Papers

Supervised fine-tuning (SFT) is the de facto standard for adapting large language models (LLMs) to target domains, but it often degrades the model's general capabilities, a phenomenon known as catastrophic forgetting. Existing approaches typically modify the SFT loss to mitigate forgetting, but they inevitably operate along a domain-generality trade-off. In this work, we step outside this trade-off by decoupling the two capabilities at the model level: we keep the original base model for general capability, and selectively invoke the SFT expert only when domain-specific knowledge is required. Specifically, we propose CPR (Critical-Point Routing), a token-level routing framework between a base model and its expert derivative, based on critical tokens where the base model fails but the expert succeeds. We train a lightweight hierarchical router that estimates the expert-call probability per token, and pair it with a tailored inference procedure that combines momentum smoothing and threshold gating. Across diverse model-domain configurations, CPR achieves state-of-the-art across all settings, surpassing SFT expert by 1.4-5.5% in domain performance while recovering its general-capability drop from 3.4-14.5% to at most 0.5%, with minimal overhead from invoking the expert on only one-third of tokens.

Graph-based models have been shown to provide a structured representation for complex multi-domain energy systems but face limitations when edge power flows depend on non-adjacent states or when a single edge carries multiple power-flow types driven by different inputs. This paper proposes two general extensions to address these limitations: a recursive state-to-input feedback scheme that embeds non-adjacent state dependencies into edge inputs without altering the graph structure, and a parallel edge decomposition method that represents composite interactions using sets of single-input edges while preserving energy conservation at the vertices. The extended framework is demonstrated on a lithium-ion battery module consisting of 36 parallel cells, and the resulting model predicts module temperatures with errors below 1{\deg}C. Validation on this electro-thermal battery system demonstrates the effectiveness of the extended framework for multi-domain systems that cannot be represented by previously established graph-based formulations, and indicates its potential for broader application to complex energy systems in control and design studies.

Plane geometry problem (PGP) solving has become a critical benchmark for multimodal reasoning because it requires accurate visual perception and precise multi-step symbolic deduction. Although test-time scaling (TTS) has demonstrated remarkable success in general mathematical reasoning, it fails to scale effectively under the symbolic-program paradigm for plane geometry. We identify two key obstacles: limited reasoning diversity induced by rigid symbolic programs and insufficient explicit visual grounding before symbolic deduction. To address these issues, we propose Multi-Trace Synthesis (MTS), which converts each symbolic program into heterogeneous reasoning traces, including executable Python scripts and CoT-augmented variants. We further propose Perception-Augmented (PA) training, which parses diagrams into structured semantic clauses before deduction, and Consensus-Guided Multi-Trace Ensemble (CG-MTE) for efficient self-adaptive inference. Experiments on three geometry benchmarks show that our method consistently improves PGP-solving across model scales and achieves strong performance against both general-purpose MLLMs and specialized geometry solvers. Under test-time scaling, CG-MTE achieves comparable accuracy to high-budget self-consistency while reducing sampling cost by up to 8x. Code and data are publicly available at https://github.com/Jason8Kang/ReTTS-PGPS.

Graph positional encodings are widely used in graph neural networks and graph Transformers, yet it remains unclear when the code itself can identify nodes. We study a hybrid distance-spectral encoding that combines anchor-distance profiles with quantized low-frequency Laplacian-energy coordinates. Treating the encoding as an observation map yields a simplex-refined converse, an exact collision factorization \(\kappa_H=\kappa_D\kappa_{S|D}\), and the collision information \(I_H=-\log\kappa_D-\log\kappa_{S|D}\). On random regular graphs, the criterion is made explicit through a bounded-correlation Gaussian-wave surrogate; for actual Laplacian-energy coordinates, we give the distance-conditioned spectral collision condition sufficient for conditional actual-coordinate achievability. Experiments show that \(I_H/\log n\) calibrates localization success, and PE-only structural task probes on Universal Dependencies trees show that hybrid encodings better recover syntactic-tree geometry than distance-only or spectral-only baselines.

