Papers

No specification says how a governed autotelic AI agent organization, where agents pursue self-generated goals inside guardrails, should be designed and evaluated. We answer in two parts. First, we synthesize the Governed Autotelic Multi-Agent Product Organization (GAMPO) framework from a document-based qualitative evidence synthesis of 321 sources, integrating agency, agile, platform, and governance theory into a runnable specification. Second, we probe a prompt-layer instantiation of GAMPO on CHI-Bench, a long-horizon healthcare benchmark, across open and frontier models. The result is a boundary condition: governance benefit is gated by a model's spare capacity and is domain- and model-specific. On capacity-constrained open models the full procedure yields no reliable benefit, whereas a single "verify your writes" sentence doubles task success (pass@1 2/20 to 4/20). At the frontier the same scaffold lifts prior-authorization 24% to 40% but nets zero on another model, a gap traced to a stable recommendation-override disposition. A second result refines the first: replacing the generic procedure with an answer-blind, per-task definition-of-done, keyed only to the case's own policy and published standards, never the hidden key, raises prior-authorization to 84% under best-of-five self-consistency (68% single-attempt, confirmed by a held-out board) and utilization-management to 44%, while care-management meets a subjective content-quality wall. The contribution is a named, auditable framework and capability-gated evidence that governance should be sized to spare capacity, and that at the frontier a case-grounded specification beats a uniform procedure. Findings are exploratory: partial instantiation, small per-cell samples (n = 5-25), and single trials.

Vision-language models (VLMs), such as CLIP, have achieved strong performance across multimodal tasks by aligning visual and textual representations in a shared embedding space. As VLMs are increasingly used for high-stakes domains, failure prediction becomes critical for risk-aware deployment and human intervention. Existing failure prediction methods typically rely on confidence scores or auxiliary classifiers. Although these methods are effective on predicting VLM failures, they provide limited interpretability. In this work, we investigate the use of Sparse Autoencoders (SAEs) for interpretable failure prediction in VLMs. We formulate failure prediction as a classification task over sparse SAE latent activations and introduce a three-stage failure-aware training pipeline that encourages the learned latent directions to remain interpretable while becoming more informative for failure prediction. Our experiments show that the resulting framework outperforms the evaluated baselines in failure prediction. Further analysis suggests that failure-aware training encourages SAE latent directions to capture more class-specific concepts. We also use the SAE to provide a concept-level analysis of how model representations change during failures, revealing a shift from class-specific concepts toward more ambiguous or style-related concepts. Finally, we explore how the learned SAE latent directions can support runtime failure recovery.

Overfitted codecs have demonstrated promising performance for image and video compression. In particular, for image compression, the Cool-chic family of models has shown competitive performance against scene-agnostic models, with orders of magnitude lower decoding complexity at the cost of a longer overfitting process. However, these overfitted image codecs are not fully optimized toward the rate-distortion objective: their network weights remain in full precision during training, and the associated quantization parameters are selected in a separate post-training stage. Furthermore, their synthesis operates at a single scale, which overlooks cross-scale redundancy. In this paper, we propose MIRC, an overfitted image codec in which every coded component, including the latents, the synthesis network, and the entropy models, is quantized and entropy coded under a single rate-distortion objective, adopting the end-to-end compression pipeline of the neural video representation codec NVRC. We further introduce a multi-scale representation with cross-stage parameter sharing, which improves coding efficiency at a small transmitted overhead. On the CLIC2020 professional validation set, MIRC achieves a 10.5% BD-rate saving against VVC (VTM 22.0). Moreover, MIRC offers a family of configurations spanning 1.2 to 2.9 kMAC per pixel, so the decoding budget can be selected to match the deployment target.

Scalable video coding (SVC) encodes a video into a layered bitstream of a base layer and one or more enhancement layers, enabling decoding at different bitrate/quality/resolution operating points for diverse devices and network conditions. SVC has been incorporated into major video coding standards and has recently attracted growing interest for both scene-agnostic and scene-adaptive neural video codecs. Among the latter, implicit neural representation (INR) based codecs overfit a compact neural network to an individual video, offering fast decoding and competitive coding efficiency. However, research on scalable INR-based compression remains in its infancy: existing methods introduce additional network layers, which couple the bitrate with the decoding complexity and cannot match strong scalable/non-scalable codecs. This paper proposes S-NVRC, a scalable INR-based video codec that jointly supports fine-grained bitrate and decoding complexity scalability from a single embedded bitstream. It adopts a coarse-to-fine prefix for feature grids and a nested prefix for network layers, which scale bitrate and decoding complexity, respectively. With a single encoding, S-NVRC spans a wide range of bitrate and decoding complexity and outperforms SHM-12.4 and the multi-layer VTM-20.0 by 43.7% and 5.6% in BD-rate on the UVG dataset.

