Papers

Dynamic vocabularies built on-the-fly via Lempel-Ziv-Welch (LZW) compression let a language model represent the same text in fewer tokens, but prior work shows that this efficiency comes at a cost to reasoning and perplexity. We show that this quality gap is not a fundamental price of compression but can be substantially closed through targeted improvements to the compression layer and LZW codec, without modifying the underlying pretrained language model. Our analysis reveals four key issues: (1) LZW's compression algorithm creates an implicit conflict between the input embedding and the output embedding, a tension that a single shared embedding cannot resolve; (2) naively applying RoPE to LZW-compressed sequences ignores that merged tokens can span multiple tokens; (3) merging certain tokens (notably digits) disproportionately hurts downstream accuracy; and (4) dynamic vocabularies introduce an inherent segmentation ambiguity that confounds standard evaluation. We address these through decoupled input/output dynamic embedding layer, size-aware RoPE positioning, and a modified LZW codec supporting selective merges and multi-view segmentations. Across four models, our recipe improves perplexity on every corpus we evaluated as well as mathematical reasoning accuracy, closing roughly half the quality gap between zip2zip and the uncompressed baseline at a similar compression ratio on general text. This represents a significant step toward lossless adaptive-vocabulary compression for language models.

Thyroid nodules affect a large proportion of the general population. Although most are benign, a non-negligible fraction is malignant. Ultrasound imaging is the primary diagnostic modality, and accurate nodule segmentation is a prerequisite for CAD systems to classify malignancy and assess metastatic risk.\\ In this work, we propose a thyroid nodule segmentation framework that integrates a DeepLabV3+, channel-attention modules and post-hoc conformal prediction, providing both accurate mask and pixel-level uncertainty estimates. We address thyroid ultrasound nodule segmentation as the first stage of a pipeline that also includes malignancy classification and metastatic risk prediction. We adopt DeepLabV3+, whose Atrous Spatial Pyramid Pooling (ASPP) captures multi-scale context, with an EfficientNet-B5 backbone for its accuracy-efficiency trade-off. Additionally, we test three channel-attention modules: Squeeze-and-Excitation, CBAM and Grouped Channel Attention (GCA), paired with two loss formulations (Dice-BCE and Boundary).\\ Since binary masks do not indicate prediction confidence, we wrap the trained model with a split conformal prediction layer that produces pixel-level uncertainty maps with statistical coverage guarantees.\\ We evaluate all configurations on the TN3K public benchmark using a fixed split and directly compare them with state-of-the-art methods. The plain backbone already improves upon previously reported results, with CBAM + Dice-BCE emerging as the best overall configuration.

Psychology has proposed many principles of cognition. Each, however, was developed in isolation, leaving their combined explanatory power unknown. We assembled a small set of established principles into one model, which we call COntextual Retrieval and Encoding (CORE). CORE consists of a hierarchy of episodic memories into which prediction errors are written and whose encoding, retrieval, and forgetting depend on context. Fitted jointly to 7,036,442 choices from 137 experiments, the model predicts human choices better than black-box models and reproduces 75% of reported behavioral effects when simulated. Moreover, its latent variables show classic signatures of Bayesian learning and event segmentation. Taken together, these findings suggest that a small cognitive core accounts for much of human behavior. They also open a path toward making psychology more than the sum of its parts, by building computational models on the entire experimental record.

Large Audio Language Models (LALMs) achieve strong performance on speech-text tasks, but standard audio encoders produce hundreds of tokens for even short audio clips. Because self-attention scales quadratically in sequence length, this inflates inference-time compute substantially. Current training-free token compression methods prune or merge tokens based solely on coarse attention or feature similarity. However, treating audio tokens as generic, isotropic features smooths over spectral transitions, destroying fine-grained phonetic transitions essential for decoding speech. In contrast, speech signals exhibit highly structured temporal patterns where information density peaks at acoustic transitions. We propose AudioTETRIS, a training-free, two-stage framework that uses the audio encoder's representation geometry as an acoustic prior for token importance before applying attention-based compression. Instead of discarding encoder structure after the encoding stage, AudioTETRIS propagates this prior directly into the LLM decoder to compute a per-token fused score. This score drives progressive top-$k$ token selection across decoder layers without requiring additional training or auxiliary models. Evaluated on Qwen2-Audio-7B, Phi-4-Multimodal, and Audio Flamingo 3, AudioTETRIS retains near-uncompressed performance on several speech tasks using only 25% of audio tokens, matches or surpasses full-token accuracy at moderate compression rates, and consistently outperforms existing training-free baselines.

