# Research Papers — 2026-09 · bundle page 8 of 266 > Full-text markdown of papers #351–400 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/8 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/7.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/9.md) ### Ready2Blend: From Natural-Language Instructions to Composable Alignment Prompts Authors: Jeesu Jung, Hwan Chang, Juseon Do, Jeonghwan Choi, Jinho Choo, Sungwoo Nam, S. K. Hong, Hwanjun Song Published: 2026-09-30 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.39365v1 Continual alignment requires LLMs to adapt to new requirements without forgetting previously acquired behaviors. Natural-language instructions are flexible and composable but offer only indirect control, whereas post-training provides stronger adaptation at the cost of repeated parameter updates. We introduce Ready2Blend, which combines the flexibility of natural language with learned alignment. AlignFormer maps each requirement to a fixed-length alignment prompt stored in a modular prompt bank, while the backbone and prior prompts remain frozen. Composability regularization transfers the semantic geometry of textual requirements into prompt space, enabling inference-time blending and reweighting. Across two practical continual alignment settings, Ready2Blend is the only frozen-backbone method that matches post-training-based alignment methods, reaching $93.1$-$98.5\%$ of a joint-training reference with competitive retention, while requiring only a few prompt tokens and up to $4.3\times$ less training time. Its modular design further enables weighted personalization and order-free composition without retraining. Code will be released upon acceptance. --- ### Rethinking Multi-Image Re-Representation in Multi-Image Understanding Authors: Gengyuan Zhang, Xiao Han, Xinyu Xie, Tong Liu, Volker Tresp Published: 2026-09-30 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.39363v1 Multi-image understanding requires MLLMs not only to recognise the content of individual images, but also to organise visual evidence distributed across them. We study this problem through multi-image re-representation, viewing prompted Chain-of-Thought reasoning and agentic visual tool use as different ways of re-organising visual evidence during reasoning. We introduce Mosaic, a general-purpose multi-image visual harness that enables an MLLM to actively construct visual intermediates with ten composable image operations. We compare five re-representation settings on existing multi-image benchmarks and on MosaicBench, a new grounding-focused benchmark for fine-grained multi-image understanding. Our experiments show that the relative benefits of textual and visual re-representation are strongly task-dependent. Visual re-representation is particularly effective for tasks requiring precise visual evidence, including hypothesis testing, precision comparison, and orientation-sensitive reasoning, while tasks dominated by higher-level semantic content show smaller or less consistent gains. Building on this finding, we train MosaicAgent-8B to use Mosaic with reinforcement learning using only accuracy and format rewards. Without demonstration trajectories or rewards for specific tool-use, the agent learns to compose visual operations over multiple steps and exhibits diverse problem-solving patterns unpromptedly. Code and data will be released at https://github.com/gengyuanmax/Mosaic. --- ### LampAttention: Look-Ahead Mixed-Precision FlashAttention for Dedicated Accelerators Authors: Stanislav Budzinskiy, Marian Gloser, Tolunay Yilmaz, Ying Hong Tham, Yuanyi Lin, Wenyi Fang, Fan Wu, Philipp Petersen Published: 2026-09-30 Categories: cs.LG, math.NA arXiv: https://arxiv.org/abs/2609.39361v1 While most attention logits can be computed in low precision without degrading numerical stability, current attention kernels fail to exploit this phenomenon. We introduce a novel hardware-algorithm co-design in the form of mixed-precision FlashAttention. Our method accumulates key-query products and evaluates their exponentials in 8-bit formats, then adaptively identifies sensitive sub-blocks and recomputes them in 16-bit formats. We propose the specifications for a dedicated accelerator capable of executing this pipeline efficiently. Simulated experiments with Qwen3 and Gemma 3 show that rerouting a selective minority of sub-blocks to high precision is sufficient to recover the baseline model performance. --- ### Autoresearch in Mixed-Integer Linear and Nonlinear Programming Authors: Yuwei Gu, Yaoxin Wu, Tong Guo, Wen Song, Zhiguang Cao Published: 2026-09-30 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.39360v1 Despite recent progress in autoresearch, applying it to practical operations research problems, typically formulated as NP-hard mixed-integer linear or nonlinear programs (MILPs or MINLPs), remains challenging because effective research requires systematically managing competing ideas and long-horizon experimental trajectories. We introduce AutoMIP, a reusable agent skill for organizing long-horizon autoresearch in mixed-integer programming through idea pooling and algorithm tree search. AutoMIP maintains a persistent pool of complementary candidate ideas while organizing executable experiments into an algorithm tree, enabling the agent to preserve unexplored hypotheses, refine promising algorithms, and switch to alternative methodological directions based on historical states. On MILP and MINLP benchmark cohorts, AutoMIP achieves the highest final success rates among the evaluated autoresearch frameworks. On MIPLib, AutoMIP discovers new best solutions for 31 of 60 instances, surpassing existing autoresearch frameworks. On MINLPLib, it achieves new best solutions for 52 of 60 instances. Ablation studies further demonstrate the complementary contributions of idea pooling and algorithm tree search, highlighting the importance of jointly maintaining diverse research ideas and structured experimental trajectories for long-horizon autoresearch. --- ### Working Around the Compute Ceiling: Byte-Exact Memory in Galahad Makes LLM Reading a One-Time Cost LLM Reading a One-Time Cost Authors: Sietse Schelpe Published: 2026-09-30 Categories: cs.CL, cs.AI, cs.IR, cs.LG, cs.PF arXiv: https://arxiv.org/abs/2609.39358v1 A transformer language model performs a bounded amount of computation per token, and recent work by Vishal Sikka, former CEO of Infosys, argues that this bound limits which tasks a model can carry out or verify (arXiv:2507.07505). We ask how much of the budget beneath that ceiling is spent on work the model has already done. Serving is stateless across requests: a model that answers a second question about a document recomputes the document's attention state from the first token. On seven real-world datasets, 98.7% of prompt tokens were text the model had already read. We present Galahad, a memory layer for vLLM, SGLang and llama.cpp that makes this reading a one-time cost. Taliesin saves the model's key-value (KV) state for a block of text and loads it on the next request that contains the same bytes, instead of recomputing it. Blaise keeps the documents themselves and passes the model only the section a question needs. On a recall test with 100 facts hidden in a 97,000-token corpus (Gemma 4 31B), Taliesin alone let the model attend to the whole corpus and answered 98 of 100 on llama.cpp at 3.0 s and 572 J per question, against 10 of 100, 9.3 s and 2,754 J for the same model without Galahad, which could hold only the last 12,000 tokens. With Blaise added, the model read about 668 tokens per question and answered 100 of 100 on all three runtimes at 0.59-0.64 s and 200-213 J; a tuned RAGFlow pipeline answered 77. Storing the corpus is a one-time cost of about 100 s and 28 kJ, whose energy is recovered after 13 questions. Restored state is bit-identical: all 262,144 output logits matched after restart, rehydration and hot-load. Galahad worked with all 30 models we tested under vLLM, and it fails closed: any load that does not pass its checks is recomputed. Together these results move LLM serving from stateless to stateful inference. --- ### Robustifying Asynchronous SGD via Soft Throttling Authors: Kaoru Otsuka, Maxime Meyer, Yuki Takezawa, Makoto Yamada, Anastasia Koloskova Published: 2026-09-30 Categories: cs.LG, cs.DC, math.OC arXiv: https://arxiv.org/abs/2609.39357v1 Asynchronous SGD is a popular algorithm for distributed learning where each client's gradient update is applied on arrival. This leads to a speed-up, but also an increased vulnerability to attacks, as fast clients can dominate the total update. We introduce Throttle, a Byzantine-robust generalization of asynchronous SGD where the key idea is to exponentially down-weight updates from faster clients by a factor $q$. Both asynchronous SGD ($q=1$) and synchronous Byzantine-robust SGD ($q\to\infty$) correspond to specific settings of Throttle. We provide a theoretical analysis of the convergence rate and validate the robustness to attacks both theoretically and empirically. Remarkably, our experiments show that this down-weighting mechanism can also improve performance over standard asynchronous SGD even in the non-Byzantine setting. --- ### Hiding in Plain Sight: Decoupling Pretext from Actuation for Skill Poisoning in LLM Agents Authors: Wenxin Wu, Lingyong Yan, Lei Sha, Shuaiqiang Wang, Jiashu Zhao Published: 2026-09-30 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.39352v1 LLM agents increasingly rely on reusable Skills for complex, multi-step tasks, creating a critical supply-chain attack surface where poisoned Skill content steers agent decision loops under benign requests. Existing skill poisoning attacks either colocate actuation with its contextual pretext or distribute actuation across multiple Skills, but do not explicitly separate the rationale for execution from the operation itself. In this work, we reveal that untrusted agent decisions fundamentally depend on two conceptually distinct Risk-Realization Factors (RRFs): an actuation factor (specifying what concrete operation is performed) and a pretext factor (providing the situational rationale for why the agent must perform it). Guided by this abstraction, we propose a coordination-based attack paradigm: decoupling pretext from actuation. Rather than fragmenting the malicious actuation, we preserve it as an intact operation within a downstream Steering Skill, while delegating the pretext factor to an upstream Grounding Skill that subtly alters persistent environment artifacts through routine utility operations. The intact