This study examines the ability of large language models (LLMs) to predict the risk of weather-related forced outages in the distribution grid in a zero-shot framework, without labeled training data. The problem is formulated as a binary severity classification task across three forecast horizons (3h, 6h, 12h), using six years of outage records and high-resolution weather data for a utility service area in central Texas. Four zero-shot LLMs are benchmarked against two supervised classifiers across two input configurations: one using current weather observations and the other using weather forecast data. Results show that supervised models outperform LLMs on macro-F1 and precision, while newer LLM generations achieve competitive scores. Beyond accuracy, LLMs offer complementary strengths in actionable reasoning and geographic scalability, suggesting that combining them with supervised models may be the best practice.

Wi-Fi-based human activity recognition (HAR) has become an important part of integrated sensing and communications, paving the way for a range of context-aware services. However, most existing Wi-Fi-based HAR systems rely on deep learning (DL) models that are computationally and memory intensive in both training and inference, which poses significant challenges for real-world deployment. Conventional training requires simultaneous updates of millions of parameters, leading to prohibitive memory consumption. In this paper, we propose a novel quantum-assisted memory-efficient training framework (Q-MET) designed to improve efficiency in both training and inference. Q-MET utilizes a hybrid quantum classical neural network to indirectly generate parameters for HAR models, significantly reducing the trainable parameter count compared to direct optimization. To further support the deployment on resource-constrained devices, we integrate structured pruning during the training phase. Experimental results demonstrate that Q-MET achieves a 90% to 95% reduction in trainable parameters compared with conventional backpropagation-based DL training while maintaining or even exceeding classical classification accuracy. Additionally, Q-MET supports lightweight inference through structured pruning, achieving 75% to 85% model sparsity with less than 2% loss in classification accuracy. To the best of our knowledge, this work represents the first quantum-assisted approach to simultaneously tackle memory inefficiencies in both the training and inference stages of HAR systems.

Multi-agent LLM pipelines increasingly assign roles, including execution and verification, to models of different capability tiers. This is done because running a flagship model at every stage is expensive. Previous literature has established that verification stages are not always beneficial, but holds reviewer capability roughly fixed relative to the executor. We vary it. We replace the reviewer with models spanning a capability range down to one that cannot solve the problems at all, and measure the outcome of every individual rejection. This is done across a constant set of 100 olympiad mathematics problems. A cross-family mid-tier reviewer improves final accuracy by 12 percentage points, from 52 to 64 percent (p = 0.0005), with zero damaged answers. Same-model self-review attains the highest error-detection rate of any condition (0.85 recall) yet yields no significant gain: it rejects 2.1 times as often for a third the repair rate (15 against 43 percent, p = 0.0074) and falsely rejects 35 percent of its own correct answers against 2 percent for the cross-family reviewer (paired p = 0.000015). The low damage rate of self-review proves to be an artifact of revision inertia rather than reviewer quality: of 18 falsely rejected correct answers, the three where the executor complied all became wrong, while the fifteen it ignored survived unchanged. Below a capability floor the role becomes inert: our weakest reviewer changed zero of 100 final answers while doubling token cost. These findings describe a single executor-reviewer configuration on 100 problems and should be read as a controlled pilot rather than a general claim about verification stages.