Diagnostic information in multi-view medical imaging studies, like echocardiography, is distributed across views, each capturing different aspects of cardiac structure and function. In clinical practice, these views are first interpreted through quantitative measurements, which then support specific diagnoses. However, existing echo representation learning methods do not explicitly model measurements as an intermediate representation between views and diagnosis. As a result, individual measurements are not explicitly represented in a form that downstream diagnostic tasks can selectively use, hindering accurate diagnosis. Inspired by this hierarchical clinical workflow, we introduce EchoTab, a measurement-grounded representation learning framework that represents multi-view echocardiography studies using measurement-specific tokens as an intermediate representation between echo views and diagnosis. Each measurement token selectively attends to the views relevant to its estimation and is trained through tabular reconstruction to predict the corresponding measurement value. The resulting tokens form a factorized echo representation, which encode clinically meaningful structural and functional properties and can be selectively used for downstream diagnosis. Unlike the commonly used contrastive approaches, EchoTab does not require negative samples or large batch sizes. Experiments on two datasets demonstrate that EchoTab learns effective measurement-grounded representations that improve diagnosis classification in multi-view echocardiography and its learned attention aligns with established measurement-view relationships used by cardiologists. Our code is accessible at https://github.com/michelleespranita/EchoTab.

Android's inter-process communication (IPC) mechanisms (Activities, Services, BroadcastReceivers, ContentProviders, and URI scheme handlers) form the primary interface between applications and between applications and the system. We present a systematic study of security enforcement comprising 120 evaluations (99 inter-component interface evaluations across five interface categories, 10 platform security-control evaluations, and 11 dedicated positive controls) in over 30 Google and AOSP packages on Android 17 (API 37). Using an expected-versus-observed framework under a zero-permission attacker model, we evaluate whether each boundary's enforcement mechanism functions as designed. We find 97 enforcement gaps (80.8%) and 23 enforcement holds (19.2%). From these observations, we derive a 14-category enforcement gap taxonomy organized across manifest, code, platform, and architectural layers. Within our purposively selected corpus, observed enforcement outcomes differ substantially across interface categories: evaluated Activities and BroadcastReceivers exhibit a 100% gap rate, while ContentProviders, which benefit from framework-provided defenses such as setStrict(), achieve a 27.6% holds rate. We find that all seven evaluated platform compatibility changes with security intent are either disabled or gated at a targetSdkVersion that none of the applications in our evaluated corpus targets. We document 18 custom URI scheme registrations handling sensitive data (authentication tokens, payment credentials, medical information) despite the known unverifiability of custom schemes (RFC 8252 Section 8.1). Our study includes 23 positive controls, calibration cases demonstrating that the evaluation procedure distinguishes functioning enforcement from observed enforcement failures.

The collective-risk dilemma has become a paradigm for studying cooperation in a wide variety of social experiments, initiatives that generate increasingly rich datasets describing individual actions over successive decision-making processes. Beyond the purely descriptive analysis of such empirical datasets, a methodological framework is required to interpret and reproduce the observed dynamics, a scheme that may be skewed by exogenous biases. Here, we introduce a data-driven, non-parametric method for simulating individual decisions together with a validation protocol for assessing the fidelity level of the outcomes they generate. Our methodological contribution is based on an empirical heuristic ensemble: agents are represented by a finite set of behavioural heuristics, probabilities conditioned on different observable features of the game state, endowed with a hard-voting arbitration that determines their per-round final decisions. This reproducible generative framework allows for behavioural inference in different collective-risk dilemma setups: Behavioural heuristics are seeded directly from empirical data assessing conditional probabilities, without a priori assumptions; a plurality vote with uniform random tie-breaking resolves between competing heuristics each round, adding no free parameters; validation follows a pattern-oriented approach, comparing synthetic and empirical macro-indicators against single-heuristic and uniform-random baselines; synthetic data generated allow exploration of scenarios or experimental setups that have not been empirically realised; as a worked example, we instantiate our methodological approach on the xAire citizen science collective-risk dilemma experiment run in Barcelona.