actuation thus hides in plain sight, appearing completely legitimate and task-driven only when evaluated against the fabricated pretext. Building on this formulation, we develop an automated framework that discovers authentic execution dependencies, synthesizes coordinated pretext-actuation skill pairs, and iteratively refines poisoned skill instructions via runtime closed-loop feedback. Extensive evaluations across single-session and persistent cross-lifecycle scenarios demonstrate that decoupled skill poisoning achieves high attack success, exposing a critical blind spot in isolated Skill security audits. Our automated framework code is available at https://github.com/Wenxin-buaa/CoordPoison.git. --- ### On the Complexity of Preference-Based Bandits Authors: Ahmed Ben Yahmed, Marc Abeille, Clément Calauzènes Published: 2026-09-30 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.39351v1 We study preference-based bandits with general reward function classes, where a learner sequentially selects pairs of arms and observes binary preference feedback governed by the Bradley--Terry model. This setting naturally arises in applications such as recommender systems, tournament ranking, and learning from human feedback, where relative preferences are easier to elicit than absolute rewards. The observation model inherits the logistic bandit challenge of handling the problem-dependent constant $κ$, which accounts for the non-linearity of the link function and can grow arbitrarily large. Moreover, prior work has predominantly focused on linear or kernelized reward models, precluding the use of richer function classes. To address these limitations, we consider general reward function classes and introduce the \emph{locally sensitive eluder dimension}, a novel complexity measure tailored to the logistic structure of preference feedback that yields fine-grained regret guarantees without unfavorable dependence on $κ$. Building on this notion, we propose \textbf{GINOP} (Generic INformative OPtimism), an algorithm that constructs log-loss confidence sets and jointly selects arm pairs to balance optimism and informative exploration. We establish a first-order regret bound that, in contrast with what previous results suggest, demonstrates that learning with preference feedback is as statistically efficient as learning from direct reward observation. Finally, we corroborate our theoretical findings with empirical evaluations against competitive baselines. --- ### HAPMoE: Heterogeneity-Aware Automatic Parallelism Planning for Mixture-of-Experts Models Training Authors: Mengyuan Fan, Peizhuang Cong, Zixiao Huang, Si Xu, Tong Qiao, Yanghao Li, Jing Yang, Tong Yang, Quanlu Zhang, Yu Wang Published: 2026-09-30 Categories: cs.DC, cs.LG arXiv: https://arxiv.org/abs/2609.39350v1 As model sizes continue to scale, distributed training has become inevitable. Automatic parallelization techniques can derive efficient training parallelism strategies at low cost while achieving superior performance. The difficulty of this problem is jointly determined by the complexity of the model and the underlying compute cluster. Meanwhile, mixture-of-experts (MoE) models are increasingly emerging as the dominant architecture and the rapid evolution of accelerator hardware has made cluster heterogeneity commonplace, posing substantial challenges to automatic parallelization. However, existing approaches typically target either MoE architectures or heterogeneous clusters, failing to generalize to scenarios where both challenges coexist. To this end, we present HAPMoE, a heterogeneity-aware automatic parallelism planner for MoE training. HAPMoE builds a lightweight MoE-aware cost model and efficiently searches a six-dimensional parallel space, producing parallel plans directly deployable on Megatron-LM. Experiments show that HAPMoE improves end-to-end training throughput by up to 3.2$\times$ over baselines across heterogeneous clusters. Its non-uniform pipeline partitioning yields an additional up to 78% gains, and its pruning-enhanced dynamic programming algorithm completes the search within 1 minute, demonstrating high efficiency and practical value in complex hardware environments. --- ### Offline Guidance, Online Reasoning: Reusing LLM Feedback for Small Language Models Authors: Bohan Zhang, Linan Yue, Weibo Gao, Pengyu Chen, Hong Guo, Yanqi Hao Published: 2026-09-30 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.39346v1 Large language models (LLMs) offer strong reasoning capabilities but are often costly to access through commercial APIs, while small language models (SLMs) are easier to deploy locally yet remain weaker in reasoning. This capability-deployment gap has motivated LLM-SLM collaboration, which aims to improve SLM reasoning using LLM capabilities while preserving the deployment advantages of SLMs. Existing approaches mainly follow two paradigms. Knowledge distillation uses LLM-generated answers and reasoning trajectories to train SLMs offline, but requires parameter updates and additional training. Alternatively, online collaboration routes difficult problems to an LLM or leverages LLM-generated guidance and corrections when an SLM encounters difficulties. Although effective, online collaboration requires repeated LLM access. Moreover, the guidance produced for a particular problem is discarded after inference and cannot benefit subsequent problems involving similar reasoning states. In the paper, we focus on a more constrained setting in which the LLM is accessed only offline, the SLM parameters remain fixed, and online inference is performed solely by the SLM. To this end, we propose Reusable Latent Correction (RLC), which converts one-off natural-language guidance from a black-box LLM into persistent corrective experiences in the hidden space of an SLM. RLC stores these experiences in an external bank and retrieves them according to the SLM's current reasoning state, enabling the SLM to reuse LLM-derived corrections during inference without any online LLM calls. Experiments across multiple reasoning benchmarks and SLM scales show that RLC consistently improves SLM reasoning without parameter updates or online LLM calls. Code is available at https://github.com/ZBH031/reusable-latent-correction. --- ### Who Said What, and Will It Be Remembered? Evaluating Persistent Speaker Attribution Across Meetings Authors: Shantanu Vispute, Aditya Mishra, Siddhartha Saxena Published: 2026-09-30 Categories: cs.SD, cs.AI arXiv: https://arxiv.org/abs/2609.39344v1 Speech transcripts used as long-term memory must preserve both words and stable speaker identities. Existing meeting-transcription metrics either ignore speakers or remap anonymous speakers independently in each recording, so they cannot measure whether the same person retains one identity across meetings. We evaluate persistent speaker attribution with Speaker Identified cpWER (SI-cpWER), which scores a corpus under one global speaker-ID assignment. The benchmark covers five commercial diarize-then-identify cascades, two open academic baselines, and ThyVoice on the full 129-meeting CHiME-8 NOTSOFAR evaluation set in clean and noiseaugmented form, plus CHiME-6. ThyVoice is our end-to-end reference system; it repairs overlap and gates the evidence used to create and update voiceprints. Requiring persistent identity changes the commercial ranking: ThyVoice records lower SI-cpWER than every evaluated commercial cascade in all three conditions and the lowest mean in the full panel, 47.13 versus 54.75 for the next system. Complementary lexical, diarization, per-recording attribution, and speaker-clustering diagnostics characterize upstream error surfaces in the final attributed record. These results show why persistent attribution must be evaluated directly in systems that reuse conversations across time. --- ### The Golden Path Hypothesis: Reusable Schedules in Diffusion Caching Authors: Dong Wang, Wenwu Tang, Francesco Corti, Yun Cheng, Lothar Thiele, Olga Saukh Published: 2026-09-30 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.39343v1 Diffusion caching accelerates generation by replacing transformer computation with cached or predicted features at selected denoising steps. We introduce the Golden Path Hypothesis (GPH): under fixed inference conditions, prompt-independent cache schedules can achieve final-output quality comparable to the best prompt-specific schedules across prompts. We investigate the GPH across ten caching methods, four image and video models, and three cache ratios. Prompt-adaptive methods repeatedly select a small number of schedules, and reusing their most frequent schedules on new prompts closely matches the quality of prompt-specific choices. Exhaustive evaluation of 1.4 million schedules on four examples further identifies prompt-independent schedules that remain competitive on unseen prompts. To explain this transfer, we analyze denoising trajectories and the accumulation of caching errors. Latent-state trajectories exhibit similar structures across datasets and seeds, while an exact error decomposition shows that accumulated effects of earlier errors predict final latent-state error better than local approximation errors. This motivates searching for end-to-end schedules using final-output quality. With only a small set of examples, the resulting golden paths transfer across prompts and datasets, and can be tuned to the desired quality objective, including reconstruction fidelity or perceptual similarity. --- ### Belief-Based Maximum Occupancy Principle and Active Inference Authors: Manolis Mylonas, Rubén Moreno Bote Published: 2026-09-30 Categories: q-bio.NC, cs.AI arXiv: https://arxiv.org/abs/2609.39342v1 Intrinsic motivation plays a central role in adaptive and goal-directed behavior by conferring agents reward-independent objectives and biases useful to act in noisy and uncertain environments. Active Inference addresses the problem of acting in a partially observable environment through a principled framework for belief updating and action selection. A key component of Active Inference is the specification of prior preferences, which shapes behavior by encoding desirable future outcomes. An intrinsic motivation approach called the Maximum Occupancy Principle (MOP) proposes that agents act so as to maximize occupancy over future paths of states and actions, with no preferences or epistemic targets. Despite its simple formulation, MOP gives rise to rich and adaptive behaviors that combine exploratory variability with goal-directed dynamics. In this work, we extend MOP to partially observable environments