Deciding whether two supplier records belong to the same corporate family is a prerequisite for spend consolidation, credit exposure aggregation and sanctions screening. It is usually treated as entity matching, but the tasks differ: a family link connects records that are deliberately different entities, and the evidence often appears in neither record. We introduce CorpFam, a public benchmark of 54,864 candidate pairs over 10,307 corporate families, derived from 6,638,350 US federal award records in which every supplier self-reports its ultimate parent to a government registry. Pairs are stratified by name visibility: whether the names are identical after normalisation, share a distinctive token, or share none. Because strata have positive rates from 10.2% to 97.3%, we report per-stratum recall, base-rate invariant, rather than F1, which is not. The strongest of 5 matchers recovers 100.0% of identical pairs and 4.2% of invisible ones; no method exceeds 4.7% on the latter. The failure begins before matching. Blocking decides which pairs a matcher sees, and we evaluate 7 schemes spanning phonetic keys, attribute keys that ignore the name, and semantic nearest neighbours. None reaches three percent on invisible pairs, and their union recovers 6.8%. 93.2% of these links never enter the candidate set, so no matching-stage improvement can reach them. The links are real: against SEC Exhibit 21 subsidiary schedules, which share no provenance with procurement registration, 64.2% of invisible links are corroborated, against 0.16% under permuted parents and 0.41% against the same parent's wrong exhibit: two unrelated nulls agreeing to within 0.25 points. Corporate-family resolution is a retrieval problem misfiled as a matching problem; the intervention point is candidate generation, not ranking. The benchmark, adjudication log, and code reproducing every number are released.

Aerodynamic surrogate models trained on high-fidelity CFD data reproduce numerical predictions of both scalar outputs and entire fields accurately, yet their predictive fidelity is limited by systematic discrepancies between CFD and experimental observations. We present an experimentally grounded correction framework that adapts a CFD-trained deep learning surrogate using wind-tunnel PSP measurements. A Geotransolver surrogate trained on 2,300 high-fidelity CFD simulations of the NASA CRM wing-body configuration, spanning geometric variation, Mach 0.70-0.85, and angles of attack 0 to 4 degrees, reproduces the CFD integrated aerodynamic forces and pitching moment to R2 > 0.99 but does not match the experimental data. To incorporate experimental information without retraining the surrogate, a correction network is trained on spatially registered PSP measurements at two freestream Mach numbers (0.70 and 0.85) across the same angle-of-attack range, learning the discrepancy between the surrogate-predicted and experimentally measured surface-pressure distributions. At Mach 0.85 the correction substantially improves agreement with PSP, particularly at the wing suction peak, shock location, and subsequent pressure recovery, reducing both the magnitude of the prediction error and the fraction of wetted surface on which it exceeds 0.05 in Cp, and it does so from a limited experimental dataset without modifying the pretrained surrogate parameters. On held-out angles of attack the grounded surrogate agrees with measurement to within 2.3-2.7% of the measured Cp range, and outperforms direct interpolation between the measured conditions at every state tested. Experimental measurements can therefore ground a large-scale simulation-trained surrogate by learning systematic CFD-to-experiment discrepancies while preserving its generalization capability and computational efficiency.

Softmax is a key operation in Transformer attention, but its exponentiation and normalization add significant overhead in compute-in-memory (CIM) accelerators, especially when analog attention scores must first be converted to the digital domain. This work presents a tunable-temperature analog softmax circuit in GlobalFoundries 22-nm fully depleted silicon-on-insulator (FDSOI) technology that operates directly on CIM-generated score voltages without intermediate analog-to-digital conversion. Each input score is converted into a time-domain event using a shared falling ramp. The corresponding comparator transition samples an RC-decaying reference to generate an exponential weight, which is then processed by an in-circuit normalization stage. In contrast to analog softmax circuits that rely on transistor weak-inversion behavior for exponentiation, the proposed architecture controls the softmax response through the ramp slope and RC time constant, enabling programmable effective temperature. The 128-element architecture is evaluated using transistor-level and post-layout extracted simulations, including multi-level input vectors, capacitance variation and mismatch, process and temperature variation, monte carlo analysis, and shared-interconnect parasitics. The complete 128-element implementation occupies 9453.42~$\mu\mathrm{m}^{2}$ including the shared global ramp circuitry, while each replicated softmax element occupies 70.2~$\mu\mathrm{m}^{2}$. The circuit achieves a 242.97-ns evaluation latency at 13.44~mW total power, corresponding to 25.5~pJ per output element. The simultaneous 128-element evaluation achieves an RMSE of 24.46~mV relative to the ideal softmax response. The extracted circuit characteristics are further incorporated into a MemTorch-based hardware-aware Transformer model, where the proposed softmax achieves a validation loss within 2.5% of the ideal-softmax baseline.