Reinforcement Learning from Verifiable Rewards (RLVR) is a key driver for Large Reasoning Models, but it is vulnerable to "zero-reward collapse", a training failure where a sudden increase in task difficulty causes the learning signal to vanish. We argue that the conventional approach of discarding these failed trajectories is wasteful, as they contain valuable diagnostic information about the model's cognitive gaps, a bright side that can be leveraged for recovery. Existing solutions, however, often regress to static, "off-policy-style" interventions like pre-defined hint libraries or perform post-hoc refinement, failing to provide the adaptive, forward-looking guidance needed to escape the collapse. To address this, we propose a paradigm shift towards on-policy pedagogical intervention and introduce MENTOR, a multi-agent framework that acts as a live mentor. MENTOR diagnoses the root cause of failure in real-time and dynamically generates targeted, instructional prompts to guide the learning agent's next attempt. As a plug-and-play module, MENTOR integrates seamlessly into existing policy optimization pipelines like GRPO without modifying the training infrastructure. Extensive experiments on math and code generation show that MENTOR effectively mitigates zero-reward collapse, outperforming state-of-the-art baselines. Notably, it achieves massive average relative gains of up to 78.3% on Llama-3.2-3B-Instruct over its base model.

Prediction-market prices combine traders' judgments about future outcomes, but their errors may be concentrated in particular price ranges and change as settlement approaches. Prior cross-category evidence shows that Polymarket weather markets have larger pricing bias than other major categories, without separating low-price longshots from high-price favorites. This study compares the two ranges at different horizons in daily maximum-temperature markets. Event-level probability scores, cubic B-spline recalibration, and Buy-NO tests are used to examine the form and effect of the bias. Longshot overpricing mainly occurs at distant horizons, whereas favorite overpricing becomes much stronger during the final 1-12 hours and is nearly one order of magnitude larger. Event-level scores improve toward the daily temperature peak, while simple recalibration yields no statistically significant gains. The Buy-NO results in the corresponding time windows also show that the main pricing error is concentrated at the favorite end.

The proximal Galerkin algorithm is a structure-preserving solver for variational inequalities that results in discretization-independent nonlinear iteration counts. However, in large three-dimensional simulations, the arising Newton saddle point systems cannot be feasibly solved by direct methods. To leverage an iterative linear solver, we propose a robust and scalable block preconditioner. Instead of using the usual dual Schur complement that acts on the latent variable, we design a preconditioner for the primal Schur complement. This leads to a block preconditioner that requires two approximate inverses of the latent block and only one of the primal Schur complement. The latent block represents an algebraic equation which, after a small perturbation, can be cheaply inverted. We then introduce two choices of approximations for the primal Schur complement: a PDE-operator-based and an algebraic approach. Either approximation can be efficiently inverted by geometric multigrid. We investigate the effectiveness of these strategies when proximal Galerkin is applied to obstacle, Signorini, and strain-constrained elasticity problems and also compare with a monolithic multigrid approach. Choices of standard relaxation methods, in the block preconditioner route, lead to robust iteration counts with respect to the mesh size and discretization degree. Moreover, in the context of the Signorini, specialized relaxation and prolongation also result in effective solvers for nearly-incompressible elastic materials. The linear solves are parameter-robust in 2D, with only mild growth in 3D. This allows for the solution of a three-dimensional contact problem, with a Poisson's ratio of 0.499, discretized with over ten million degrees of freedom.