Individual tree measurements derived from Light Detection and Ranging (LiDAR) mounted on Unmanned Aerial Vehicles (UAV) provide valuable information for forest inventory, ecosystem monitoring, and sustainable forest management. Recent advancements in machine learning have increased the demand for annotated datasets to develop and evaluate point cloud-based approaches, especially for individual tree segmentation. However, publicly available annotated UAV-LiDAR datasets in temperate forests are limited. In this article, we present CedarCypress3D, a manually annotated UAV-LiDAR dataset collected in Japanese cedar (Cryptomeria japonica) and Japanese cypress (Chamaecyparis obtusa) plantations in Japan. The dataset consists of UAV-LiDAR point clouds and field survey measurements from 34 circular plots across two sites with different topographic characteristics, along with terrestrial LiDAR point clouds available for a subset of 22 plots. A total of 1,627 trees were measured in the census field survey and manually annotated to match the corresponding trees in the UAV-LiDAR point clouds. For the subset of plots with terrestrial LiDAR data, semantic labels (i.e., stem and non-stem) were additionally assigned to tree points in the UAV-LiDAR data. CedarCypress3D provides high-quality annotated UAV-LiDAR data for developing and evaluating individual tree instance segmentation and semantic segmentation methods in temperate planted forests. The dataset can also support research on tree attribute prediction and multi-platform LiDAR analysis. The dataset is publicly available at https://doi.org/10.5281/zenodo.22168721.

Large language model (LLM) agents are increasingly deployed in long-horizon, interactive, and stateful environments. In these settings, a single wrong action, such as refunding the wrong purchase, can cause irreversible task failure and must be intercepted before execution. Such failures may not appear in every single run, but can emerge across repeated trials, making reliability across steps and trials critical. However, ensuring agentic reliability is challenging: even frontier LLMs struggle to explain why an action may be wrong, especially in long, intertwined trajectories governed by domain-specific policies. Much recent work relies on prompt-based critique agents, while optimization-based methods lack a systematic way to produce rich verification rationales for training. We address this gap with CAST, a critique-aware training framework that converts sparse task outcomes into action-level supervision for critique learning and policy optimization. CAST analyzes agent trajectories to synthesize structured rationales explaining action validity under partial observability. The resulting critique model is used to construct critique-aware training data for optimizing the policy model. Fine-tuning Qwen3-family models on dynamic tool-calling benchmarks, CAST improves reliability across domains, outperforming GPT-OSS-120B by over 10% pass^4 on Retail tasks and yielding an additional 9% improvement on Telehealth in an out-of-domain setting. These results demonstrate that critique-aware training improves the robustness of LLM agents in realistic dynamic environments.

Citation accuracy has been studied for decades because of its importance to research reliability. Content-level citation verification assesses the reliability of scholarly claims. Recent work adopts a two-stage retrieval-classification framework inherited from fact-checking. However, this design overlooks the complexity of the raw citing claim and introduces three issues into the verification system, namely scope mismatch, perspective mismatch, and proposition entanglement. These issues increase the difficulty of retrieval and classification, thereby limiting model performance. Motivated by this gap, we propose claim normalization, which applies three rewriting strategies to the raw citing claim before retrieval and classification, allowing each downstream model to perform a single, well-defined task. Building on this method, we develop Claim-Normalized Citation Verification (CNCV), a new three-stage framework consisting of claim normalization, evidence retrieval with grounding, and citation classification. We evaluate CNCV across 18 classifiers using a factorial experiment on human-annotated citation instances. Compared with the prior two-stage framework, CNCV improves macro F1 by an average of 12% for encoders and 10% for generative LLMs, driven by improved evidence quality, the dominant factor identified in our experiments. Evidence retrieved from automatically normalized claims yields downstream classification performance statistically equivalent to that obtained with manually annotated evidence.

Vision-Language-Action (VLA) models are reshaping autonomous driving (AD) by unifying perception, reasoning, and control through language, enabling semantic grounding, interpretable decisions, and better long-tail generalization. But language is expensive onboard: latency and memory budgets are tight, and autoregressive decoding is inherently sequential. This work reframes the central question as when and where language should act at inference, since inference cost recurs at every deployed frame while training cost is paid once. We introduce the Language Residue taxonomy to organize methods by their inference-time use of language: train-time-only supervision (L1), latent non-textual reasoning (L2), conditional invocation (L3), and full per-frame generation (L4). We review representative methods and tag each across five deployment axes (latency, parameters, memory, FLOPs, tokens), analyzing them on major open- and closed-loop driving benchmarks (e.g., nuScenes, NAVSIM, Bench2Drive). We further trace how efficient methods from NLP/LLM are adapted in AD, identifying the constraints and motivations driving these adaptations. A continuously updated repository will be available at Github.