Existing post-hoc explainers for machine learning classifiers primarily focus on feature attribution, assigning importance scores to individual features. While valuable, this approach struggles to articulate the complex, combinatorial patterns that often drive a model's decision-making process. To overcome this limitation, we introduce ProToMEx, a new paradigm for explainability that leverages Probabilistic Topic Models (PTMs). Our model-agnostic framework learns latent ”topics” that represent distinct, high-level reasons for a classification, moving beyond simple feature importance to reveal underlying semantic structures. ProToMEx naturally provides both global explanations of a model's overall behaviour and local explanations that can disentangle multiple co-existing reasons for a specific prediction. We demonstrate empirically that ProToMEx not only produces explanations of comparable fidelity to popular methods like SHAP and LIME but also drastically reduces the amortised computational cost of generating local explanations, making it highly suitable for real-time applications. Specifically, we show that ProToMEx is ~30-40x faster than SHAP and LIME over standardised tabular datasets and synthetic datasets.

Latent world models have become increasingly popular as a method to predict and plan in latent space rather than pixel space. Recent architectures, such as LeWorldModel (LeWM), jointly train the encoder and predictor using regularization techniques like SIGReg to prevent representation collapse. Even with such regularization preventing representation collapse, we identify a new world model failure mode of physical representation laziness, particularly noted in highly dynamic environments. For these lazy cases, the learned latent states do not collapse but nonetheless fail to represent key physical properties, causing ubiquitous downstream planning failure. To resolve this issue, we propose training-time auxiliary supervision with a lightweight "Fourier auxiliary head", which enforces physically-informed structuring of the latent space with no additional inference-time cost and can be generalized to any environment. Experimentally, we show that the auxiliary head substantially improves planning success rates in dynamic environments where the baseline LeWM exhibits physical representation laziness. It also leads to modest improvements in other environments, even when the baseline does not exhibit physical representation laziness. We further observe superior planning performance being accompanied by higher latent space correlations with key physical properties, indicating both the ability of our method to physically structure latent states and the potential planning-side benefit to the learned representation being physically structured. We also see in low-data regimes, auxiliary supervision is particularly impactful in increasing success rate. These findings support the use of our Fourier auxiliary head method to improve both overall success rate and data efficiency, while avoiding representation laziness in latent world models.

Low-bit key–value (KV) caches reduce the memory required for autoregressive decoding, but the resulting quality loss depends on the model and quantizer. We keep the quantizer fixed and distill the floating-cache model's behavior into low-rank Q/K/V projection updates while the student executes a physically packed incremental cache. Across three seeds, 4-bit affine-cache adapters recover $54.24\%\pm2.47\%$ of the held-out perplexity gap on TinyLlama-1.1B and $75.96\%\pm4.04\%$ on Gemma-4-12B. On the same frozen NF4 Llama-3.1-8B base, one validation-selected run per quantizer recovers $60.42\%$ under KIVI K2V2 and $37.61\%$ under KVarN K4V2, while preserving 180-case associative retrieval. Gemma's score on an official 4K/8K RULER subset rises from 42.80 with the unadapted 4-bit cache to 48.33 after adaptation (46.15 floating), with substantial task heterogeneity. Finally, a 2-bit rank–token sweep reduces TinyLlama's 2-bit PPL from 576.10 to $11.4000\pm0.0059$ across three seeds, versus 10.3988 floating, but restores only 11–12 of 180 retrieval cases. These results show that low-rank projection adaptation can recover held-out quality across fixed cache formats, while perplexity recovery need not restore long-context retrieval.

Spell correction for Japanese search queries presents unique challenges due to the co-existence of four writing scripts (Latin/romaji, hiragana, katakana, and kanji) and the distinct error patterns each script induces. We present a compact BART-based sequence-to-sequence model (3 encoder + 3 decoder layers) designed for low-latency spell correction of Japanese music search queries. The core contribution lies in a script-aware synthetic misspelling generation pipeline that produces realistic training data by combining keyboard-layout models (QWERTY and flick input), phonetic confusion priors mined from real query logs, voiced/unvoiced consonant alternations, and kana case errors. A key design decision is normalizing mixed-script catalog titles to a single canonical script before misspelling synthesis, which we show is critical for reducing model hallucinations. We train a custom byte-level BPE tokenizer on the target music catalog to handle all four scripts in a unified vocabulary. Experiments on a curated evaluation set show that our model achieves an exact-match accuracy of 41.09% and a character error rate (CER) of 11.62%, outperforming edit-distance baselines and achieving the lowest character error rate among all evaluated systems while maintaining sub-4ms inference latency on a single GPU. We further analyze performance across individual scripts and mixed-script queries, demonstrating the effectiveness of script-aware data augmentation through systematic ablation studies.

