX 16 GB for closed-loop onboard-compute real-to-sim hardware-in-the-loop (HIL) evaluation, achieving 14.61 Hz model inference and 11.13 Hz end-to-end decision updates. These results demonstrate the effectiveness and efficiency of LightVLN for aerial navigation.

---

### Look Where You Say You're Looking: Self-Grounded Attention for Visual Reasoning

Authors: Uri Berger, Gal Chechik, Gal Dalal
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05023v1

We introduce Self-Saliency, a method for training Vision-Language Models (VLMs) to increase the alignment between their visual attention and the image regions mentioned in their reasoning. Self-Saliency uses a grounding model to localize the objects mentioned in each reasoning step and treats the resulting areas as supervision for the model's visual attention. Previous work on steering visual attention determines target image regions based solely on the image and question. In contrast, we show that conditioning the target regions on the model's generated reasoning improves downstream performance. For proper evaluation, we build a unified, broad suite of 25 visual reasoning benchmarks, where we reproduce the results of previous methods. We find that Self-Saliency significantly outperforms both prior attention-steering methods and baselines that ground image-level text, achieving both a better average rank and a better mean score. Post-training analysis shows that the model primarily adapts its reasoning text to existing attention patterns, producing shorter steps that refer to larger regions. Nevertheless, when controlling for generated text, attention to grounded regions increases significantly across the relevant layer. Finally, we identify a consistent geometric bias in VLM visual attention toward the image border. However, our ablations show that Self-Saliency's gains cannot be explained by simply aligning attention with the center of the image, highlighting the importance of aligning visual attention with the regions mentioned in the model's reasoning.

---

### Operational Abstractions of Neural Network Concepts via Topological Representations

Authors: Mathieu Pont, Christoph Garth
Published: 2026-10-04
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2610.05017v1

Concept-based methods provide a semantic level for interpreting and manipulating learned representations, but existing editing approaches are typically specialized to particular interventions and do not provide a common and editable representation of concept organization. To achieve this, we introduce Topological Concept Representations (TCR), a post-hoc operational abstraction that jointly characterizes the concepts encoded in a learned representation and their relationships. TCR constructs an intermediate concept space from concept recoverability and interaction scores, and compactly encodes its organization through topology. Interventions are expressed through modifications of this abstraction that are propagated back to the underlying learned representation. This allows different concept-level operations to share the same optimization framework and separates the desired concept organization from the mechanism to achieve it. We establish stability and reparameterization-invariance properties of TCR and its connections to existing concept-editing formulations. We use TCR to disentangle concepts as a preprocessing step for existing erasure methods, improving worst-group accuracy by 21.89 on average at comparable concept leakage. We further use TCR to transfer concepts from teacher to student models, improving concept recoverability by up to 5.54 while also improving or maintaining competitive test top-1 accuracy.

---

### MetaKernelBench: Measuring GPU Kernel Knowledge Transfer Beyond Code

Authors: Xueyi Chen, Shiyu Liu, Xin Jin, Yuhua Zheng, Xin Li, Haolei Bai, Junhan Zhu, Huan Wang
Published: 2026-10-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.05014v1

Recent GPU kernel optimization agents retain what they learn in knowledge bases or as distilled skills. Kernel benchmarks score each attempt's implementation for correctness and speed but leave the reuse value of retained experience unmeasured. We introduce MetaKernelBench, which measures whether experience distilled from an attempt in one kernel domain-specific language (DSL) improves a fresh attempt at the same problem in another. Its 74 problems are fused subgraphs in six families, each posed as a pair of CuTe DSL and TIRx variants that differ only in the DSL. The agent first attempts each variant solo and is instructed to distill what it learns into a natural-language skill, which is transferred whether or not the source attempt passes verification. The skill is the only extra input to a skill-conditioned attempt by the same model in the other DSL. We compare each skill-conditioned attempt with the solo attempt on the same variant under matched per-attempt budgets, scoring correctness and end-to-end runtime. Across six models and both directions, paired lift over solo attempts ranges from -19% to +29%. Four models gain in both directions, yet regressions occur on 16% to 45% of problems in every model and direction. Outcomes follow the source attempt's result relative to the target's solo attempt rather than source success alone, improving in 71% of comparisons when the source stands above and regressing in 54% when it stands below. MetaKernelBench complements implementation-quality metrics by measuring same-problem cross-DSL kernel knowledge transfer.