We characterise the measurement error of a free- hand consumer structured-light scanner used to derive three- dimensional geometric wound descriptors. Rigid wound-care phantoms cannot change, so every difference between repeat scans of one site is measurement error; all repeats come from a single scanner unit, nine sites and 23 scans, so this charac- terises one instrument. The 95 percent repeatability limit for reconstructed surface area is a factor of 4.8, with a confidence interval from 3.0 to 7.0, so rescanning an unchanged site can shift the reading from a 79 percent decrease to a 377 percent increase. Forty-four of 45 descriptors fall below an intraclass correlation of 0.50, and none of 248 descriptor and pipeline combinations reaches 0.75. An error budget formed by holding the analy- sis region fixed, leaving sensor and reconstruction untouched, bounds the share of variance attributable to how much surface the operator captured at 75 percent for surface area, 91 percent for hull area and 95 percent for bounding-box diagonal; only hull volume is majority instrumental, at 48 percent, so a better sensor would buy little. No wound is delineated anywhere in the chain, so the comparison against the four-week area reduction used clinically to predict healing, a ratio near 2.1, is a lower bound on the noise an unsegmented pipeline must overcome, not a measurement of wound-area reproducibility. Standardising the analysis region cuts the limit to 2.13, meeting that ratio rather than clearing it. Statistical outlier removal imposes a measured systematic area deficit near 11 percent.

People with a poor sense of direction (SOD) struggle to build cognitive maps for effective spatial navigation, and existing navigation tools prioritize efficiency over spatial learning. To understand how navigation strategies differ by ability, we conducted a landmark attention study with 20 participants (ten good SOD, ten poor SOD) who navigated across four Tokyo neighborhoods in virtual reality (VR). We found systematic group differences in both gaze behavior and the types of landmarks they verbally identify as effective. Based on these findings, we built LandmarkLens, a mixed-reality (MR) navigation system that uses a vision-language model (VLM) to identify and highlight navigation-relevant landmarks. A follow-up study with eight poor-SOD participants showed improved performance in scene recognition, suggesting that guided landmark attention can support landmark-level spatial knowledge acquisition for people with poor SOD, a first step toward broader spatial learning.

Preference optimization is widely used to align large language models with human preferences, but preference-data composition may also influence privacy-relevant memorization. We examine whether adding synthetic privacy-preference pairs to Direct Preference Optimization (DPO) is associated with lower canary-based memorization signals without modifying the objective or introducing a formal privacy mechanism. We propose Privacy-Pressure Preference Mixing (P3M), a data-composition protocol that varies the amount of privacy-preference data while keeping helpfulness and harmlessness preference data fixed. We evaluate a non-privacy Baseline and privacy-mixing ratios of 0.5, 1.0, and 2.0 using Gemma 3 270M-IT across five random seeds and validate the same four conditions using 4-bit-quantized Gemma 2 2B-IT across three seeds. Overall, under the tested conditions, privacy-preference mixing is associated with lower mean canary suffix log-likelihood proxy values across both model settings and lower aggregate membership-inference attack performance relative to the Baseline in the mixed-source 2B evaluation. Specifically, across the privacy-aware 2B configurations, the mean area under the receiver operating characteristic curve (AUROC) ranges from 0.596 to 0.629, and the mean area under the precision-recall curve (AUPRC) ranges from 0.541 to 0.575, compared with 0.804 and 0.790, respectively, for the Baseline. However, the reduction in membership distinguishability does not hold uniformly across data sources. Moreover, the relationship between the privacy ratio and harmlessness preference accuracy varies by model setting, whereas helpfulness preference accuracy remains broadly stable. These findings suggest that P3M should be viewed as a lightweight empirical protocol for examining privacy-utility-safety trade-offs rather than as a formal privacy guarantee or a defense against extraction attacks.