Self-supervised pretraining has transformed language and vision, but its value for molecular graph neural networks remains contested. We ask whether pretraining on a large unlabelled corpus improves molecular property prediction. We adapt LeJEPA, a predictor-free joint-embedding predictive architecture regularised by Sketched Isotropic Gaussian Regularisation (SIGReg), to molecular graphs, evaluating GPS and Chemprop-style D-MPNN encoders on the Wong et al. [1] antibiotic-activity dataset and ogbg-molhiv using a multi-seed, bootstrap-based protocol. Pretraining improves learned representations but does not robustly improve finetuning. A frozen probe on pretrained embeddings exceeds random initialisation on both tasks (ogbg-molhiv ROC-AUC 0.788 vs 0.665; +0.123), reaching the published self-supervised band, but this does not translate into finetuning gains. On the antibiotic scaffold split, a canonical partition is significant (delta AUPRC +0.041, p = 0.010), but the effect vanishes across five partitions (pooled +0.013, p = 0.095). Finetuning is null on the random split, ogbg-molhiv, and D-MPNN. The representational edge is nevertheless recoverable. Embeddings saturate at ~16-32 effective dimensions, whereas Morgan fingerprints improve to 1024 bits. At matched dimensionality, fingerprints lead validation (0.799 vs 0.782 at 128 dimensions) but trail shifted test scaffolds (0.759 vs 0.788). Truncating embeddings and combining them with a 1024-bit Morgan fingerprint raises ogbg-molhiv ROC-AUC from 0.805 to 0.832 (delta +0.027; 95% CI [+0.003, +0.054]; p = 0.014); an untrained encoder gains nothing (delta -0.003). Thus, pretraining supplies complementary information best realised through feature-level combination, while finetuning gains are weak and partition-dependent.

Automatic Speech Recognition (ASR) systems are widely deployed in safety-critical settings but remain vulnerable to data-poisoning backdoor attacks. Existing ASR backdoors typically use phrase-level triggers paired with a fixed target sentence, creating strong artifacts (e.g., repeated transcripts or triggers placed in non-speech regions) that simple preprocessing can mitigate. We propose GhostWord, a word-level, time-localized ASR backdoor that uses codebooks mapping short ($\approx$400\,ms) acoustic triggers to target words. During poisoning, we inject a trigger into the forced-aligned time span of a chosen source word in the audio and replace only that word in the transcript, enabling precise semantic flips and composable sentence manipulation while avoiding many-to-one label artifacts. Across Common Voice (v23 English, v24 Lithuanian) and multiple backbones (Whisper-Small/Medium, MMS, SpeechT5), GhostWord achieves an average attack success rate of 89.3% and transfers across languages and models. Adapting optimization-based defenses (ABL, ANP, SAU, I-BAU) reveals a sharp robustness–accuracy trade-off: attack success drops from 89.3% to 29.1% while clean WER rises from 21.5% to 45.0%, consistent with our theoretical analysis showing that, in high-vocabulary models, backdoor suppression structurally tends to degrade clean performance. The source code is publicly available at https://github.com/rohban-lab/GhostWord

Deep neural network (DNN) has been widely applied in various industries. Specialized chips are being discussed for the purpose of achieving lower power consumption with higher throughput. Hardware variations introduced during the process of chip manufacturing are the main reason for affecting the inference accuracies. In this paper, we propose hardware-conscious software training (HCST) method which enables high inference accuracies even under the influence of hardware variations.

Certifiable methods have arisen as a means to guarantee global optimality of solutions to non-convex problems using convex semidefinite programming (SDP) relaxations. The most performant of these methods use a local solver to obtain the candidate solution, and then certify its optimality using efficient linear algebra techniques. However, for many problems of interest in robotics, this local-solve-then-certify approach is impeded by a form of degeneracy in the relaxation, leaving a costly optimization of the relaxation as the only recourse. In this paper, we introduce our Central-Path Certifier (CP-Cert), a certifiable method explicitly tailored to certify candidate optima to problems that exhibit this form of degeneracy. Using a candidate as a starting point, our approach seeks a nearby region of the feasible space – known as the central path – where a valid certificate can be readily obtained. The approach is kept efficient by exploiting indirect linear algebra techniques, problem sparsity, and parallelism. We apply CP-Cert to both matrix-weighted pose registration and pointcloud data association, whose novel SDP relaxation is of independent interest. On simulated examples, we explore the properties of this novel relaxation and show that CP-Cert is fast and scalable, achieving runtimes that are up to three orders of magnitude faster than state-of-the-art direct solvers. Finally, we combine these contributions into a certifiable, outlier-robust pose-estimation pipeline, which we apply to real-world data.