---

### EMBER-Bench: Benchmarking Cross-Event Causal Memory in Long-Horizon Embodied Tasks

Authors: Aoyang Cai, Boning Zhao, Shaoxuan Xie, Dahui Gao, Huan Yang, Zhongyuan Wang, Zhiwei Yu, Guocai Yao
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05013v1

Lifelong physical agents must reason over extended interactions where past events continue to shape the world long after they disappear from view. Beyond recalling what happened, agents must infer how history changes the current state and constrains future actions. Yet existing embodied and video-memory benchmarks largely focus on historical retrieval and summary, leaving such history-dependent causal reasoning underexplored. We introduce EMBER-Bench, an egocentric benchmark for cross-event causal reasoning in long-horizon embodied tasks, for which we newly created the task design, video recording, and data annotation. It contains 189 household tasks and 699 QA pairs, spanning task progress, failure recovery, external interventions, and compound long-horizon tasks with distant dependencies and prerequisites, with fine-grained event and causal-chain annotations. EMBER-Bench evaluates reasoning in both directions: next-action prediction selects the next action from history, and causal traceback, given that action, identifies the historical event that makes it necessary. Input ablations that add action logs or privileged cause-and-consequence annotations to the video indicate which kind of historical information models fail to use. Among the 16 evaluated models, the highest overall accuracy is 61.2%, compared with a mean of 98.3% across two human evaluators. At paired decision points, correct traceback is not associated with correct next-action prediction. Adding action logs yields a gain of 1.6 points, whereas cause-and-consequence annotations yield an additional gain of 13.0 points on top of that. These results suggest that extracting causal information from past events and converting it into constraints on current actions remains a key difficulty for long-horizon embodied agents. Project Page: https://zhaoalexgoat.github.io/EMBER-Bench/

---

### FreeLoc: Online Floorplan Localization via Diffusion-Aided Pose Refinement

Authors: Haocheng Peng, Boyang Zhou, Jiarui Hu, Xiyue Guo, Ziyang Zhang, Boming Zhao, Yifan Gao, Xiao Li, Hujun Bao, Zhaopeng Cui
Published: 2026-10-04
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2610.05011v2

Floorplans provide compact and widely available geometric maps for indoor localization, but existing high-performing floorplan-based methods still convert them into dense scene-specific offline databases, tying accuracy, storage, and runtime to the sampling resolution of the discretized pose space. We present FreeLoc, an online RGB-based floorplan localization framework that treats the floorplan as a directly queryable geometric map. FreeLoc introduces an efficient online geometric querying and diffusion-aided refinement scheme, which retrieves plausible pose anchors through on-the-fly floorplan ray querying and refines them into accurate continuous pose estimates. For sequential localization, FreeLoc develops an online likelihood construction strategy that bridges single-frame localization and probabilistic temporal fusion by constructing likelihoods from coarse-sampled candidates and refined pose hypotheses, enabling histogram-filter-based temporal fusion without offline databases. Experiments demonstrate real-time online inference and state-of-the-art performance in both single-frame and sequential localization, while real-world results validate practical deployability in indoor robotic localization scenarios.

---

### PortraitAes: Intent-Conditioned Structured Portrait Aesthetic Assessment

Authors: Junzhou Xie, Haozhong Xiong, Xunyun Tian, Kaile Du, Tianchen Yu, Qiang Li, Wei Liu, Jiaming Liu, Ruihua Huang, Yang Shi, Guangcan Liu
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05010v2

Portrait aesthetic assessment assigns comparable scores according to how effectively human-centered images fulfill their photographic intent. These scores support data filtering, candidate selection, and preference modeling in image-generation pipelines. Existing methods typically predict a single aesthetic score or use general-purpose MLLMs without conditioning on photographic intent. This omission matters because the same blur, pose, lighting, or framing choice may serve one photographic intent but undermine another. These models thus learn context-agnostic aesthetic priors and yield inconsistent, inaccurate, misleading judgments for portraits with distinct photographic objectives. We introduce PortraitAes-Bench, an 11K-scale benchmark that decomposes this task into intent-conditioned subjudgments. Expert-authored rubrics define nine photographic intents, six first-level dimensions, and 22 secondary criteria. They support a structured pipeline for intent routing, specialist assessment, verification, and score fusion. Following this structure, we train PortraitAes with multi-task supervision. We then improve score comparability through Gaussian score calibration and within-dimension cross-image ranking. On the standard benchmark, PortraitAes achieves a Pearson correlation of 0.924 and a Spearman rank correlation of 0.934. On the hard-case set, its Pearson correlation is 0.829 and its Spearman rank correlation is 0.795. Across both sets, PortraitAes outperforms the evaluated general-purpose MLLMs and specialized aesthetic baselines.