Stochastic programming (SP) is a powerful framework for decision-making under uncertainty, but its practical adoption in industry is often hindered by the difficulty in understanding the causal relationships that drive optimal solutions. In the two-stage SP, strategic first-stage decisions are coupled with operational second-stage recourse decisions. When the number of scenarios under consideration is large, understanding the direct link between the uncertainty realization and optimal recourse strategy becomes computationally and cognitively demanding. Common approaches to improve interpretability include trained classification trees or scenario reduction, replacing the large scenario set with a representative subset. This is often achieved through post-hoc clustering (e.g., k-means) based on uncertainty realizations or optimal recourse decisions. While useful, these methods only provide a statistical approximation of the solution space and may fail to reveal the underlying structural properties of the recourse problem that drive optimal first-stage decisions. This work introduces a novel, deterministic approach to explainability using multiparametric programming (mp) within a Benders decomposition framework. We reformulate the recourse subproblem as a multiparametric linear program, generating an explicit map of Critical Regions (CRs), which are polyhedral partitions of the uncertainty space. This allows us to cluster scenarios analytically rather than statistically. We demonstrate this methodology on a supply chain planning under demand uncertainty. Our results show that 100 stochastic scenarios map to exactly six critical region clusters. This mapping allows us to explain optimal capacity planning decisions as a precise trade-off between specific operational modes, providing a fully transparent interpretation of the stochastic solution.

Speculative decoding accelerates large language model inference by using a draft model to generate candidate tokens, which are verified by the target model in a single forward pass. Verification proceeds sequentially and discards every position from the first rejection onward, yet existing draft training relies on token-level imitation of the target with a fixed per-position weighting that reflects neither property. We introduce Verification-Aware Training (VAT), a plug-in framework that simulates verification at every training step and turns the resulting accept and reject patterns into supervision. VAT consists of two components: (i) a verification head, a lightweight jointly trained binary classifier that supervises the draft model on whether each position survives sequential verification; (ii) verification-adaptive weighting, which replaces the fixed weighting schedule by keeping full weight up to each sample's first rejection point and re-anchoring the decay to start there. VAT modifies only the training objective, so it can be layered on top of existing methods without changing the draft architecture, the target model, or the inference procedure. Applied to EAGLE-3 and DFlash on Qwen3-4B, Qwen3-8B, and LLaMA-3.1-8B, VAT improves average acceptance length by up to 11.4% and wall-clock speedup by up to 8.7%, with consistent gains across math, code, and chat benchmarks. Code will be available at https://github.com/naver-ai/vat

Retrieval-augmented language models can fail to respect negative constraints when the retriever supplies evidence about concepts the user explicitly excluded. Beyond explicit negation, queries may ask for answers that include one concept while excluding another, or for entities that belong to a category but differ from a closely related instance. Because the excluded concept still appears in the query text, dense retrievers may assign high similarity to documents about that concept even when the user asks to avoid it. We introduce E-SENS, a training-free reranking method for negation-sensitive retrieval. E-SENS extracts a compact trap query for the excluded side and subtracts trap-query similarity from the original-query retrieval score. On ExcluIR, E-SENS shows a clear recall-violation trade-off across four embedding models and reduces trap retrieval at recall-preserving settings.

This work presents $\textbf{Lapis}$, a $\textbf{l}$inear-$\textbf{a}$ttention-based $\textbf{pi}$xel-$\textbf{s}$pace generative framework that achieves efficient and high-fidelity depth estimation with one-step diffusion. While generative frameworks have significantly advanced monocular depth estimation with superior detail fidelity, the $\mathcal{O}(N^2)$ complexity of standard attention and the multi-step denoising process introduce prohibitive computational costs when scaling them to high-resolution image applications. Although linear attention and one-step prediction are intuitively viable, directly applying them leads to poor structural consistency, detail loss, and noise. Lapis rectifies these limitations through a coarse-to-fine hierarchy. Specifically, a Patch-level Consistency Module restores structural coherence by integrating semantic and spatial priors. Subsequently, a Pixel-level Refinement Module recovers sharp geometric boundaries via skip-connection-based pixel correspondence. Furthermore, to mitigate sampling noise inherent in one-step diffusion, we leverage the manifold assumption and adopt a direct $\mathbf{x}$-prediction strategy to target the clean data manifold. Extensive evaluations on multiple benchmarks demonstrate that Lapis consistently achieves state-of-the-art (SOTA) accuracy and boundary sharpness across various resolutions, reducing inference latency by up to 7.6$\times$ at 1080P and 10.9$\times$ at 1440P resolution compared to previous SOTA generative models.