and introduce a Bellman reformulation of the Expected Free Energy for Active Inference, both incorporating belief-based inference over hidden states as part of the agent state. The Bellman formulation enables tractable offline computation via value iteration over the full belief-state space. We compare the resulting behaviors in a set of minimal experimental settings with uncertain food sources. We find that MOP agents switch between goal-directed (food seeking) behavior and exploration between different food sources, depending on their energy available and their belief state. In contrast, Active Inference agents mostly inhabit regions around a single food source, a strategy having both high pragmatic and epistemic value. We finally compare with Empowerment, which is shown to be qualitatively similar to Active Inference. --- ### Understanding as No-Arbitrage: Bounded Dutch Books as a Definition and Training Objective for Language Models Authors: Daniel Dragonevskiy Published: 2026-09-30 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.39341v1 Does a language model merely predict tokens, or does it understand what it says? We make this question measurable by defining "understanding" through the lens of no-arbitrage. A model understands a vocabulary to a certain degree if a computationally bounded trader cannot extract guaranteed profit by betting against the model's probabilities on logically related claims (a "Dutch book"). We establish three theoretical results: first, because full logical coherence is computationally intractable, understanding is inherently graded, not absolute. Second, we prove that the exact optimum of standard next-token prediction is inherently incoherent across different question formats; the flaw lies in the training objective, not the architecture. Third, we show that uncertainty accumulates predictably along reasoning chains, making unjustified overconfidence an arbitrage opportunity in itself. To address this, we introduce Arbitr, a training framework where an adversarial trader penalizes the model for logical inconsistencies, paired with a calibration anchor to prevent uninformative collapse. Across five pre-registered experiments on Qwen2.5 and Phi-3.5 models, we demonstrate that standard models are highly exploitable across different phrasings. Arbitr reduces this exploitability by orders of magnitude without sacrificing task accuracy, and the effect successfully transfers to unseen logical patterns and new model families. Crucially, we uncover a scaling illusion: at 7B parameters, near-zero measured incoherence often coincides with extreme, unjustified confidence. We conclude that while Arbitr enforces rigorous logical consistency, coherence is a necessary condition for knowledge, but not a sufficient one --- ### ElectrolyteFM: Unifying Electrolyte Property Prediction through Cross-Property Knowledge Learning Authors: Jiaxin Yu, Shuo Wang, Peng Wang, Yongcai Wang, Deying Li Published: 2026-09-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.39340v1 Electrolyte formulation design requires balancing multiple physicochemical properties, yet existing models often focus on a limited subset. Learning each property in isolation can overlook transferable chemical information, whereas indiscriminate sharing can introduce cross-property interference. Our directed transfer analysis shows that jointly learning two property prediction tasks can improve or degrade prediction relative to separate training, with asymmetric transfer effects between the tasks. We propose ElectrolyteFM, a unified multi-property prediction model which can more accurately predict multiple properties of each electrolyte by effectively identifying and utilizing property-specific features and knowledge shared across properties. More specifically, ElectrolyteFM learns property-specific representations independently and captures cross-property knowledge through a separately trained expert pool. A router selects relevant shared information for each formulation and target property, and property-specific residual adapters convert this information into corrections to the corresponding representation for prediction. Experiments on Electrolyte12 show that ElectrolyteFM reduces normalized mean absolute error averaged across 12 electrolyte properties by 14.8% relative to the strongest electrolyte-specific baseline. On an independent sodium-electrolyte dataset unseen during training, it reduces conductivity mean absolute error by 6.7% relative to the best-performing baseline. --- ### A Unified Dual Method for Matching Problems Authors: Guillaume Houry, Ferdinand Genans, Jean Feydy, François-Xavier Vialard Published: 2026-09-30 Categories: cs.CG, stat.ML arXiv: https://arxiv.org/abs/2609.39339v1 Matching problems are ubiquitous in data science as they enable the alignment of structured objects and distributions. While existing solvers are often tailored to specific matching formulations, we unify a broad class of such problems within a common mathematical and optimization framework based on duality theory. Theoretically, we demonstrate that matching objectives decomposable as a difference of convex (DC) functions can be recast as implicit registration problems. This connection links matching to another well-studied class of objectives and yields a dual formulation amenable to natural optimization strategies. We then apply these findings to quadratic matching (QM) problems, which admit DC decompositions and for which we provide extensive convergence guarantees. Our framework applies to Gromov-Wasserstein (GW), as well as its unbalanced formulation and several variants, which are increasingly popular QM problems. Numerically, we implement our algorithms at scale for various data modalities such as graphs, point clouds, meshes, and word embeddings. Finally, our modular approach allows us to explore new formulations such as fracture matching, broadening the scope of problems that can be addressed within this framework.. --- ### Learning Beyond Full Imitation: Task-Preserving Knowledge Distillation Authors: Qianfeng Yuan, Wenbing Tao Published: 2026-09-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.39338v1 Knowledge distillation transfers knowledge by encouraging a student to match a teacher's predicted class probabilities. These probabilities express not only confidence in the correct class, but also relations among incorrect alternatives. Yet closer imitation does not necessarily yield a better student. A student may already distinguish the correct class more sharply than its teacher, so further imitation can require giving back discrimination it has acquired. Our main result is an exact separation between full imitation and conditional learning. When the correct class's score advantage over each alternative must be preserved, full teacher-to-student KL minimization is blocked exactly when the student assigns no more probability than the teacher to every incorrect class. Crucially, the teacher's relative probabilities among incorrect classes remain fully learnable. We characterize the exact price of this transfer: a minimum increase in correct-class log-odds that compensates for the largest conditional-probability mismatch. Label fitting and conditional matching can therefore be completed even as full teacher KL diverges. This separation motivates task-preserving knowledge distillation (TPKD), which keeps the label gradient intact and minimally corrects the conditional gradient so that its output update preserves the label step's gains against every incorrect alternative. The corrected conditional direction retains more than half of the original first-order conditional descent at the same step size, with a tight bound. For a fixed positive conditional target and sufficiently small constant output steps, label and conditional errors vanish together. Experiments trace this learning from exact head updates to ordinary network training. TPKD reaches 88.05% accuracy on CIFAR-100 and 93.81% on CLINC150, improving over standard distillation by 0.47 and 0.35 percentage points across three seeds. --- ### WinoTS: Wavelet-based Self-Distillation for Time Series Models Authors: Noam Major, Kathy Razmadze, Yoli Shavit Published: 2026-09-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.39337v1 Self-supervised pre-training of time series models is currently dominated by next-token prediction and reconstruction objectives. In continuous-valued domains, these paradigms often waste model capacity on high-frequency, point-wise noise at the expense of learning invariant structure. While invariance-based self-distillation has proven highly effective in computer vision, its application to temporal data remains largely underexplored. Effectively adapting such methods to time series requires carefully designed augmentations: spatial operations like cropping can shift the timing of repeating cycles or distort the signal, while basic jittering may provide limited variation. We introduce Wavelet-based self-distillation for time series (WinoTS), an invariance-based pre-training paradigm designed specifically for temporal signals. At its core, WinoTS leverages time-frequency augmentations to construct multi-scale structural views without distorting underlying signal dynamics. Across extensive evaluations, WinoTS outperforms state-of-the-art baselines in long-term forecasting, cross-domain zero-shot transfer, and unsupervised anomaly detection. Notably, linear probing on frozen WinoTS representations frequently surpasses fully supervised models trained from scratch. Systematic ablations demonstrate that WinoTS is a flexible, architecture-agnostic framework yielding gains across time series backbones, and establish that time-frequency transformations provide a principled alternative to vision-style spatial augmentations. --- ### How Many Samples Are Enough for Learning Across Domains? Authors: Hong Zheng Published: 2026-09-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.39336v1 Understanding the fundamental mechanisms of learning is essential for designing systems with strong generalization. Recent studies have shown that increasing the number of training domains, or enlarging the distribution shift among them, improves generalization when each domain contains sufficiently many data samples. However, the conditions under which the data samples can be considered sufficient remain unexplored. In this work, we fill this gap by establishing criteria for per-domain sample requirements based on the presented learning bounds. These criteria not only reveal an inverse linear scaling