Recent results [arXiv:2602.04378, arXiv:2602.22608] have established worst-case objective-error lower bounds of order $1/T^2$ after $T$ linear minimization oracle (LMO) calls, matching the upper bound [arXiv:1406.1305] in its dependence on $T$ and thereby resolving a long-standing open question. Upon close inspection, however, both lower bounds are somewhat brittle. The first [arXiv:2602.04378] applies only to the standard Frank–Wolfe algorithm, and, as we show, its prescribed slow exact-line-search trajectories can amplify sufficiently small initialization errors exponentially, corresponding to roughly one additional bit of initialization accuracy per iteration. The second [arXiv:2602.22608] obtains lower bounds for any deterministic LMO-based first-order method via a resisting oracle that adversarially adapts to the queries of the algorithm. In both cases it is conceivable that randomization may circumvent these lower bounds; in particular, neither establishes a lower bound for unrestricted randomized first-order LMO methods. We close this gap by proving an $\Omega(L\operatorname{diam}(S)^2/T^2)$ distributional lower bound for randomized first-order linear minimization oracle methods over strongly convex sets. While our construction reuses the family of [arXiv:2602.22608], our proof characterizes the conditional distribution of the hidden permutation after adaptive queries and bounds the expected decrease of a combinatorial statistic controlling optimizer variance. We further show that the same error threshold holds with probability at least $3/4$ and yields a stopping-time tail bound.

Uncertainty quantification for electric vehicle (EV) charging demand is essential to turn the increasing pressure of EV adoption on power systems into a source of flexibility. We study a large-scale dataset of public charging stations in Scotland, where demand exhibits a hierarchical structure, organized across multiple aggregation levels, from individual charging points to regional and national scales. The charging infrastructure is continuously evolving, with new stations deployed and existing ones occasionally removed, which calls for forecasting and uncertainty quantification methods tailored to this evolving structure. To address these challenges, we combine hierarchical forecasting with recent advances in conformal prediction with missing data. The conformal procedures considered produce valid prediction intervals at every node of the hierarchy and the algorithm CP-MNR-Nested successfully accounts for the time-varying hierarchical structure of the data.

We'll call a map from basic to clopen sets of binary sequences an encoding if it preserves uniform Lebesgue measure and the disjointedness relation. For some notion of randomness $R$ we can ask if there is a computable encoding $E$ such that for every non-Martin-Löf-random (non-MLR) sequence $\alpha$, the sequences in $E(\alpha)$ are non-$R$. Thus we can use $E$ and randomness test for $R$ to construct a randomness test for MLR. We show that there is no such $E$ for partial computable randomness (PCR), but there is one for Kolmogorov-Loveland randomness. Indeed, there is a total computable $E$ and a pair of total computable non-monotonic betting strategies such that on every sequence in $E(\alpha)$ at least one of the strategies wins whenever $\alpha$ is non-MLR.

Detectors of machine-generated text score almost perfectly on scientific-literature benchmarks, implying we can reliably distinguish with these detectors human and machine-generated documents. We looked at ten scientific-document subsets from six popular resources, and on several of them that apparent ease is a side effect of how the documents were put together. A logistic regression using 27 superficial features reached 0.833-0.997 macro-F1, and on the three arXiv subsets it outperformed every detector we evaluate with F1-scores 0.951-0.997. The strongest cue to the detector is the PDF extraction artifact: human texts retain their original line breaks, whereas generated texts are extracted as a single block of prose. This cue, together with three other formatting features, tell the classes apart at 0.996 AUC. To distinguish benchmarks that contain such a signal from detectors that rely on one, we introduce format counterfactuals: just re-wrapping a machine-written abstract without changing a single word reverses 20-27% of a public detector's correct decisions. Normalizing formatting before training reduces the generalization gap from 0.32 to 0.22. On format-clean benchmarks, it also raises out-of-domain macro-F1 from 0.33 to 0.65 for detectors trained on artifact-containing subsets.

Pre-trained robot policies, such as Vision-Language-Action (VLA) models, provide a strong foundation for generalist robotic manipulation, but adapting them to new tasks with only a few demonstrations remains challenging. Directly fine-tuning a pre-trained policy often achieves strong in-distribution performance, yet it can distort the representations learned during pre-training and weaken the model's generalization ability under out-of-distribution shifts. We propose Retrieval-Anchored Representation Alignment (RARA), a two-stage adaptation method, to address this issue. First, RARA retrieves representation features from pretraining with similar task intent and motion, and aligns the policy's representations of the target demonstrations with the pre-trained representations of the retrieved demonstrations. Second, RARA fine-tunes the policy on the target demonstrations while limiting how much its representations of the retrieved demonstrations change during fine-tuning. By reducing the representation mismatch before action learning and preserving pre-trained knowledge during adaptation, RARA limits unnecessary representation drift during fine-tuning. Experiments with Diffusion Policy and GR00T on 13 tasks in RoboTwin, RoboCasa, and on a real robot show that RARA improves the average success of fine-tuning by 9.5 points in-distribution and 13.6 points out-of-distribution, and outperforms all compared baselines. Project page: https://shua-kang.github.io/RARA_robot/