Counterfactual fairness audits of clinical language-model agents report a flip rate: how often an action changes when only the patient's demographic descriptor changes. Part of that rate is not demographic. A stochastic agent also changes its own action when nothing changes, and a flip rate cannot be interpreted without knowing how often. We measured it. Re-running one condition ten times over sixteen synthetic vignettes at default sampling changed a clinical agent's action in 8.7 percent of replicate pairs, from 2.2 percent for intensive-care escalation to 17.9 percent for controlled-substance caution, an output given no operational criteria. Across six models from five vendors, pooled floors ranged from 2.5 to 23.7 percent; in this panel neither disclosed size, vendor, nor hosting ordered them. The floor depends on the decoding configuration: majority voting over five draws removed 39 percent of it (95 percent confidence interval (CI) 18 to 64); at temperature 0 three of four locally served models showed no disagreement, but a hosted model still did. We also show that, for a binary action, the flip rate expected under no demographic effect equals the floor and a real effect adds only its square, so a flip rate inside the floor is not evidence of fairness, and direction must be tested with a signed paired test. We give a four-step reporting procedure and release FairMedAgent, the harness, with its protocol, vignettes and analysis scripts, so any team can measure the floor for its own agent.

Large Language Models (LLMs) increasingly use user context such as memory, profiles, and role prompts to personalize their responses. This personalization can affect evidence-based judgment: the same evidence may lead to different conclusions under different user contexts. Finance provides a high-stakes setting to study this problem because decisions often depend on interpreting long and complex documents. We test this using 3,575 SEC filings across twelve LLMs. We compare persona-conditioned retrieval, neutral retrieval, and memory-framed context to separate the effect of evidence selection from the effect of interpretation. We find that most user-context spillover comes from how models interpret the same evidence under different roles, rather than from retrieving different evidence. We then test two simple mitigation strategies: expressing the same investor mindset as a user profile instead of an assistant role, and separating evidence-based and personalized outputs. Both reduce spillover, but neither removes it completely, and their effectiveness varies substantially across models.

Vision Transformers (ViTs) typically process every image using a fixed input resolution and model width, even though many images can be classified with substantially less computation. We introduce ProgResViT, an input-adaptive ViT that performs inference progressively across multiple rounds. The first round processes a low-resolution image with a narrow subnetwork. Inference terminates when the prediction is sufficiently confident; otherwise, the model reuses the representations produced in the current round and proceeds with a higher-resolution input and a wider subnetwork to refine its prediction. As all rounds share a single backbone, we propose Progress-Conditioned Soft Gating (PSG), which conditions token fusion and layer outputs on the current round, block, and input resolution. On image classification, applying ProgResViT to DeiT yields better accuracy-compute trade-offs than adaptive-width, adaptive-depth, and dynamic-token baselines. With knowledge distillation, a DeiT-based ProgResViT achieves 84.9% top-1 accuracy, slightly exceeding the reported DeiT-III-S accuracy under a comparable evaluation setting. We show that the same design also provides favorable accuracy-compute trade-offs for self-supervised DINO representations and downstream semantic segmentation. Code is available at https://github.com/ds-kiel/ProgResViT.

Writing proficiency manifests in how students develop content, organize ideas, choose words, and use language. Despite growing interest in LLM-based student simulation, whether LLMs can reproduce such multidimensional variation in extended writing remains largely unexplored. In this work, we explore if language models can realistically simulate student writing, and introduce SWIM, a task that formulates Student Writing sIMulation as proficiency-conditioned essay generation. We evaluate prompting, supervised fine-tuning (SFT), and reinforcement learning (RL) methods for writing simulation using automated essay scoring as a measure of profile alignment. Extensive experiments reveal that prompting provides limited proficiency control, even for strong proprietary LLMs with rubric-grounded strategies. In particular, while models can adjust content-oriented traits, they struggle to reproduce the lexical, grammatical, and organizational variation in different proficiency levels. SFT substantially improves alignment, while RL with the proposed proficiency-alignment reward yields further gains across all writing traits and essay prompts. Our findings suggest that explicit supervision enables substantially stronger profile alignment than prompting alone, while authentic low-proficiency writing remains challenging to reproduce.