law between the number of training domains and the number of samples required per domain, but also explain the fundamental rationale behind the assumption of data sufficiency, thereby providing theoretical guidance for assessing the adequacy of existing datasets and constructing datasets. This differs from classical learning theory, as the number of samples required is highly dependent on the number of training domains. Additionally, we prove the close relationship between in-domain learning and out-of-domain generalization through the presented generalization bounds, and lastly discuss some key arguments. --- ### TexTailor: Texture-Preserving Video Virtual Try-On via Adaptive Garment Conditioning Authors: Zijing Qin, Jun Zhou, Ruicheng Zhang, Jiaqi Hou, Zunnan Xu, Ronghui Li, Zhenyu Xie, Xiu Li Published: 2026-09-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.39335v1 Video virtual try-on has attracted increasing attention due to its broad potential in digital fashion and intelligent e-commerce. However, existing methods primarily focus on low-resolution settings and still face substantial challenges when extended to high-resolution scenarios. These limitations can be attributed to two main factors: (1) the insufficient utilization of rich garment reference information, and (2) the lack of explicit positional modeling between garment and video representations during cross-modal interaction, which weakens fine-grained local correspondence. To address these issues, we propose TexTailor, a high-fidelity video virtual try-on framework built upon a pretrained video Diffusion Transformer. Specifically, we introduce a timestep-adaptive modulation mechanism to dynamically adjust garment visual representations throughout denoising. We further develop a frame-aligned positional encoding strategy to strengthen garment-to-video correspondence, together with a multi-source injection design that reduces interference among heterogeneous conditions. Extensive experiments on multiple video virtual try-on benchmarks, including the high-resolution Eevee dataset, demonstrate that TexTailor achieves competitive performance in garment detail preservation, temporal consistency, and overall video quality. --- ### Taming Speculative Search for Test-Time Scaling in LLM Serving Authors: Jinwoo Jeong, Woohyung Choi, Myeongjae Jeon, Jeongseob Ahn Published: 2026-09-30 Categories: cs.DC, cs.CL, cs.OS arXiv: https://arxiv.org/abs/2609.39334v1 Test-time scaling has recently emerged as a powerful approach for improving LLM reasoning by allocating additional computation during inference, substantially enhancing accuracy on challenging tasks such as mathematics and coding. To accelerate the exploration of reasoning paths, recent studies proposed speculative execution. However, we show that supporting speculative execution poses two unique challenges for LLM serving systems: (1) an explosion in the search space of candidate paths and (2) frequent, fine-grained verification tasks for candidates. To address these challenges, this paper proposes SpecScale, a serving system for efficient speculative execution. We introduce three techniques to reconcile the trade-off between latency and computational overhead: (1) early pruning of low-quality candidate paths, (2) deduplicating computation across redundant candidate paths, and (3) deferring fine-grained verification tasks. We evaluate SpecScale on challenging reasoning benchmarks, including MATH and Olympiad. Our results show that SpecScale significantly outperforms both non-speculative and recent speculative approaches, delivering substantial improvements in throughput and latency while preserving answer quality. --- ### NarrativeSteward: Coordinating Delegation, Guidance, and Verification in Agent-Assisted Interactive Narrative Authoring Authors: Wenjin Wang, Jiazhen Lei, Yuxin Sha, Nuwa Xi, Meng Zhao, Xingxi Yin, Qi Liu, Yuliang Shen, Zixun Sun Published: 2026-09-30 Categories: cs.HC, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.39333v1 Autonomous AI agents can turn authors' goals into interactive narratives by independently organizing and carrying out generation and revision. As agents generate and revise extensive content, authors struggle to grasp its overall structure, local details, and relationships, complicating continued guidance. We present NarrativeSteward, an authoring environment that organizes outlines, worldbuilding, and narrative graphs as linked artifacts for agent implementation and author guidance. Agent dialogue and project-wide structural review help authors understand the evolving work and guide local and cross-layer revisions, while change records and execution verification help authors assess the resulting work. Technical tests validated the system's change records, recovery mechanisms, and execution diagnostics. In a 12-participant within-subject study, NarrativeSteward supported easier formulation of revision requests and inspection of changes, and greater perceived understanding of changes and story structure, than general-purpose agents. Qualitative findings show how reviewing the work and feedback helps authors develop requirements and guide subsequent delegation. We open-source NarrativeSteward at https://github.com/Tencent/NarrativeSteward. --- ### PatchKV: Weight-Space Compensation of KV Cache Authors: Chanryeol Lee, Chanhyuk Lee, Yeonwoo Choi, Donggyun Kim, Seunghoon Hong Published: 2026-09-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.39329v1 Long-context inference with Large Language Models (LLMs) is bottlenecked by the linearly growing memory of the key-value (KV) cache. Existing compression methods reduce the cache through token eviction or approximation, but degrade sharply at aggressive compression budgets. We propose PatchKV, a training-free framework that compensates KV cache compression methods by carrying part of the context in the model's weights. PatchKV pairs an off-the-shelf compressed KV cache with a context-specific weight patch, which is computed once at context-loading time and served for downstream queries for the context. The weight patch is derived in closed form via ridge regression, by aligning the block-wise activations of context-derived reference query tokens under the full cache and the compressed cache. Once merged into the model, the patch leaves the forward graph and per-query inference cost unchanged in the single-context, multi-query setting. Across long-context QA (SCBench with up to 170K tokens, SQuAD, NIAH) and math (GSM8K) benchmarks on three model architectures, PatchKV consistently improves cache compression methods, suggesting an alternative direction to compensate them at aggressive budgets. --- ### Discrete Score Matching Enables Causal Discovery from Count Data Authors: Euijong Song, Hyewon Park, Gunwoong Park Published: 2026-09-30 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.39326v1 Count data pose a challenge for score-matching-based causal discovery: derivatives are unavailable, and simply replacing them with finite differences does not generally suffice for causal discovery. We generalize SCORE's constant-curvature criterion (Rolland et al., 2022) by conditioning on the node's value, yielding the conditional curvature score (CCS) for ordering. We also extend curvature-based parent recovery through the off-diagonal curvature score (OCS), enabling directed acyclic graph (DAG) recovery with both scores constructed from score functions for continuous data and concrete scores for counts. In the bivariate setting, zero CCS exactly characterizes a semiparametric generalized linear model (GLM) conditional form in which the conditional family need not be specified in advance, unlike in classical GLMs. For bivariate semiparametric GLM DAGs under our regularity condition, canonical-parameter nonlinearity is necessary and sufficient for identifiability. In multivariate DAGs, this nonlinearity enables DAG recovery through CCS and OCS. Our framework identifies a new class of semiparametric GLM DAGs that strictly contains the nonlinear Gaussian ANM class identified by SCORE. We introduce DISCO (DIscrete SCOre), a count-DAG recovery algorithm that estimates CCS and OCS using discrete diffusion. Experiments demonstrate accurate DAG recovery across Poisson, negative binomial, binomial, and mixed-family settings, as well as scalability to 1,000-node DAGs on a single GPU. --- ### WorkGenesis: Building the Worlds That Teach Agents to Work Authors: Xinyu Zhu, Fenyi Liu, Yuzhu Cai, Shuo Tang, Rui Ye, Linfeng Zhang, Siheng Chen Published: 2026-09-30 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.39325v1 The ability of Large Language Model (LLM) agents to complete daily and professional work is receiving increasing attention. Training such agents requires realistic work scenarios. Expert-authored occupational work is costly and slow to produce, while unconstrained synthesis often yields tasks with weak factual grounding or internally inconsistent requirements. To bridge this gap, we introduce WorkGenesis, a framework that constructs executable occupational work from real-world artifacts through two core technical innovations: (1) Evidence-Based Work Construction, which grounds each unit of work in real-world evidence by retrieving public files guided by O*NET occupational knowledge and synthesizing the surrounding context, companion materials, work request, and itemwise rubric around them; and (2) Execution-Guided Consistency Verification, which renders a reference deliverable inside the constructed work, attributes every unsatisfied rubric item to the agent, the task, or the rubric, and uses task and rubric defects as feedback to iteratively repair the work until it passes the audit. Experimental results demonstrate that Fx-Work-35B, trained with simple supervised fine-tuning (SFT) on only 20K units of work synthesized by WorkGenesis, achieves the highest scores among all comparable-scale baselines on the five reported metrics across GDPvalAA-v2, APEX-Agents-AA, and JobBench (31.00 versus 24.79 average score), and even surpasses frontier models such as the 1.6T DeepSeek-V4-Pro-Preview. These results show that WorkGenesis provides scalable training data for working agents. --- ### MotionWeave: Learning Motion-Centered Future Dynamics for Vision-Language-Action Policies Authors: Jingqiu Wang, Yan Wang Published: 2026-09-30 Categories: cs.RO, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.39324v1 Vision-Language-Action (VLA) models have recently incorporated world models to provide richer dynamic supervision beyond sparse action