---

### Physics-Augmented Graph Transformers for Patch-Antenna Forward and Inverse Design

Authors: Avi Epstein, Snir Nehemia, Haim Suchowski, Lior Wolf
Published: 2026-10-04
Categories: cs.LG, physics.comp-ph
arXiv: https://arxiv.org/abs/2610.05004v1

Full-wave electromagnetic (EM) simulation enables accurate patch-antenna analysis but is computationally expensive for large-scale forward prediction and inverse design. We present a mesh-native, physics-augmented graph-learning framework that treats radiation-pattern prediction as signal reconstruction on an irregular surface mesh. For the forward problem, a GPS graph transformer is trained with Physics-Augmented Intermediate Supervision (PAIS), an auxiliary node-level objective that predicts complex surface currents, the physical intermediate linking geometry to radiation. PAIS improves multiple GNN backbones at no inference-time cost, while shuffled-current and non-physical controls show the gain comes from physical correspondence. Direction-conditioned decoding and a differentiable radiation-integral consistency loss further exploit this structure. On an 80,000-sample CST benchmark, GPS+PAIS reaches MSE 0.17 / PSNR 19.67, generalizes to a PCA split, and transfers zero-shot to canonical patches. For inverse design, surrogate-filtered diffusion beats nearest-neighbor retrieval by 32% relative MSE.

---

### How corner is a corner case? Percentile control for highway scenario generation

Authors: Jiaxi Liu, Hang Zhou, Hangyu Li, Yifan Wang, Keke Long, Chengyuan Ma, Bin Ran, Xiaopeng Li
Published: 2026-10-04
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.05003v1

Generating corner-case scenarios with appropriate adversity in a simulation environment is critical for testing an autonomous vehicle (AV) software stack's safety performance before deployment. Existing autonomous-driving scenario generators can enforce specific behavior, adversity, or feasibility conditions, but they provide limited control over how extreme a generated scenario is relative to plausible futures in the same traffic context. This study represents the adversity of a generated scenario as its percentile in the conditional distribution of future risk given the observed history. This view supports calibrated answers to two questions: how "corner" a generated corner-case scenario is and how its "cornerness" can be fine-tuned. To this end, we formulate history-conditioned risk-percentile requests and learn a reference risk distribution that maps each requested percentile to a physical risk target. We then use a percentile-conditioned joint diffusion model with sampling-time risk guidance to generate multi-agent futures, together with a reference-based criterion for evaluating percentile realization. Experiments use the minimum post-encroachment time (PET) between the ego and its surrounding vehicles as the risk surrogate on highD. On the primary evaluation set, our method realizes 1,422 of 1,440 requests within a 0.05 percentile tolerance (98.75%), with mean percentile error 0.00673 and PET-target error 0.00991 seconds. The resulting interface connects context-relative risk specification, physical realization, and evaluation through a common risk scale. Project website and videos of generated scenarios are available at https://hhj233.github.io/CornerPercentile/.

---

### VisualErase: Dual-Branch Visual Trajectory Redirection for Robust Concept Erasure in Text-to-Image Diffusion Models

Authors: Qianlong Xiang, Miao Zhang, Kun Wang, Yupeng Hu, Junhui Hou, Liqiang Nie
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.05000v1

Concept erasure is essential for the safe deployment of text-to-image diffusion models, as they may reproduce harmful, copyrighted, or privacy-sensitive content learned from unconstrained large-scale data. Existing methods typically erase unwanted concepts while preserving general generation capability by redirecting target-related text-to-image mappings. However, recent studies show that erased models may still retain visual generative trajectories of target concepts, leaving them vulnerable to adversarial recovery attacks and revealing a fundamental gap between redirecting text-to-image mappings and truly removing visual knowledge. To bridge this gap, we propose VisualErase, a new paradigm that redirects concept-bearing visual generative trajectories toward explicitly defined concept-removed outcomes. To enable this redirection, we use structure-preserving image editing to construct content-aligned, concept-removed counterparts for source images, providing explicit visual endpoints that retain non-target content. We then derive a denoising target from each source-to-counterpart pair and use a dual-branch redirection loss to align both text-conditioned and unconditional predictions with this target, since conditional supervision alone does not explicitly constrain generation without textual guidance. To mitigate the adverse effects of concept erasure on non-target generation, we jointly optimize the redirection loss with a counterpart retention loss that matches denoising predictions from the frozen pretrained model. Across style, celebrity, and nudity erasure, VisualErase limits the maximum attack success rate over seven attacks to 0%, 8%, and 0.1%, respectively, while retaining general generation quality. These results highlight the importance of visual trajectory redirection for robust concept erasure beyond text-to-image mappings alone.