This paper introduces a novel, efficient class of parallel direct and indirect iterative schemes to solve general obstacle and free boundary value problems. The uniqueness of the solution for the direct parallel method is established under the assumption that the model problem yields an $M$-matrix. The convergence analysis of the indirect approach is predicated on minimizing the discrete energy functional at each iterative stage of the linear approximation process. Under the aforementioned setting, we establish theoretical convergence results for both smooth and nonsmooth energy functionals defined over a convex set, in the sense of Ferris and Mangasarian \cite{ferris1994parallel}. For the numerical computation of the one-dimensional obstacle problem, we adopt a direct generalized parallel approach based on the SPIKE algorithm \cite{polizzi2006parallel}. To accelerate convergence and reduce computational complexity, a fast recursive version of the scheme is generalized to the nonsmooth case. For two- and three-dimensional obstacle problems, we employ a directional splitting method that treats each directional subproblem as a one-dimensional obstacle minimization problem. By framing the energy functional minimization as a parabolic time-dependent problem and utilizing an Armijo time-stepping rule during the solution update process, we successfully obtain results for higher-dimensional obstacles. Additionally, this study explores the possibility of extending the algorithm into a constrained quadratic programming optimization solver. We also evaluate the framework on image deblurring phenomena under the aforementioned setting, successfully recovering the original images. Finally, numerical illustrations are provided to validate the theoretical results.

Large language models (LLMs) exhibit in-context learning capabilities, where they can learn new tasks from prompt contexts without weight updates. We compare the learning efficacies of two prominent modes of in-context learning: (1) learning from descriptions of rules (instruction following); and (2) learning from examples of input-output demonstrations (few-shot prompting). Through five learning tasks that cover diverse domains (games, arithmetic, linguistic inferences), we compare two modes of learning (rules vs. examples) specifying the same underlying task. We furthermore explore model and task properties that modulate the learning efficacies. We find that models generally learn more reliably from rules than from examples alone, and additional examples on top of rules or simply scaling up the number of examples do not lead to consistent and significant gains. Instruction tuning amplifies the benefit of rule-based learning while keeping example-based learning capacities intact. Surprisingly, we find no privileged effect of example-based learning in base models, and rules still lead to gains in algebraic task domains. Overall, the comparative efficacy of rules over examples is larger when the task recruits algebraic abstractions and computations, and smaller when the task requires distributional sensitivity and/or recruits parametric knowledge.

Precipitation forecasts shape decision-making across the global economy, particularly in sectors such as agriculture. However, unlike variables such as temperature, precipitation is highly intermittent and localized, making it difficult to forecast. While recent advances in AI weather prediction systems have enabled them to surpass physical models on globally averaged metrics, improvements in mean error rarely translate to actionable forecasts of severe flooding or dry crop fields. Furthermore, most of these models are trained and evaluated against a reanalysis data product, ERA5, which has well-known biases. Here we retrain AIFS, ECMWF's widely-used, open-source operational 0.25{\deg} probabilistic graph-transformer weather model, on satellite-based precipitation observations. Our model, Laxmi, improves global probabilistic accuracy by 19% and resolves systematic distributional biases in ERA5. Specifically, Laxmi reduces drizzle overprediction by 33% for amounts less than 3 mm per day. It also mitigates extreme rainfall underprediction, improving the global 95th percentile Brier skill score by 57%. Across a case study of 10 Indian tropical storms, Laxmi delivered the most accurate forecast of 150 mm event-total precipitation in 7 events, compared to 1 for AIFS and 2 for the leading physical model, IFS. Our results demonstrate that incorporating observation-based precipitation data directly into training can substantially improve forecasts.

We study a governed approach to enterprise analytics: a language model interprets the question, while deterministic policy selects and runs a pre-approved analytical program that returns both results and evidence. We show that this restriction can remain expressive within a defined analytical class, using relational operations plus aggregation, comparison, windows, ranking, and similarity. Fixed meaning, policy, data, and execution rules also make results replayable. Across 440 runs, three 8B models generated SQL and selected tools at runtime, while Qwen3-8B interpreted intent only and policy executed the approved program. None of 330 runtime-planning episodes matched the full answer-and-evidence contract across all test datasets; the policy-executed analyzer matched 110 of 110. This is a configuration-specific result, not evidence that runtime agents cannot succeed under other designs.