labels. However, explicitly predicting future images or videos may include control-irrelevant appearance, while guidance derived from holistic future visual representations and shared global action features may fail to establish timestep-specific correspondence between actions and local visual changes. To address this issue, we propose MotionWeave, a motion-centric future-dynamics framework for action-chunk prediction with two modules: the Action-Induced Motion Grounder (AIMG) and the Horizon Residual Composer (HRC). Specifically, AIMG conditions on action and proprioceptive representations to construct horizon-specific queries that localize interaction regions associated with each future action timestep from current visual tokens. HRC extracts differences between interaction representations at adjacent horizons, encodes them as temporal motion cues, and injects them into action tokens through a gated residual. During training, robot-arm masks rendered from future frames are used to construct KL-based motion-grounding supervision, while inference uses only the current observation. On six MetaWorld tasks, MotionWeave achieves a 75.3% average success rate, an absolute gain of 8.6% over π0 (66.7%), especially on sustained-interaction tasks. Our code is available at https://github.com/autu-mn/MotionWeave. --- ### HiWE: Hierarchical World Knowledge Model with Visual Keypoint Enhancement for Zero-Shot 3D Path Planning Authors: Guoqing Ma, Mingqi Yuan, Chen Gao, Jiayu Chen, Shan Yu Published: 2026-09-30 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.39323v1 Robot demonstration generation requires a system to identify where an interaction should occur, plan a feasible motion, and execute the required contact. HiWE connects these decisions through a point-based interface between visual grounding and language-based planning. PointVLM is instruction-tuned to associate task-relevant objects with image coordinates using a mixture of point annotations, segmentation-derived samples, robot observations, and visual question answering data. Depth measurements lift these predictions into a semantic 3D representation. A language planner, 3DLLM, uses this representation to specify end-effector waypoints and gripper commands, while a hybrid grasping module resolves local grasp poses. The evaluation covers 14 simulated manipulation tasks and four physical-robot tasks, together with ablations of the visual training data, spatial inputs, and grasp selection. Here, zero-shot execution refers to deployment without task-specific demonstration training; the visual model uses existing robot data during fine-tuning. This paper describes the original point-based formulation of the framework; its relationship to the subsequent GeneralVLA extension is detailed in the introduction. --- ### GRPO Training Dynamics for Small Language Models Authors: Rajat Ghosh, Vaishnavi Bhargava, Henry Wong, Aryan Singhal, Debojyoti Dutta Published: 2026-09-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.39321v1 Group Relative Policy Optimization (GRPO) has emerged as a memory-efficient reinforcement fine-tuning (RFT) technique for reasoning-intensive tasks. How- ever, GRPO training dynamics on small language models (SLMs) remain poorly understood, limiting its reliable adoption and reproducibility in open and resource- constrained environments. In this work, we present a systematic study of GRPO fine-tuning for SLMs ranging from 1.5B to 7B parameters under a practical single- node 8xA100 compute budget. Our study spans multiple model families and reasoning domains, including mathematics, coding, and multiple-choice question answering (MCQ) in science. Across these settings, we analyze how group size affects policy convergence, training stability, and downstream benchmark per- formance. We further characterize tensor-level update dynamics during GRPO training and investigate whether the choice of LoRA target modules and layers can improve the performance of GRPO-tuned models. While our initial GRPO-tuned models outperform their base counterparts on approximately 80% of mathematical benchmark evaluations, they demonstrate limited capability on MCQ and code reasoning tasks. Guided by our mechanistic evaluations, we refined our LoRA and reward-shaping configurations to improve performance in latter domains. These findings provide practical guidance for GRPO training for SLMs. --- ### Rethinking Generative Image Compression at Extremely Low Bitrates Authors: Tianyu Zhang, Zhaoyang Jia, Houqiang Li, Dong Liu Published: 2026-09-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.39315v1 Generative image compression produces visually plausible reconstructions at low bitrates, yet their behavior as the rate approaches zero remains largely unexplored. When pushed below normal operating rates, representative codecs undergo semantic collapse: rather than gracefully losing source-specific detail, they produce malformed or unrecognizable content. Our analysis identifies two factors. As the bitrate decreases, reconstruction losses increasingly conflict with semantic objectives on gradients and visual results, while pixel-space and reconstruction-oriented VAE diffusion models become less efficient on semantic preservation. Guided by these findings, we introduce RAE-CoD, a compression-oriented diffusion (CoD) built in a representation autoencoder (RAE) space with direct alignment between compressed and source representations, preserving recognizable, naturally structured content for a $256\times256$ image with as few as 16 bits. We evaluate this framework using five vision foundation models (VFM) and a blinded vision-language model protocol. On MSCOCO-30K, RAE-CoD stands out from all evaluation. At 0.001-0.008 bpp, it reduces relative VFM feature MSE and Fréchet Distance ratio by at least 25.7% and 69.1% over the best competitors. Meanwhile, semantic recognizability and quality of the reconstructions remain nearly constant while source consistency falls smoothly, replacing abrupt semantic collapse with a graceful transition toward unconditional generation. Code will be released at https://github.com/LuizScarlet/RAE-CoD. --- ### ReSAIL: Mitigating Collapse in Iterative Agent Self-Distillation Authors: Shengjie Jin, Hengbo Xu, Zelong Sun, YuJie Guo, Zhiwu Lu Published: 2026-09-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.39306v1 Iterative self-distillation enables LLM agents to learn from successive deployments, offering a path toward recursive self-improvement (RSI). Yet our experiments with existing methods reveal a collapse in deployment performance across cycles, while task performance with privileged information (PI) also declines. We address this collapse by prioritizing informative interaction steps for distillation and preserving PI-conditioned behavior as the student becomes the next teacher. We introduce Retentive and Selective Augmentation for Iterative Self-Distillation (ReSAIL), a plug-in augmentation for iterative PI-based self-distillation. ReSAIL selects interaction steps where PI most strongly changes the teacher's predictions and balances the resulting distillation losses across trajectories. It also regularizes the student's PI-conditioned output distributions toward those of the frozen teacher at selected and unselected steps to preserve PI-conditioned behavior for supervision in the next cycle. On ALFWorld and TextCraft, ReSAIL sustains substantial gains across model scales over three cycles, with an average absolute gain of 22.5% in final-cycle success rates when added to self-distillation baselines. Sensitivity-guided selection of offline data also improves action prediction accuracy for multimodal GUI agents on AITZ. These findings provide the first evidence that a more robust learning mechanism can effectively mitigate performance collapse in iterative agent self-distillation over deployment trajectories. --- ### Scale and Selection: What Makes Automatic Harness Evolution Work for Visual-Interface Robot Agents Authors: Zhijie Wei, Ferris Tan, Jinghui Wang Published: 2026-09-30 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.39304v1 When an off-the-shelf coding agent is used directly as a robot policy, observing a browser-based 3D interface through screenshots and acting by posing a virtual target gripper through a few tools, the agent's harness, its prompts, tools, and control rules, largely determines success, and until now it has been written by hand. We show that this harness can be improved automatically by another coding agent, the optimizer agent, and report two findings about what makes it work. First, the number of rollouts the optimizer agent sees per round governs whether the evolved harness is trustworthy, generalizes, and improves steadily. A single rollout is a noisy binary outcome, so with few rollouts per round a revision can be promoted on luck; enlarging the batch raises the signal-to-noise ratio of every promotion decision. Holding rounds fixed and growing the training set from 5 to 100 rollouts, held-out success rises from 47% to 67%, while small training sets overfit, reaching 70% on training tasks but only 54% held-out. Second, the optimizer agent must not be given free rein. With every revision it proposes accepted unconditionally, performance drifts downward within ten rounds as ill-judged edits accumulate; adding the most basic safeguard, Champion-Challenger selection that promotes a revision only if it strictly beats the incumbent on the same fixed evaluation set, turns the same loop into one that raises held-out success from 51% to 67% over 30 rounds. Automatic harness evolution for visual-interface robot agents is thus feasible, but its gains hinge on the rollout scale behind each decision and on how the optimizer agent's revisions are selected. --- ### Generalized Geometry Block Proximal Linearized Method for Multiblock Nonconvex and Nonsmooth Optimization Authors: Weifeng Yang Published: 2026-09-30 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2609.39301v1 This paper considers a class of multiblock nonconvex and nonsmooth optimization problems arising in many applications. Existing methods construct proximal linearized operators or their variants within standard Euclidean geometry to solve this class of problems, forcing their block variable updates to rely on the standard inner product and its induced norm. Nevertheless, this construction fails to capture the geometric structure of the target problem, leading to low numerical efficiency. To overcome these drawbacks, we propose a generalized geometry proximal linearized operator for updating block