---

### Beyond Overparameterization: Provable Learning of Input-Convex Multi-Layer Polynomial Networks with Active Queries

Authors: Jinqi Tang, Qian Chen, Shihong Ding, Cong Fang
Published: 2026-10-04
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2610.04999v1

The theoretical understanding of multi-layer neural networks is largely confined to overparameterized settings, which obscure parameter identifiability and incur high sample complexity. Neural tangent kernel (NTK) provides a general theory for wide networks, but does not offer efficient sample-complexity guarantees. Recent feature-learning results go beyond kernel methods for single-neuron, multi-index, and hierarchical targets. However, the analysis is often restricted to shallow or specific architectures and to the overparameterized regime. We break this paradigm to achieve parameter-level recovery of deep target networks, albeit by using active data queries. Specifically, we study $L$-layer polynomial networks with even degree-$k$ monomial activations and nonnegative higher-layer weights. This structure makes the target network input-convex, while the optimization landscape remains highly nonconvex with respect to the parameters. Leveraging input convexity and active queries, we propose \textbf{ASPIRE} (\textbf{A}ctive \textbf{S}am\textbf{P}ling for \textbf{I}terative \textbf{R}ecovery via \textbf{E}igendirections), a layerwise sampling-based diagonalization algorithm that recovers all network parameters to $δ$-accuracy with sample complexity $ \widetilde O_{k,L}\left(d^{L^2+O(L)}δ^{-2e}\right) $ in polynomial time. To our knowledge, this is the \emph{first} parameter-recovery guarantee for deep target networks whose exponent grows only polynomially with depth, as well as the \emph{first} justification for the effectiveness of using high-quality data in neural network training, with a remarkably \emph{exponential} separation.

---

### EmoRSS: Mitigating Emotion-Induced Over-Refusal in Large Language Models

Authors: Shuyi Miao, Yaojin Ma, Chenhang Cui, Xiaohao Liu, Dang Jisheng, Shengda Zhuo, Fei Shen, Tat-Seng Chua
Published: 2026-10-04
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.04998v1

Emotional expression can influence the safety decisions of large language models (LLMs), offering a potential avenue for improving safety alignment. Existing studies have mainly focused on how emotional expressions facilitate attacks under harmful requests, while overlooking their effects on benign requests. We find that emotional expression can also systematically increase refusal tendencies on benign requests, leading to unnecessary over-refusal. Based on this observation, we propose emotion-guided refusal subspace steering (EmoRSS), an activation-steering method that mitigates emotion-induced over-refusal while preserving refusal behaviour on harmful requests. Specifically, we first identify a refusal-sensitive layer using layer-wise linear probes and construct a refusal subspace from sparse autoencoder (SAE) features aligned with the probe direction. Next, we use paired regular and emotional requests with the same queries to estimate the mean activation shift in the features defining the refusal subspace. Finally, we decode this shift into an activation intervention vector and apply it in the reverse refusal direction during inference, without updating the backbone parameters. Experiments on two LLMs show that, when requests contain emotional expressions, our method achieves a more favourable trade-off between refusing harmful requests and answering benign ones than prior over-refusal mitigation baselines, while better preserving general task performance.

---

### Pessimistic Minimax Learning for Public-Private Information Games under Unilateral Coverage

Authors: Shuze Daniel Liu, Claire Chen, Jiuqi Wang, David Simchi-Levi
Published: 2026-10-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.04997v1

We study offline learning in two-player zero-sum contextual games with public and private information, motivated by strategic settings such as auctions and negotiations with private valuations. We introduce unilateral prescriptive concentrability and show that asymmetric information can change offline coverage through its effect on equilibrium behavior. For finite state-action spaces, we develop a pessimistic algorithm with an $\tilde{O}(1/\sqrt{n})$ exploitability rate, matching the standard sample-size dependence for fully observed minimax games. We further develop a pessimistic policy mirror descent framework, PPA-PMD, for general function approximation and obtain a unified $\tilde{O}(1/\sqrt{n} + 1/\sqrt{T})$ exploitability rate with no-regret actor updates. Together, these results provide the first theoretical framework for offline equilibrium learning under public-private information constraints.