This paper presents a path-following guidance framework for a turn-rate-constrained unmanned aerial vehicle (UAVs). The proposed guidance strategy uses commanded course angle as a tanh-hyperbolic shaping function of the lateral deviation from the desired path. Stability characteristics illustrating asymptotic convergence to the desired path are established using Lyapunov analysis. Analytical expressions for the turn-rate and corresponding maximum turn-rate bound are derived, enabling direct gain selection for a turn-rate-limited vehicle. Numerical simulations are performed for straight-line following scenarios under both ideal and non-ideal initial heading conditions. Furthermore, a comparative study against representative arctan-, arcsine-, and arccos-based vector-field guidance laws is conducted using analytically feasible gains selected from their respective turn-rate bounds. Comparative numerical results demonstrate the proposed formulation achieves reliable boundary convergence while maintaining smooth steering behavior and a compact analytical gain-design framework.

Feature selection is a central problem in high-dimensional financial regression, where large numbers of engineered variables and strong correlations among predictors reduce model stability and interpretability. $L_1$ regularization (Lasso) is widely used for sparse regression, but its shrinkage can be restrictive when informative signals are spread across correlated variables. Exact sparse regression avoids this shrinkage but requires solving a combinatorial optimization problem that is computationally difficult and is typically fit to minimize in-sample rather than out-of-sample error. This paper presents a feature selection framework based on Monte Carlo Tree Search with Generalized Nested Rollout Policy Adaptation (GNRPA). Feature selection is formulated as a sequential decision process, and candidate subsets are scored by their out-of-sample validation performance rather than training error. The method also supports a lexicographic objective accounting for subset size and multicollinearity. We study this approach on a contemporaneous explanatory regression task: explaining the weekly return of the NASDAQ-100 index from 1,072 engineered features spanning market, macroeconomic, and factor data. This isolates the value of subset selection under multicollinearity from the separate problem of predictive forecasting. On this task, the proposed method identifies sparse feature subsets with higher out-of-sample explanatory power than an optimized Lasso baseline, reaching an out-of-sample $R^2$ of 0.6961 versus 0.6313 for Lasso. These results suggest that Monte Carlo Tree Search is a practical alternative to convex regularization for sparse explanatory feature selection in financial regression, with potential value for factor attribution and risk decomposition where transparency matters as much as out-of-sample fit.

Materials scientists increasingly combine information from several public databases, each with its own interface, schema, units and calculation conventions. MatQuery is a prototype and experimental infrastructure for studying how a natural-language request can be turned into queries over such databases and how the returned values can be interpreted. This report describes what has been built and the preliminary results obtained so far. The prototype holds frozen, self-verifying local copies of six public materials databases (about 10,000 records each), a catalog of declared fields derived from them, a database-independent query representation with an adapter for each database, and an eight-step pipeline in which a language model proposes and deterministic program steps check, execute and assemble. A benchmark generator produces questions whose reference queries are verified by execution on the frozen copies and whose wording is checked by back-translation and review. An evaluation harness runs direct prompting, CHESS and Alpha-SQL on the same frozen basis. Preliminary experiments on an 800-question development set show that direct generation with three demonstrations reaches 82.4% strict execution accuracy (56.5% without demonstrations), that a low-cost scorer selects the exact set of databases for 85.0% of complete requests with well-calibrated probabilities, and that band gaps differ by more than 0.5 eV in 31.6% of composition-matched cross-database clusters. Two pilots show the limits of the current approach: a probe-then-ask policy gave answers rated valid by both reviewers in only 11 of 50 sessions, and only 115 of 300 sampled record pairs could be compared by structure. The complete pipeline is in smoke testing and has not yet been evaluated at scale.