variables, and develop the Generalized Geometry Block Proximal Linearized (GGBPL) method based on this operator. Compared with existing proximal linearized operators, the proposed operator allows the block surrogate functions to be constructed using arbitrary inner products and general admissible metrics, thereby enabling the GGBPL method to adapt its updates to the geometric structure of various problems. We also introduce the inertial version of GGBPL, named the inertial GGBPL (iGGBPL) method. We further establish a new unified convergence framework under this generalized geometry, within which we prove that our methods guarantee convergence of the objective function values, establish global convergence of the generated sequence to a critical point, and derive the convergence rate of our methods. We also establish an $\mathcal{O}(\varepsilon^{-2})$ iteration complexity bound for obtaining an $\varepsilon$-stationary point. We apply our methods to two nonconvex and nonsmooth problems: sparse nonnegative matrix factorization with $\ell_0$-constraints and sparse nonnegative CP decomposition with $\ell_0$-constraints. Numerical results demonstrate the superior numerical performance of our proposed methods over several state-of-the-art methods. --- ### BMASH: Ball-Motion-Aware Soccer Header Spotting Authors: Ahmed Endris Hasen, Muhammad Shahzad Khan, Nikolaos Passalis, Jenni Raitoharju Published: 2026-09-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.39300v1 Recent advances in computer vision have made broadcast sports videos increasingly useful for event analysis, performance assessment, and player-safety applications. In soccer, however, header spotting remains a challenging problem due to the subtle and short-lived nature of header events. This paper focuses on soccer header spotting: identifying moments in broadcast videos where the ball contacts a player's head. We first adapt and evaluate Video Swin as a strong action-recognition baseline for this task, and then introduce BMASH, a ball-motion-aware fusion framework that integrates detector-derived ball features. BMASH combines Video Swin action representations with ball-presence and motion features from frame-level soccer-ball detection, integrating player-action context with ball dynamics to distinguish headers from visually similar events. We evaluate BMASH using game-level splits with separate test matches and rotating validation folds, considering both centered-window classification and continuous full-video spotting. Results show that Video Swin provides a strong baseline for header spotting, while BMASH improves clip-level AP and ROC-AUC over the corresponding Video Swin baseline. In continuous full-video spotting, BMASH achieves a comparable event-level F1-performance with a different precision--recall trade-off. --- ### MiniRep: Robust Reputation-Based Aggregation for Multi-Agent Debate Authors: Jiaming Zhang, Yuwan Liu, Yue Huang, Sisi Duan Published: 2026-09-30 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.39297v1 Autonomous agents powered by large language models (LLMs) are rapidly evolving into an open agentic ecosystem. To support trustworthy collaboration, industry initiatives increasingly assess agent reputation from past behavior and provide performance leaderboards. However, reputation derived from past performance may not reliably predict an agent's behavior on new tasks, particularly when malicious agents can adapt their behavior and influence other agents during collaboration. We study reputation in multi-agent debate (MAD), where multiple agents answer the same query, debate to improve their answers, and aggregate them into a final output. We present MiniRep, a reputation-based aggregation system for MAD under malicious agents. To ground our threat model in established research, we construct an attack taxonomy drawing on reputation-system attacks and software-testing mutation operators, covering strategic exploitation of reputation and subtle corruption of agent proposals. Guided by this taxonomy, MiniRep evaluates agents based on both their behavior on the current task and their reputation over time, while preventing groups of agents with highly similar responses from dominating the final decision. We assess MiniRep across diverse tasks, LLM-agent compositions, corruption placements, and attack types drawn from our taxonomy. Our experimental results show that, MiniRep outperforms both conventional MAD aggregation and conventional reputation-based approaches on MATH no matter being attacked or not. Also, under a heterogeneous 10-agent setting on MATH, MiniRep outperforms all baselines in all 28 attack conditions. --- ### Semantic-Aware Joint Source-Channel Optimization for Encoder-Agnostic Digital Video Communication Authors: Xiangben Zhu, Caili Guo, Yang Yang, Chuanhong Liu, Meiyi Zhu Published: 2026-09-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.39296v1 Video semantic communication has attracted increasing attention as a promising approach to improving video transmission efficiency. However, most existing approaches rely on computationally intensive deep learning-based video encoders and decoders, which hinders their deployment in resource-constrained scenarios. To address this issue, we propose a lightweight semantic-aware joint source-channel optimization (SAJSCO) scheme that can be integrated into existing digital video communication systems as a plug-in module. Specifically, we develop a video communication system model in which the transmitter jointly optimizes source and channel coding parameters based on the inter-frame semantic importance of the input video and estimated channel state information. On this basis, we formulate an optimization problem that maximizes semantic importance weighted video reconstruction quality under a maximum bitrate constraint. To solve it, we first quantify inter-frame semantic importance using a cosine similarity-based metric with a shifted window mechanism. We then develop a multi-actor proximal policy optimization (MPPO) algorithm to solve the formulated problem by jointly adapting the source compression rate and channel coding rate. The learned policy can be directly applied to different video encoders without encoder-specific retraining or fine-tuning. SAJSCO achieves Bjøntegaard Delta rate reductions of 34.86\% and 18.01\% when integrated with H.265, a conventional video encoder, and DCVC-RT, a deep learning-based video encoder. Over-the-air experiments on a hardware testbed further demonstrate a PSNR gain of up to 1.448 dB with H.265 and an LPIPS reduction of up to 0.033 with DCVC-RT compared with the respective best-performing fixed-parameter baselines. --- ### ANI: Adaptive Numerical Injection for Unifying Semantic and Arithmetic Representations in Numerical Reasoning Authors: Jinsung Jeon, Seung-won Hwang Published: 2026-09-30 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.39294v1 Precise numerical reasoning with Large Language Models (LLMs) is essential for expanding their applicability to complex real-world tasks. However, text-based tokenization often fragments numbers, significantly hindering precise arithmetic reasoning. Meanwhile, numerical embeddings, despite arithmetic precision, rely on context-agnostic substitution that disregards the semantic role of numbers as identifiers. To combine the complementary strengths, we propose \textbf{ANI (Adaptive Numerical Injection)}, a hybrid framework that governs the selective injection of numerical features based on the semantic context. By employing a context-aware gating mechanism, we selectively inject numerical embeddings (specifically FoNE) into the latent space, explicitly preserving nominal identifiers while enhancing quantitative operands. Through extensive evaluations across various LLMs, we demonstrate that ANI enhances MATH performance by 9.5 points over the official reference model, while maintaining robust performance on general linguistic benchmarks. --- ### Physics-Informed Method of Group Data Handling: Adaptive Construction of Functional Representations with an Application to the Navier-Stokes Equations Authors: Mykhailo Minin Published: 2026-09-30 Categories: cs.LG, physics.flu-dyn arXiv: https://arxiv.org/abs/2609.39291v1 Physics-informed computational methods usually optimize parameters within a functional representation whose structure is fixed in advance. This work proposes a Physics-Informed Method of Group Data Handling (PI-GMDH), in which representations of coupled physical fields are progressively constructed during solution. Candidate functional directions are evaluated through the first variation of the complete physical and observational objective, introduced in packages, and followed by block-coordinate damped Gauss-Newton coefficient optimization. The framework is demonstrated with tensor-product Chebyshev functions on the incompressible Navier-Stokes equations using a two-dimensional time-dependent Taylor-Green benchmark. Under the tested configuration, adaptive PI-GMDH reached validation and held-out test losses of 5.299e-19 and 5.296e-19 with 204, 201, and 175 active functions for u, v, and p. Complete degree-by-degree and all-terms PI-GMDH variants, together with selected PINN and KAN reference configurations, are used to examine the effect of structural construction policy. The results show that, for this controlled synthetic benchmark, selective progressive construction can provide a favorable combination of accuracy, representation size, and wall-clock time. The comparison is illustrative rather than a claim of universal superiority over alternative physics-informed approaches. --- ### EngramBench: A Capability-Grounded Benchmark for Skill-Evolution Harnesses Authors: Zhixuan Tan, Pengjie Gu, Zhao Li, Yihan Hu, Xu He, Dong Li, Jianye Hao Published: 2026-09-30 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2609.39284v1 While large language models have achieved remarkable success in isolated code generation, authentic software engineering requires sustained reasoning, complex state management, and continuous cross-domain abstraction. However, current evaluations of skill evolution in autonomous agents suffer from a critical identifiability problem: they structurally confound genuine capability abstraction with rote solution leakage (i.e., copying highly similar code from historical training data). To resolve this, we introduce EngramBench, a rigorous, capability-grounded benchmark governed by the strict axiom of capability overlap without solution overlap. Comprising 30 diverse learning tasks and 13 unseen