Current video-to-music (V2M) models lack semantic control and fail to penalize instruction violations, largely due to their reliance on reconstruction objectives and the representational bottleneck of static cross-modal conditioning in Diffusion Autoregressive (DAR) architectures. To resolve this, we introduce VIBE, a novel text-and-video-to-music (T+V2M) generation model that leverages: (1) Conditioning Connection, a depth-wise cross-layer conditioning mechanism that dynamically bridges the planning and diffusion refinement heads and (2) a comprehensive reward modeling taxonomy, optimizing for both hard, verifiable constraints (e.g., tempo, key) and soft, subjective qualities (e.g., musicality, multimodal alignment) with a structured 5-stage training curriculum. Upon evaluation using audio-visual alignment, instruction following, and audio quality metrics, along with a subjective human evaluation study, we observe that VIBE demonstrates enhanced controllability and instruction adherence while performing comparably to most evaluated baselines on generation fidelity and multimodal alignment.

Systematic generalization remains a significant challenge in deep learning. In particular, combinatorial generalization - generalizing to new configurations of known factors of variation - is effortless for humans but difficult for standard neural architectures that rely on statistical correlations rather than explicit structural representations. We introduce a new architectural component that embeds structured inductive bias into deep learning: an attention mechanism operating over tensor-product representations (TPRs). Through controlled experiments on compositional tasks, we show that this TPR-attention mechanism outperforms existing architectural components in combinatorial generalization. These results highlight the value of integrating explicit compositional structure into neural attention and point toward a promising path for models capable of systematic generalization.

Vision-language-action (VLA) driving methods increasingly combine multi-trajectory imitation learning with group-relative policy optimization (GRPO), making trajectory selection critical to final performance. However, some high-scoring trajectories that improve imitation can degrade subsequent GRPO by inducing advantage estimates misaligned with the current policy's feasible behavior distribution, driving updates away from safe and compliant behaviors. To address this, we propose a novel framework that aligns multi-trajectory supervision with policy optimization. To address the policy gradient bias induced by infeasible noisy trajectories outside the feasible region, augmented trajectories are constrained to a neighboring manifold of the ground-truth feasible region, and a Pareto-optimality criterion is adopted in place of the conventional aggregate score, retaining only non-dominated candidates and thereby filtering out conflicting samples at the source. To ensure that expanded trajectory supervision is effectively absorbed during policy optimization, we introduce two complementary mechanisms: feasibility-first advantage assignment and dynamic distillation. The former adapts Pareto credit to the feasibility composition of each rollout group and guides fully infeasible groups toward safe references. The latter updates teacher trajectories across refinement rounds to continually transfer useful supervision. Together, they progressively translate the benefits of expanded supervision into policy improvement. On NAVSIM v1 and v2, our method achieves 91.4 PDMS and 89.1 EPDMS, respectively, under single-trajectory inference, and recovers 440 of 658 initially failed scenes, 11.1% higher than the original GRPO baseline.

We develop a variational–Markov construction of detailed-balance master-equation discretizations for Fokker–Planck equations directly from their energy–dissipation laws. Rather than discretizing the differential operator, we represent mass transfer between neighboring grid points by two directional jump rates. On each edge, the discrete energy–dissipation law determines the net flux, while detailed balance fixes the rate ratio; a local reconstruction of edge quantities from neighboring grid-point values then uniquely determines both rates. A logarithmic-mean reconstruction recovers the classical Scharfetter–Gummel/Wang–Peskin–Elston rates, while alternative reconstructions yield other reversible schemes. The resulting semi-discrete systems conserve mass, preserve nonnegativity, satisfy detailed balance, and dissipate a discrete free energy. The same edgewise construction extends to state-dependent and degenerate mobilities, nonlocal interaction energies and higher-dimensional problems. The corresponding implicit and semi-implicit fully discrete schemes are linear, mass-conservative, positivity-preserving, and energy-stable, with a time-step restriction required only for nonlocal interaction energies whose kernel is not negative semidefinite. Numerical experiments confirm the expected spatial convergence rates of all detailed-balance schemes and verify their equilibrium accuracy, positivity preservation, and free-energy decay with discontinuous potentials, saturation, nonlocal interactions, and two-dimensional problems.