Using a zoom-in tool is an important foundational part of modern visual agents, because it allows to efficiently handle tasks involving high-resolution images. Most previous methods need an extensive warm-start supervised fine-tuning phase for teaching models zoom-in. We show that this is not necessary by proposing a new intrinsic reward for learning tool use in MLLMs without the need for additional labels or warm-start SFT. Our InfoNCE-style reward uses a curriculum of increasingly hard negative tool calls as a contrastive training signal. Empirical experiments on $V^*$, HRBench and MME-RealWorld show that our approach is competitive while being more efficient. When used as a drop-in replacement for SFT, we even outperform all baselines. To directly measure the zoom-in ability of models, we further introduce the scalable synthetic Muffin&Chihuahua (M&C) dataset. Each image consists of a grid with every cell either showing a muffin or chihuahua. Leveraging the M&C dataset's unique region of interest labels, we find that recall is the metric that most strongly correlates the zoom-in region with final task performance. Our model and code for reproduction is publicly available under https://github.com/UKPLab/emnlp2026-zoom-in

Secret key generation from correlated observations at Alice and Bob, in the presence of an eavesdropper Eve, underpins physical-layer security. Classical pipelines quantize by hand, amplify privacy afterwards, and optimize no objective tied to a key rate. We propose Variational Probabilistic Quantization (VPQ): neural encoders that map the correlated sources directly into a discrete key alphabet, trained by a variational adversarial objective whose entropy, mismatch, and leakage terms match the three terms of the one-way secret key rate. A linear code-offset secure sketch then reconciles the encoder outputs into an identical key without a separate privacy amplification step. We prove that the VPQ losses lower-bound the one-way secret key capacity of the induced source, and derive in closed form the optimal worst-case key rate over the source class of a given alphabet size and mismatch probability, attained by finite-field linear sketches. On Gaussian fading channels, VPQ leaks less to a correlated eavesdropper than one classical and two recent learning-based baselines, and Reed–Solomon reconciliation operates within the predicted finite-blocklength gap.

Plan accuracy alone cannot show whether a speech-delivery decision follows its source: a fixed prior may match the original label yet fail to respond appropriately when a cue changes. VoxReason provides a 100-case verifier benchmark that holds each utterance fixed, edits one designated source-label cue, and scores cited evidence, eight plan fields, and the permitted response. On a source-key-disjoint test of 24 cases, a source-emotion prior reaches plan-slot accuracy 0.958, but none of the 24 edited neutral targets appears in its training labels; its required-change accuracy is 0.000. This diagnoses the support boundary of this prior, not its performance on supported edits. In a complementary 32-case emotion-disjoint test, the prior has seen all edited neutral targets but neither original test emotion; its plan-slot accuracy is 0.219 and required-change accuracy is 1.000. The partitions reuse and overlap the same 100 cases, so these deterministic diagnostics are not independent cohorts or learned-planner results. The benchmark evaluates derived labels and structured plans, not audio input, generated speech, or listener judgments.

This paper investigates over-the-air (OTA) computation enabled online federated learning (FL) in low-Earth orbit (LEO) satellite networks. Specifically, we consider a dual-layer OTA aggregation architecture, where ground devices upload analog model updates to serving satellites via uplink OTA aggregation, and satellites forward the aggregated signals to a data processing center through the second round OTA aggregation. Then, we formulate a long-term data-utilization maximization problem in which devices continuously collect new data and untrained samples gradually lose freshness. The problem is subject to the satellite beam budget, transmit-power limit, and global mean squared error (MSE) constraint that governs end-to-end aggregation distortion. This yields a coupled mixed-integer nonlinear programming (MINLP) problem, involving tightly coupled discrete beam-hopping decisions and continuous power control. Due to the combinatorial action space and nonconvex constraints, the problem is NP-hard and computationally intractable. Furthermore, the time-varying satellite topology and dynamic data generation render it a sequential decision-making problem, necessitating adaptive online scheduling. To address these issues, we cast the problem as a Markov decision process and develop a proximal policy optimization (PPO)-based deep reinforcement learning framework that jointly optimizes adaptive beam hopping and power control, using an MSE-aware reward to balance data utilization and aggregation accuracy. Numerical simulation results verify that the proposed algorithm consistently outperforms other benchmark schemes, achieving superior long-term data utilization and faster FL convergence while satisfying the MSE requirement.