transfer tasks, EngramBench challenges agents to navigate interactive, multi-hour development cycles driven by LLM-simulated users. Our extensive evaluation across 48 multi-hour execution trajectories -- corroborated by human-expert validation -- reveals a profound insight into procedural memory. We demonstrate that static skill banks do not magically bypass the "last mile" of exact code implementation, which remains bottlenecked by the base model's inherent reasoning limits. However, they serve as an indispensable execution compass. By navigating agents away from catastrophic, token-heavy trial-and-error, genuine capability abstraction slashes redundant context bloat and reduces overall coding time by over 55%. Ultimately, EngramBench shifts the evaluation paradigm from trivial pattern matching to the verifiable measurement of deep, cross-domain capability transfer. --- ### Faithful Dual-constrained Erasure for Robust LLM Safety Alignment Authors: Jiaqing Li, Shide Zhou, Zhibo Zhang, Yuxi Li, Tianlong Yu, Kailong Wang Published: 2026-09-30 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.39279v1 Machine unlearning has emerged as a crucial mechanism for removing hazardous knowledge and enforcing safety alignment in Large Language Models (LLMs). However, recent studies reveal a persistent security risk: unlearned models remain highly vulnerable to retraining attacks, where suppressed malicious behaviors rapidly resurface after benign fine-tuning. In this work, we investigate the optimization dynamics of unlearning and identify that this vulnerability stems from shallow alignment. Rather than effectively erasing target knowledge, models often exploit a shortcut by activating previously dormant parameters to act as spurious suppressors, forming a fragile inhibitory shell over intact malicious representations. To address this issue and enforce authentic memory deletion, we propose FDCU, a novel dual-constrained subspace projection framework. FDCU restricts parameter updates through a highly scalable, element-wise dual-masking rule: it preserves general knowledge manifolds via Fisher Information and strictly prohibits the abnormal activation of spurious suppressors via the Principle of Minimal Functional Intervention (PMFI). By reliably blocking the model's ability to superficially hide knowledge, FDCU promotes the authentic dismantling of target representations. Extensive experiments across specific knowledge erasure and safe output control tasks demonstrate that FDCU achieves state-of-the-art robustness against retraining attacks while maintaining near-lossless general utility, ensuring durable safety for LLMs. --- ### A Tilted Bowl Is Not a Slippery Slope: Compressing Looped Models Authors: Steven Kolawole, Pearse Jim, Opegbemi M. Busoye, Glory Bagai, Virginia Smith Published: 2026-09-30 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.39277v1 Looped models reason by applying the same block of weights many times, so compressing that block saves memory traffic on every loop. Compressed looped models, however, often collapse, and the collapse is usually blamed on rounding error that accumulates from loop to loop. In this work we test that account on more than 30 models from five families and find, to our surprise, that it holds only for loops that never settle. When a loop settles, a fixed rounding error does not accumulate. It moves the point where the loop settles, much as tilting a bowl moves where a ball comes to rest, and the answer is lost only when the shift is larger than the readout tolerates. This picture lets us predict which models fail from a single label-free measurement, and it tells us why failed models recover: their loops still settle, so a few final loops with 8-bit weights bring the answer back. Motivated by these findings, we build a controller that stops when the model's halting head fires and then finishes with 8-bit loops. On Sudoku-Extreme and Maze-Hard it beats fixed-depth inference by up to 15 points under a third of the weight traffic. --- ### ReTaCo: Residual-Target Control for On-Policy Distillation Authors: Zixiang Ni, Zhuo Hu, Renjie Cao, Weijie Ren, Binqin Shi, Weijia Zhang, Shuheng Cao, Zhicheng Shi, Zhenhao Zhang, Haomin Wen, Zhiyuan Hu Published: 2026-09-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.39275v1 On-policy distillation (OPD) trains a student on its own generated prefixes with token-level teacher feedback, but transmitting or storing the teacher's full-vocabulary distribution at every token is costly. Entropy-aware OPD (EOPD) adds forward supervision to reverse KL to help the student recover plausible tokens it underestimates, using only the teacher's top-$k$ probabilities to limit cost. Because EOPD renormalizes these probabilities, its target assigns no mass to the omitted vocabulary. We prove that the resulting loss keeps pushing the student's top-$k$ mass toward one even after the student matches the teacher's relative probabilities within the top-$k$ set, so the teacher itself is not a stationary point whenever the omitted tokens have positive teacher probability. We propose ReTaCo (Residual-Target Control), which keeps the top-$k$ tokens individually and groups the remaining tokens into one residual symbol, and pairs this forward target with a single-sample estimator whose expectation equals the full-vocabulary reverse KL. With teacher top-$k$ mass $m$, the residual target is $(1-β)(1-m)$ for $β\in(0,1)$: $β=0$ preserves the teacher's mass, and larger $β$ moves more mass onto the top-$k$ tokens without changing their relative probabilities. At a fixed prefix, we prove that the population objective has a unique optimum whose top-$k$ mass lies between $m$ and $m+β(1-m)$ and increases monotonically with $β$; at $β=0$, underestimated top-$k$ tokens still receive non-vanishing recovery gradients. Numerical optimization confirms these predictions, and across three teacher-student pairs, ReTaCo outperforms EOPD on most mathematics and code benchmarks. --- ### MegaAvatar: Controllable Talking Avatar Generation Authors: Junyao Gao, Sibo Liu, Weidong Zhang, Cairong Zhao, Jun Zhang Published: 2026-09-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.39273v1 This report presents \textbf{MegaAvatar}, a controllable talking avatar generation framework built on top of the Wan2.2-TI2V-5B model. Compared with previous talking-avatar methods that mainly rely on audio or reference-image conditioning, we introduce additional SMPL-X-derived 3D guidance, enabling global control over body pose and head motion. Specifically, we render the driving SMPL-X sequence into dense mesh frames and encode them with a lightweight 3D convolutional encoder, whose outputs are injected into the latent tokens to provide overall motion control. Furthermore, we extend Wan2.2-TI2V-5B with additional audio and face cross-attention modules to enable fine-grained expression control and preserve the input identity, respectively. In addition, we implement an audio-to-SMPL-X model to predict an SMPL-X sequence conditioned on the reference image and input audio, allowing MegaAvatar to support audio-driven inference without user-provided SMPL-X frames. Experiments show that MegaAvatar achieves high-quality talking avatar generation with controllable body and head motion, speech-synchronized facial expressions, and consistent identity preservation. MegaAvatar also supports inference with flexible resolutions and video lengths. Codes, dataset, models will be avaliable in https://github.com/Jeoyal/MegaAvatar --- ### An Island-Based Parallel Biased Random-Key Genetic Algorithm for the Three-Dimensional Trailer Loading Problem Authors: A. del Río, L. Díaz, L. C. de Vicente, J. Cameselle, B. Fernández Published: 2026-09-30 Categories: cs.NE, cs.DC, math.OC arXiv: https://arxiv.org/abs/2609.39272v1 The Three-Dimensional Trailer Loading Problem (3D-TLP) involves determining the optimal placement and orientation of heterogeneous items within the confined space of a trailer while maximizing volume utilization and satisfying a wide range of complex logistical and safety constraints. The 3D-TLP is NP-hard, rendering exact optimization approaches computationally impractical for large-scale industrial applications. To address this challenge, we propose an enhanced Biased Random-Key Genetic Algorithm (BRKGA) accelerated through a novel island-based parallelization framework, PANGEA. The proposed method combines the search efficiency and robustness of BRKGA with a multi-population evolutionary scheme for genetic algorithms. This island-model strategy promotes population diversity, mitigates premature convergence, and significantly reduces computational times. The proposed solution was validated in a real trailer loading process, providing an effective solution approach for real-world large-scale logistics. --- ### RW-Flow: One-Step Generation on Compact Manifolds via Wasserstein Gradient Flows Authors: Ualibyek Nurgulan, Seungwoo Yoo, Prin Phunyaphibarn, Minhyuk Sung Published: 2026-09-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.39271v1 Manifold-valued data, and consequently the distributions they induce, are prevalent across many domains, ranging from the locations of geospatial events, such as earthquakes, to biomolecular torsion angles that encode information about three-dimensional structure. While diffusion and flow-based generative models have been successfully extended to compact manifolds, sampling typically requires tens or hundreds of sequential network evaluations. We introduce RW-Flow, a theoretically grounded framework for learning one-step generative models on compact manifolds via Wasserstein gradient flows. The main challenge is identifiability: driving the velocity field to zero should guarantee that the model distribution matches the target distribution. We establish a necessary and sufficient condition for identifiability on compact, connected Riemannian manifolds. We specifically show that, for a symmetric, Lipschitz-continuous cost function, the velocity field induced by the Sinkhorn divergence is identifiable if and only if the associated Gibbs kernel is nondegenerate. This characterization provides a general principle for designing identifiable costs on compact manifolds. It also reveals that the squared geodesic distance, the natural manifold analogue of the squared Euclidean distance, does not always guarantee identifiability. Across benchmarks involving geospatial events, protein side chain torsion angles, RNA backbone torsion angles, and general manifolds discretized as triangular meshes, RW-Flow outperforms