Humanoid tennis requires reactive whole-body coordination under rapidly changing ball dynamics, yet collecting task-specific motion-capture demonstrations that cover diverse approach states, contact locations, and timing conditions is costly. Inspired by how humans acquire new skills by combining reusable motor experience with task-specific demonstrations, we present ATHLETE, a framework for learning reactive humanoid tennis from generic locomotion data and ordinary tennis videos. ATHLETE combines LAFAN1 locomotion with video-derived striking motions and uses Trajectory-Guided Motion Matching (TGMM) to construct diverse task-conditioned whole-body references. These references are transformed into a reactive policy through physics-based teacher–student learning with online DAgger distillation, followed by PPO fine-tuning for the tennis task. The resulting policy responds directly to robot and ball observations without requiring reference-motion inputs at deployment. On a real humanoid robot, ATHLETE achieves forehand and backhand hit rates of $90.00\%$ and $86.67\%$, respectively, and enables multi-shot rallies with human players. These results demonstrate that ATHLETE can learn dynamic humanoid tennis skills from reusable locomotion data and video demonstrations. Result videos and code can be found at https://anonymous.4open.science/w/ATHLETE2027/.

We study simultaneous competitive and regret guarantees for TCP acknowledgment. Each acknowledgment serves all pending packets, and the objective is to minimize the total acknowledgment and waiting costs. It is well known that no deterministic online algorithm achieves a competitive ratio better than $2$. Given $k$ randomized online experts and $0<\varepsilon<1$, we construct a randomized online algorithm whose realized total cost is always at most $(2+\varepsilon)$ times the offline optimal cost $\mathrm{OPT}$ and whose expected regret against the expert with the smallest realized cost is $O(\sqrt{\mathrm{OPT}\log(k+1)}+\log(1/\varepsilon))$. The algorithm requires no advance knowledge of the number of packets or $\mathrm{OPT}$. Our algorithm controls switching costs between experts and uses the gap between the permitted competitive bound and its current cost to determine when to learn. We also prove lower bounds showing that, for $k\ge 2$, this regret bound is asymptotically optimal in $\mathrm{OPT}$, $k$, and $\varepsilon$, and that $\varepsilon>0$ is necessary for sublinear regret.

Reasoning models can benefit from test-time scaling, but continued reasoning may become ineffective or harmful, a phenomenon termed overthinking. Evaluations against full thinking compare accuracy and token usage but lack references for overthinking severity and strategies' theoretical mitigation potential. To characterize overthinking at the model-task level, we first introduce Step-wise Truncation Probing (STP), repeatedly generating answers from multiple retained prefixes along each fixed sampled reasoning trajectory. Using each trajectory's earliest measured position attaining maximum empirical accuracy as a reference, we define Overthinking Ratio and Accuracy Degradation as the fraction of thinking tokens generated afterward and the accuracy decrease from this reference to full thinking, respectively. Secondly, we then formulate dynamic routing, fixed budgets, and dynamic early exit within a unified framework to quantify the theoretical mitigation potential of different termination strategies. Based on this unified framework, we use the complete STP measurements to construct a token-accuracy Pareto frontier for each strategy under a common normalized thinking-token budget. We further develop OTBench, a reusable benchmark built from the STP process, which produced 70.5B reasoning tokens and the corresponding probing measurements across six models and four benchmark groups.

We study concurrent stochastic games in which a team of agents plays against an adversarial environment. If the agents could access a shared source of randomness, then these games reduce to two-player games. Despite being common in the literature on coalition games, this assumption does not accurately model real-world systems. In this paper, we consider the setting where the agents only have access to individual sources of randomness, and must satisfy a given reachability or safety objective. The limit-sure problem asks whether the team can ensure that the objective is satisfied with probability at least $1-\lambda$ for every $\lambda>0$. For reachability objectives, we show that this problem is $\mathsf{NP}$-complete. For safety objectives, we show that limit-sure winning coincides with almost-sure winning, which is known to lie in $\mathsf{P}$. As a corollary of these results, we conclude that model checking of the qualitative fragment of the logic Individually Randomised ATL – introduced in the recent work of Brice et al. – is $\mathsf{P}^{\mathsf{NP}}$-complete.