Brazilian Portuguese remains under-served by open language models, and the few that exist are difficult to reproduce and are often compared without measures of uncertainty. We release Manacá-1B, an open decoder-only model of 1.72 billion parameters trained from scratch for Brazilian Portuguese with a fully containerized, reproducible pipeline. The pretraining is stable, with zero skipped or NaN steps and self-recovering loss spikes, and we release its full log and dynamics. We evaluate the model against nine open baselines on four Portuguese benchmarks under a single harness. Every comparison reports a standard error and a paired significance test, and the harness is validated against previously published numbers. On last-word prediction Manacá-1B is the strongest model below the 7B scale, exceeding both Tucano-1b1 and Tucano-2b4 on LAMBADA-PT with large paired margins; it is competitive on commonsense completion and near chance on multiple-choice reasoning, as are all small base models. Along the way we document a concrete evaluation pitfall: converting a SentencePiece tokenizer with case-folding normalization to the HuggingFace fast format silently drops the normalizer, routing every capitalized token to byte-fallback and depressing scores in a way that is invisible in aggregate metrics. The uncorrected tokenizer lowered LAMBADA-PT accuracy from 45.3 to 25.0; we quantify the effect and provide a one-line fix that reproduces the training tokenizer exactly. Code, raw training and evaluation logs, per-example prediction vectors, the model weights, and the corrected tokenizer are released so that every number in this paper can be recomputed.

A supraglacial lake on the Greenland Ice Sheet ends its melt season in one of four ways: it drains rapidly through a hydrofracture, drains slowly across the surface, refreezes in place, or is buried by late-season snowfall. Which one occurs decides whether the meltwater reaches the ice bed. Satellite classifiers recover the outcome accurately but only after the season closes, and how much of a season each outcome actually requires has never been measured. We measure it directly: holding the representation and the classifier fixed, we truncate the input at 14 cutoffs from May 1 to December 31, retrain at each, and record the earliest cutoff at which each outcome's per-class F1 reaches a fixed target. The outcomes resolve in a consistent order, two of them months early: rapid drainage by July 15 and slow drainage by August 1, respectively 92 and 75 days ahead of the earliest date a full-season pipeline can be computed at all, with buried and refreeze following at 44 and 30 days. Five further learners, from a majority-class floor and 54 summary statistics to a trigger-based early classifier, leave the ordering largely intact: the three that produce a per-class trajectory reproduce it in five of six cases despite end-of-season accuracies differing by up to 18 percentage points, and it survives leave-one-basin-out evaluation, though not a move to machine-labeled lakes in an unseen season. Every feature we compute at day t reads only days up to t, at a cost of at most 1.3 percentage points. A monitoring system should therefore not have one release date: rapid drainage can be flagged on July 15, three months before a full-season pipeline can be computed at all.

Candidate generation for open-source vulnerabilities is no longer scarce. AI-assisted code review now produces defect candidates cheaply, and industry programs pair them with expert human triage. The remaining scarcity is validation and impact assessment, and the gap is largest in critical-infrastructure software like 5G cores. Here, validation has four costs: verification, reachability, impact, and fix-completeness. We present Drishti, an AI-led human-directed vulnerability audit framework with four components, one per cost: (i) an anti-pattern catalog for verification, (ii) critical-path triage for reachability, (iii) concentric validation for impact, and (iv) patch-review for fix-completeness. Across audits of Open5GS and free5GC, Drishti produced three findings. The first is a pre-authentication NULL-dereference in the Open5GS NRF multipart parser, fixed upstream with a CVE requested. The second is an ASN.1-PER memory amplification in the free5GC NGAP decoder. A 2-byte input from a rogue gNodeB OOM-kills the AMF in 6.2 seconds. The third is a defective patch on CVE-2025-69248 whose defense-in-depth check is dead code before authentication.