existing one-step methods in nearly all settings under fair comparison conditions. --- ### A Time-Aware Bag-of-Receptive-Fields for Interpretable Irregular Time Series Classification Authors: Francesco Spinnato Published: 2026-09-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.39268v1 Irregular time series, characterized by non-uniform sampling intervals, missing observations, and variable lengths, are ubiquitous in healthcare, mobility, and environmental monitoring, yet effective and interpretable classifiers for this setting are limited. Existing approaches often rely on imputation, which can obscure the temporal structure of the data, or require complex neural architectures that are opaque and difficult to explain. In this work, we extend the Bag-Of-Receptive-Fields (BORF), a fast, deterministic, and interpretable transform for time series, to the irregular setting. Our key contribution is a time-weighted normalization scheme in which each observation is weighted proportionally to its associated time delta, making pattern extraction sensitive to the actual temporal distribution of samples rather than only their index position. This requires deriving an efficient sliding-window recurrence for the time-weighted standard deviation, preserving the linear time complexity of BORF. We benchmark the resulting method against state-of-the-art irregular time series classifiers on datasets from the PYRREGULAR repository, demonstrating competitive classification performance with the added benefit of human-interpretable explanations. --- ### PLRS-IC: A Dual-Calibration Framework for Chest X-Ray Vision-Language Alignment Authors: Qixing Zhao, Jinpeng Li Published: 2026-09-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.39266v1 Fine-grained vision-language alignment in chest radiography enables zero-shot classification, grounding, and segmentation without task-specific annotations. However, this alignment is fundamentally hindered by two intertwined sources of ambiguity: projection-induced visual mismatch and patient-agnostic semantic overlap. First, at the local feature level, frontal and lateral radiographs exhibit distinct appearances for the same clinical finding, rendering a shared patch-text similarity geometry inherently suboptimal. Compounding this visual ambiguity is a semantic mismatch during global contrastive optimization, where instance-level objectives penalize cross-patient pairs as strict negatives even when they share identical positive clinical concepts. To address this dual ambiguity, we propose PLRS-IC, a unified dual-calibration framework for chest X-ray representation learning. At the local alignment stage, Projection-Conditioned Low-Rank Residual Similarity (PLRS) dynamically adapts patch-text matching to projection-specific manifolds using a bounded, parameter-efficient low-rank residual. At the global optimization stage, Information-Content-Calibrated Soft False-Negative Suppression (IC-SFNS) leverages a corpus-derived information-theoretic prior to soften the penalty of semantically overlapping negatives without altering original contrastive assignments. Extensive experiments across nine public zero-shot benchmark settings demonstrate that our framework yields consistent improvements in classification, grounding, and segmentation, validating the necessity of dual-calibration in medical vision-language pre-training. --- ### Universal Cross-Prompt Adversarial Attacks on Promptable Concept Segmentation Authors: Ziqi Zhou, Yifan Hu, Yufei Song, Haowen Jiang, Xianlong Wang, Shengshan Hu, Dezhong Yao, Leo Yu Zhang Published: 2026-09-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.39265v1 The Segment Anything Model (SAM) achieves remarkable performance in visual segmentation. The latest SAM3 extends promptable segmentation to concept-level prediction, broadening the scope of segmentation foundation models. While recent works reveal that SAM and SAM2 are vulnerable to adversarial examples, the robustness of SAM3 under the concept segmentation paradigm remains unexplored. In addition, existing adversarial attacks on SAM-series models exhibit limited cross-prompt transferability. To this end, we propose AdvPCS, a universal cross-prompt adversarial attack for Promptable Concept Segmentation (PCS), including a min-max prompt optimization strategy, a global-local perception deception attack, and a temporal transition deviation attack. Specifically, we first identify the hardest-to-attack prompts via min-max bilevel optimization. In the inner maximization, we enhance diversity over candidate point, box, and text prompts. In the outer minimization, we select prompts with the highest responses based on the confidence scores output by the detector. Given the selected prompts, we apply the perception deception attack to minimize both global and local existence probabilities under joint prompting and employ the temporal memory misalignment attack to maximize inter-frame semantic inconsistency and corrupt memory pointers. Extensive experiments on four benchmark datasets show that a single universal adversarial perturbation (UAP) generated by our method generalizes across frames from different videos and achieves strong attack performance under point, box, and text prompts. In particular, it reduces the average mIoU of various PCS models on the SA-CO dataset to below 5% under text prompts, demonstrating strong attack ability. --- ### Concept Subspaces Compute Beyond the Logit Lens: A Weights-Only Test for Locating Representations Upstream of Readout Authors: Aojie Yuan, Zhiyuan Julian Su, Haiyue Zhang, Zijian Su Published: 2026-09-30 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.39263v1 A concept subspace's effect on model behavior does not establish how it relates to the output readout. We introduce a two-sided geometric diagnostic that measures an extracted subspace's overlap with the dominant right-singular directions of the unembedding matrix, evaluated against output-oriented positive controls. Given an extracted basis, the raw diagnostic requires only model weights. Our testbed is the Format-Agnostic Reasoning Subspace (FARS), a ten-dimensional basis extracted from eighteen reasoning concepts expressed in six surface forms. Across nine rank-matched estimators and twenty-six models, four activation-derived concept estimators carry only 0.38--0.80% mean energy in the top-ten readout span. Final-layer PCA carries 3.56%, exceeding FARS in 25 of 26 models. A same-layer next-token control, evaluated using a fitted linear translator for depth matching, carries approximately thirteen times more energy than FARS, with separation in all 25 tested models. Re-extracting FARS on ten disjoint concepts yields 62--100% cross-format retrieval across twenty-four generative models, demonstrating transfer of the extraction procedure rather than a fixed basis. A complementary four-model, three-seed intervention study finds model-dependent source-directed effects that remain well below full-vector replacement. Together, the geometry and intervention controls distinguish concept structure from dominant readout directions while limiting claims of causal sufficiency. --- ### Jacobian Rank Collapse in Decision-Focused Learning Authors: Aojie Yuan, Haiyue Zhang, Zijian Su Published: 2026-09-30 Categories: cs.LG, q-fin.PM arXiv: https://arxiv.org/abs/2609.39261v1 Decision-focused learning (DFL) trains predictors through downstream objectives, but a different loss need not provide an independent parameter-update direction. We characterize this restriction through the predictor Jacobian, using sparse index tracking to distinguish the covariance entries read by the optimizer from the parameter directions available to learning. Rank-one Jacobians make nonzero per-example gradients collinear; a conditional spectral bound describes near-collinearity. A batch-subspace characterization and counterexamples show why these local statements imply neither common minimizers nor collinear batch updates. Experiments examine when geometry translates into decision quality. Across 38 one-parameter equity configurations, DFL gains over MSE remain below 1.8%; a 385-parameter conditional predictor also has pointwise rank one. In validation-tuned shortest-path and knapsack experiments, full-capacity SPO+ reduces mean regret by 11.6% and 10.6%, respectively; only knapsack survives correction across eight comparisons. The capacity contrast persists on fresh datasets across batch orders and training budgets. Holding expressivity fixed, invertible coordinate scaling lowers spectral effective rank and ordinary SGD gains; compensating for the scaling restores the original trajectories. Financial forward-target controls separate forecast accuracy from decision quality; a matched neural comparison finds no aggregate DFL advantage in the tested architecture. These findings distinguish local rank restrictions, coordinate-dependent optimization and predictive accuracy. Predictor geometry helps explain available learning directions, while held-out decision quality remains the test of practical benefit. --- ### Effective Does Not Mean Useful: Conditional Functional Substitutability for Redundancy and Scaling in Transformers Authors: Jiaheng Chen, Jiaxing Li, Yucheng Xiao, Xinyong Cai, Juncheng Bu, Lan Yu, Tinghe Zhang Published: 2026-09-30 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.39259v1 Modern neural networks scale predictably, yet the mechanisms behind these regularities remain unclear. Neural redundancy is typically characterized by component importance or representational similarity, both indirect proxies. We view redundancy as an input-conditioned, dynamic relation: intermediate computational states are functionally redundant when they induce similar downstream responses. We introduce Conditional Functional Substitutability (CFS) to directly characterize such functional substitution. CFS exposes functional relations and reduction potential missed by conventional importance- and similarity-based measures. Across modalities and Transformer families, CFS reveals systematic functional reorganization with scale. Controlled scaling further shows that performance gains need not track growth in substitutability, while fixed-capacity models with more independent functional structure perform better, providing a functional account of diminishing returns. Predicted CFS further enables dynamic computation with a better performance--computation trade-off than importance-based component selection, suggesting new directions for redundancy-aware computation and more efficient model scaling. ---