We present Rationally Linearized Multilayer Perceptrons(RaLiMPs), a new approach to efficiently linearizing MLPs using low-degree rational approximations of standard activation functions. Inspired by polynomial-based MLP linearization, we show that rational approximations can provide accurate representations at substantially lower degree. RaLiMPs exploit these approximations to construct kernel representations with narrow bottlenecks, enabling efficient composition across MLP layers. We provide theoretical guarantees for the resulting approximation and evaluate RaLiMPs across applications ranging from implicit 3D models to large language models. Our experiments demonstrate favorable accuracy–efficiency tradeoffs, with RaLiMPs achieving competitive or improved accuracy while substantially reducing inference cost.

Repository-level issue-resolution benchmarks such as SWE-bench evaluate submitted patches primarily based on whether they pass the correctness tests, leaving the runtime performance of different test-passing patches largely unexplored. Evaluating patch performance is essential but challenging since designing performance workloads for each benchmark instance is costly. However, these benchmarks already provide a natural candidate: the built-in FAIL\_TO\_PASS (F2P) tests, which exercise the behavior affected by the issue fix. We therefore investigate whether the provided F2P tests can be repurposed to evaluate patch performance. Overall, we choose four frontier models, Claude Opus~5, GPT~5.6, DeepSeek V4 Pro, and Qwen~3.8 Max, paired with SWE-agent to resolve the pre-filtered measurable corpus from verified splits of SWE-bench and SWE-bench-Live, yielding 262 cases with at least two different correct solutions, including the developer ground-truth patch. We then measure each solution's runtime performance in instruction count and wall-clock time on the native F2P test, and use Welch ANOVA instead of the RSD proposed in prior work as a more robust statistical discrimination criterion. We find that this methodology exhibits notable discrimination capability in runtime performance and identifies performance differences among patches in almost half of the instances. Our empirical study shows that the runtime performance of generated patches is independent of the issue resolution rate of LLMs, with nearly reversed model rankings. The case study also shows that the performance analysis reveals architectural regression in some patches that the correctness evaluation overlooked. These findings suggest that resolution rate does not adequately characterize LLMs' capability for repository-level issue resolution, motivating future evaluations to consider patch performance alongside it.

Emotional image editing aims to change the emotion a scene conveys while preserving its essential content. Because the same scene can support different emotional interpretations, achieving an emotional goal requires visual changes grounded in the source context. Motivated by the Component Process Model, which links emotional responses to event appraisal, we investigate neural representations as a reference for these editing decisions. Using TRIBE v2, an fMRI encoding model, we analyze 400 source-target pairs across eight emotion categories. Hidden states from EmoEdit, EmoEditor, and AffectDelta correspond more strongly to predicted target-image responses in visual regions than in affect-related regions. This imbalance motivates ROI-ReFT, which combines source visual responses as a content reference with a predicted affective response change as an editing direction. Two interacting branches use both conditions to guide a frozen editor throughout generation. Predicted responses of the generated image supervise affective change and constrain visual drift. On a post hoc selected subset of 400 editing requests, ROI-ReFT achieves 91.75% target-emotion accuracy and 83.25% joint success, with the same output both matching the target emotion and reaching source-output DINO-I of at least 0.80. Our code and dataset will be publicly released.

Anticipation of vessel trajectories is an essential task when navigating at sea. This applies to both human navigators and autonomous surface vessels. Trajectory forecasting enables sub-routines such as collision avoidance to assess the situation and react accordingly in time. This paper investigates Transformer and Performer models for trajectory prediction of surface vessels in confined waters. Using widely available position data from the Automatic Identification System (AIS), candidate models are trained and evaluated. Historical information on how other vessels have moved in the local area surrounding each agent is conveyed using positional heat maps. The heat maps are generated using Kernel Density Estimation (KDE) and encoded with a Convolutional Neural Network (CNN) before entering the Transformer models. The models are trained with quantile loss to efficiently obtain a model-aware prediction interval, as this is crucial for timely collision avoidance. The proposed model architectures are shown to greatly outperform previous work while the inclusion of local heat maps enables the prediction of vessel movement that is not otherwise indicated from the previous positions.