We study violation-feedback learning of proportionally representative approval-based committees. In each round, a learner proposes a committee of size $k$. An oracle either accepts the proposal or adversarially selects a representation violation with respect to a single fixed hidden approval profile. We compare full-witness feedback, which reveals the violation level, an omitted candidate, and the affected voter group, with candidate-only feedback, which reveals only that candidate. The target notions are proportional justified representation plus (PJR+) and extended justified representation plus (EJR+). In every setting we study, the number of rejected proposals can be bounded solely in terms of $k$, with no dependence on the numbers of voters and candidates. For PJR+, the optimal deterministic and randomized rejection complexities equal $k$ under both feedback models. For EJR+, the picture is more nuanced. Under full-witness feedback, we prove an $\Omega(k^{3/2})$ deterministic lower bound and give a deterministic polynomial-time algorithm using $O(k^2\log k)$ rejections. Under candidate-only feedback, randomization achieves $O(k^2\log k)$ expected rejections via uniform random deletion, while deterministic exhaustive branching gives a $2^{O(k^2(\log k)^2)}$ rejection bound. Even with full-witness feedback, randomized learners may require $k$ rejections.

Nowcasting headline macroeconomic indicators, i.e., estimating an indicator's value for the current reference period before its official release, is critical for monetary policy and financial markets, and central banks devote dedicated teams of expert economists to producing such estimates. Large language model (LLM) agents are a promising candidate for this task, combining broad world knowledge with real-time web search and supporting queries at higher frequency than institutional nowcasts. Evaluating their nowcasting capability is, however, challenging: headline indicators such as GDP and CPI are widely reported and likely memorized during pretraining, so any evaluation on historical releases is vulnerable to data contamination. To address this, we introduce LiveMacroEval, a live, contamination-resistant benchmark in which LLM agents produce hourly nowcasts for sixteen major U.S. macroeconomic indicators over a pre-release window closing at each official release. Nowcast quality is assessed through a LiveMacro Score against announcement-window equity returns and a LiveBetting Score from simulated Polymarket-style trading, with Federal Reserve regional-bank nowcasts, the Bloomberg ECOS professional consensus, and an auto-ARIMA baseline as comparators. Over six months with four state-of-the-art LLM agents configured with web search, aggregate nowcast accuracy is broadly comparable to the institutional and professional benchmarks, with performance varying widely across individual indicators. This highlights LLM agents' potential as real-time estimators of macroeconomic conditions.

Large Language Models (LLMs) trained on extensive scientific research are increasingly integrated as assistants for scientific discovery. However, most research papers omit the fine-grained cognitive process of examining constraints, failed alternatives, and iterative decisions required to achieve the desired goal. Such cognitive processes are vital for real-world scientists working toward specific goals under constraints. In this paper, we show that LLMs, when trained to produce such cognitive traces, perform better as scientific discovery assistants than when trained solely on scientific literature. We propose COGTRL, a trajectory-level reinforcement learning framework that trains LLMs to emulate cognitively grounded reasoning by jointly optimizing cognitive traces and the scientific steps produced in an interleaved manner. Across two 3B-parameter models and two scientific domains (AI and Materials Science), COGTRL improves method quality by an average of 7.85 points over comparable 3B model baselines and achieves competitive performance relative to 70B parameter models. Moreover, analysis by domain experts shows a preference for methods generated by COGTRL over the baselines.

Diminished Reality (DR) can reduce visual clutter by removing irrelevant objects. However, removing all task-irrelevant objects may eliminate useful contextual information and reduce situational awareness. We present FocusAdapt, a context-aware DR system that predicts object-level distraction by integrating visual saliency, semantic relevance, and gaze behavior. Based on findings from a formative study, FocusAdapt selectively diminishes highly distracting objects while preserving useful context, enabling adaptive focus assistance during procedural tasks.