---

### CIPO: Counterfactual Imagination Policy Optimization for Adaptive Tool Granularity Selection

Authors: Yu Li, Yunlu Wan, Zijian Zhu, Han Luo, Chao Ren, Long-Fei Li, Lei Feng
Published: 2026-10-04
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.04991v1

Large language model (LLM) agents solve complex tasks through multi-step interactions with external tools. These interactions often contain recurring local tool sequences. Treating such sequences as composite "Skills" can shorten tool-use trajectories and reduce repeated low-level decisions. However, when atomic tools and composite skills coexist, skill use becomes a policy problem: the agent must decide whether the current state requires atomic fine control or skill-level abstraction. In this paper, we argue that effective skill use should be studied as adaptive tool granularity selection. The most direct training signal for this problem is to compare the consequences of atomic and skill choices available from the same state. Based on this view, we propose CIPO, a Counterfactual Imagination Policy Optimization framework for adaptive tool granularity. CIPO constructs executable skills through budget-constrained mining of successful tool-use trajectories and trains granularity decisions with counterfactual branch rollouts. For each base rollout, CIPO branches at the first eligible granularity decision and replaces the chosen action with a feasible atomic or skill alternative. The paired outcome difference serves as a supplementary reward for policy optimization. Experiments across multiple benchmarks and model backbones show that CIPO improves task success and decision efficiency over baselines. Further analyses show that CIPO learns effective skill use by improving the choice between atomic tools and composite skills based on the current state, without simply increasing skill frequency.

---

### When Is a Graph a Covariance? Bayesian Residual Propagation under Covariance Uncertainty

Authors: Robert Richardson
Published: 2026-10-04
Categories: math.ST, stat.ML
arXiv: https://arxiv.org/abs/2610.04989v1

Observing a prediction error at one node can help correct predictions elsewhere, but the benefit depends on the residual dependence between nodes. A graph suggests where that dependence might occur, yet does not establish its sign or strength. We develop a Bayesian model of residual covariance to determine how revealed errors should update a fixed feature-based predictor. For a fixed set of revealed nodes, we derive an exact identity for the change in expected squared error under linear residual propagation. The identity characterizes the optimal linear update and expresses the excess error of any other update as an exact quadratic. Under covariance uncertainty, the Bayes-optimal fixed linear update uses the posterior mean covariance. We average over four positive-semidefinite covariance families representing no dependence, positive dependence, negative dependence, and dependence whose sign can alternate with graph distance. Across eight graph benchmarks, four show posterior support within the distance-profile family for negative covariance between neighbors and alternating signs with distance. Among the evaluated methods, ours achieves the smallest worst-case recall gap to the best method on each graph. The method improves squared error over residual propagation on five of eight graphs, but performs substantially worse on one dense heterophilous graph. In these experiments, selecting the highest-posterior family performs similarly to model averaging.

---

### How Long, Not How Close: A Learned Temporal Metric for Planning in Latent World Models

Authors: Lama Moukheiber, Haotian Xue, Yongxin Chen
Published: 2026-10-04
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.04988v2

Latent world models plan by rolling a frozen predictor forward under candidate action sequences and ranking the candidates by the latent distance between their imagined end state and the goal. However, this ranking breaks down when the goal lies several plans away, because the latent distance measures how closely an end state resembles the goal rather than how far it remains from reaching it. To address this, we propose TEMPO, a temporal-distance planning objective that leaves the world model untouched, learns only from the recorded trajectories already used to train it, and adds negligible cost to the planner's search. TEMPO learns a small map of the frozen latent in which the distance between two states of an episode reflects the number of environment steps between them, and blends this distance into the planner's cost. It requires no rewards, policies or success labels and, being a cost rather than a model, applies to frozen world models with one latent vector per state that plan by a latent distance. We evaluate TEMPO on eleven simulated environments (e.g., maze navigation, tabletop pushing, robotic arm control and three-dimensional manipulation) with the LeWM and PLDM planners. With a small MLP that adds at most 0.3% to a plan's arithmetic, TEMPO improves both planners at every goal distance, including the one-plan setting of their evaluations, raises LeWM from 36% to 99% on TwoRoom three plans from the goal, and remains competitive on a broad range of 2D and 3D navigation, reaching and manipulation tasks.