This paper presents a data-driven approach to lossless and lossy source compression using polar codes. Classical polar source coding achieves fundamental limits with low complexity for binary memoryless sources. However, for unknown q-ary sources with memory, successive-cancellation (SC) decoding requires an explicit source model and can be computationally prohibitive. We present analytical SC recursions and show that exact implementations scale quadratically with alphabet size and cubically with state-space size. To overcome these limitations, we develop neural polar decoders trained directly from data to approximate conditional distributions of synthetic polar coordinates without requiring an explicit source model. These learned distributions enable estimation of entropy rates, mutual information, and rate-distortion functions, and support explicit polar compression schemes. For lossless compression, we identify high-entropy polarized coordinates and reconstruct the remaining coordinates using the learned SC recursion. For lossy compression, we combine neural polar decoders with a neural test channel to estimate and optimize the mutual information-distortion tradeoff and construct an explicit lossy polar code. We also extend the framework to Wyner-Ziv compression with decoder side information by learning conditional polarized distributions and a reconstruction mapping. Experiments on memoryless sources, sources with memory, and real genomic data show that the learned decoders closely match analytical SC decoders when available and remain applicable when conventional methods are computationally infeasible or the source model is unavailable.

Cross-view geo-localization remains challenging due to drastic appearance variations and geometric distortions caused by viewpoint changes. Existing methods primarily rely on discriminative feature learning or explicit geometric alignment, yet often overlook an important observation: humans tend to match scenes by integrating the spatial context of multiple landmarks rather than relying solely on isolated local visual cues. Inspired by this, we propose a Spatial Context Modeling (SCM) module, which traverses feature maps along four orthogonal directions and models interactions among the resulting directional representations at each spatial position. By aggregating complementary directional observations, SCM constructs richer scene layout representations. Since reliable spatial context depends on robust visual features, we introduce a High-Frequency Information Guidance (HFIG) module that extracts stable structural cues from the frequency domain. By emphasizing edges and contours that are less sensitive to appearance variations, HFIG encourages the model to learn more robust representations. Extensive experiments demonstrate that our method achieves competitive performance on standard cross-view geo-localization benchmarks while exhibiting strong robustness under more challenging scenarios with unknown orientations and limited fields of view (FoVs).

Recent speech LLMs can leverage rich paralinguistic cues, enabling more complex and personalized interactions. However, little work investigates how this sensitivity can lead to stereotyping, and existing evaluations largely measure demographic disparities without distinguishing beneficial personalization from stereotypical behavior. We introduce BASIS-Bench, a benchmark designed to disentangle these two aspects through three axes: (1) demographic robustness, measuring disparities in content-based tasks across speakers and demographic classification capabilities; (2) stereotypical bias, measuring whether perceived demographic cues trigger unprompted stereotypical behavior; and (3) personalization, measuring whether models can tailor responses to demographic characteristics when explicitly requested. We evaluate eight recent speech LLMs and find that, despite often recognizing demographic cues, most fail to translate them into appropriate personalization. Explicit demographic labels improve personalization but also amplify stereotypical associations already present in the textual backbones. These findings highlight the need to train speech LLMs not only to perceive demographic cues, but to learn when they should shape their responses.

Text-guided long-term human motion generation is essential for applications such as continuously operating digital avatars. Despite recent progress on short clips, existing models struggle with multi-segment instructions that involve anaphora, where constraints established in earlier segments must be preserved over time. In this paper, we introduce a novel task of long-term motion generation from such anaphoric compositions. To mitigate this, we present ELMA, an episodic long-term motion dataset with dense segment-level annotations that preserve cross-segment dependencies. Built via a fully automated collection and filtering pipeline, ELMA extracts high-quality, continuous 30-second clips from in-the-wild videos, capturing the episodic narratives necessary for long-range state modeling. Leveraging this dataset, we fine-tune state-of-the-art motion generation models to incorporate long-range context. Extensive experiments reveal fundamental shortcomings in current models when handling context-dependent instructions. We demonstrate that ELMA enables significant improvements in both global physical realism and semantic alignment, setting a robust new baseline for long-term motion synthesis.