Understanding which countries are represented in NLP datasets is essential for identifying gaps, targeting data collection, measuring progress, and informing AI policy. However, geographic metadata is very rarely available, and country-level representation is often hidden behind broad language-level claims. We introduce AtlasNLP, a country-aware atlas of over 13,000 NLP dataset records across normalized NLP task categories, tracking both the populations represented and where datasets are produced. AtlasNLP includes AtlasNLP-Gold, a human-curated reference set, and AtlasNLP-Core, an ACL-derived large-scale collection. Using this resource, we show that (1) dataset coverage is highly uneven across countries and tasks; (2) dataset production and representation are geographically asymmetric; and (3) language coverage does not imply geographic representation. These findings reveal blind spots in current dataset documentation practices and motivate more explicit geographic metadata for country-aware NLP evaluation.

Deep learning-based side-channel analysis has historically focused on single-byte targets and manually cropped traces, which risks discarding exploitable leakage. While recent work has proposed specialized architectures and resampling techniques to address this gap, the literature lacks a simple transformer baseline for simultaneous full-key attacks on uncropped traces. We present an open-source transformer implementation for uncropped full-key attacks which uses the standard transformer encoder backbone, adapting only the input and output layers to the side-channel setting. We release our implementation, training recipes, and pretrained weights for uncropped ASCADv1f, ASCADv1r, and CHES-CTF-2018 which achieve performance competitive with previously-reported results, while using less than 10GB of VRAM and requiring at most 3.34 hours of training on a single NVIDIA A6000.

Noise modulation (NoiseMod) encodes information in the variance of a noise-like waveform. Its noncoherent structure enables low-complexity links, but co-channel users are challenging because interference affects the same variance statistic used for detection. We study fluid antenna multiple access (FAMA) for two-level NoiseMod with independent random bits from each interferer, producing two possible nonzero interference variances. We adopt exact finite-sample maximum-likelihood energy detection from prior NoiseMod/TNM work. We derive the exact single-port moment-generating function and moments of the random-bit aggregate interference, including heterogeneous user powers, and prove that the port minimizing conditional BEP is $\arg\max_k H_k/(\sigma_w^2+J_k)$, where $H_k$ is desired-link power and $J_k$ is instantaneous bit-state-aware interference variance. For independent fading ports, the common interferer-bit vector couples per-port decision metrics; conditioning on the number of high-state interferers yields an exact binomial-mixture order-statistic representation and avoids the generally invalid rule $F_{Z_{\max}}=F_Z^{N_p}$. For spatially correlated operation, an integer-$\mu$ $\kappa$-$\mu$ benchmark uses full pairwise Jakes correlation in scattered fields, with channel-averaged BEP evaluated by conditional Monte Carlo. A standard-error-aware load study shows increased admissible co-channel users as spatial degrees of freedom grow. Finally, a finite-sample variance-domain sensing diagnostic quantifies the oracle gap under frozen channels and interferer states, providing a lower-bound warning on fast-FAMA sensing difficulty rather than a deployable acquisition protocol.

Bilevel mixed-integer linear optimization problems model hierarchical decision processes in which a leader anticipates the optimal response of a follower. Although expressive, these problems are computationally challenging because lower-level optimality is embedded in the leader's feasible region. Value-function reformulations replace the nested follower optimization with a constraint involving the follower's optimal value, but evaluating this value function exactly can itself be expensive. This paper introduces Graph4BiLO, a graph neural network (GNN) approach for learning bilevel value functions from variable–constraint graph representations. In contrast to fixed-length multilayer perceptron (MLP) representations, the GNN uses shared message-passing parameters and can therefore be applied across multiple problem sizes with a single trained model. The learned ReLU network is encoded exactly as mixed-integer linear constraints and embedded in an approximate single-level formulation. A repair step subsequently re-solves the follower problem for the selected leader decision to recover a bilevel-feasible follower response. We evaluate Graph4BiLO on knapsack interdiction instances with 20–100 items against the exact MibS solver and the learning-based Neur2BiLO method. Graph4BiLO obtains objective values comparable to Neur2BiLO across all tested sizes while avoiding size-specific neural networks. An additional out-of-distribution experiment demonstrates zero-shot transfer from 20-item training instances to previously unseen 40- and 60-item instances. However, embedding message passing at every graph node substantially increases the resulting mixed-integer formulation size and solve time. These results identify a central tradeoff between size-generalizable graph representations and the computational cost of embedding GNNs within optimization models.