---

### Hidden Risks of Jev: An Empirical Study of Security, Privacy, and Dual Use

Authors: Shang Wang, Tianqing Zhu, Huajie Chen, Jiayang Li, Meng Yang, Bo Liu
Published: 2026-10-04
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2610.04985v1

Jev turns natural-language questions into typed answers and probabilities with low latency and cost, enabling applications to route requests and select tools. While this interface allows Jev to integrate naturally into application workflows as a decision layer, the security and privacy implications of this emerging use remain largely unexplored. To address this gap, we conduct the first systematic study of these implications using the official Jev API and NanoJev, a local model with controllable training data and updates, focusing on three research questions: (1) What security threats arise when Jev is deployed as an application decision layer? (2) What private information can Jev reveal despite returning constrained typed outputs? (3) How can Jev's general-purpose decision capability be used for beneficial purposes or misused? Jev's decisions depend on application state and may be influenced by user-provided inputs. We therefore adapt prompt injection and adversarial suffixes to manipulate its decisions. Open-source Jev distribution and updates introduce supply-chain risks, which we examine by implanting backdoors in NanoJev through training data poisoning. Since Jev's outputs reflect both application state and information learned during training, we further adapt membership, private attribute, and internal knowledge inference attacks to recover sensitive information despite its constrained output format. Finally, Jev can serve as a general-purpose decision oracle for defensive and malicious workflows. We examine this dual use through four detection tasks covering prompt injection, jailbreak inputs, harmful content, and AI-generated text, alongside misuse scenarios involving jailbreak and model extraction. Our empirical evaluation shows that Jev remains vulnerable to the examined security and privacy threats, while its decision capability can support beneficial and malicious uses.

---

### TeleGen: Improving LLM-Based Web Application Generation via Runtime Telemetry

Authors: Yujia Luo, Haonan Zhang, Jiasi Shen, Zishuo Ding, Weiyi Shang
Published: 2026-10-04
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2610.04981v1

Large language models can generate runnable web applications from natural-language requirements, but many generated applications still fail interactive tasks. Existing generate-execute-repair pipelines execute the generated application and use task outcomes or error messages to guide code revision. However, this feedback often misses the runtime behavior between a browser action and the final task outcome, making interaction-level failures difficult to diagnose. Therefore, we propose TeleGen, an observability-enhanced framework for LLM-based web application generation. TeleGen instruments generated applications, collects runtime telemetry during task execution, and compresses raw telemetry logs into concise briefs for repair. We evaluate TeleGen on WebGen-Bench and Web-Bench. On WebGen-Bench, TeleGen improves task success from 67.7% with repair without telemetry to 76.2%, an increase of 8.5 percentage points. On Web-Bench, it improves cumulative Pass@2 from 21.7% to 29.8%. Ablation results show that runtime telemetry provides a useful diagnostic signal, while telemetry briefs make this signal more effective and less costly to use. Further analysis shows that telemetry is especially helpful for failures involving hidden execution paths, such as navigation, form workflows, and frontend-backend coordination.

---

### Geometry-Aware Preference Optimization for Text-to-Image Diffusion Models

Authors: Lei Wang, Zhen Wang, Yuexiang Xie, Yaliang Li
Published: 2026-10-04
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.04980v1

Preference alignment has become a standard practice for text-to-image diffusion models. Direct Preference Optimization (DPO) simplifies this process by eliminating explicit reward modeling. Its diffusion variant, Diffusion-DPO, has become a widely adopted baseline. Diffusion-DPO essentially encourages the likelihood of preferred samples while suppressing dispreferred ones. In this paper, we revisit DPO-style alignment methods for diffusion models from the perspective of the manifold hypothesis. Under this view, natural images concentrate near a low-dimensional manifold embedded in the high-dimensional ambient space, whereas DPO directly optimizes preference distributions in the full space without accounting for this geometric structure. This creates a mismatch in the optimization dynamics: it suppresses geometry-preserving tangential updates, while insufficiently restricting hazardous normal-direction updates. This mismatch gradually degrades image quality and diversity. To address this issue, we propose Anisotropic Geometry-Aware Preference Optimization (APO), which replaces the uniform Euclidean treatment of prediction errors with a geometry-aware anisotropic metric derived from the reference model. Concretely, APO adaptively strengthens regularization in directions where the reference denoising function is highly sensitive, while relaxing constraints in directions that permit safe semantic adjustment. This recalibrates preference optimization according to the local manifold geometry, and maintains the original manifold structure. Experiments show that APO achieves strong performance and an average win rate exceeding 60\% against various existing alignment methods across diverse benchmarks. It requires significantly fewer training steps than prior methods, and preserves generation diversity throughout training.

---