Papers

Large language model (LLM) planners can decompose natural-language instructions and select reusable robot skills, but choosing the correct skill does not guarantee successful physical execution. This gap is especially important in humanoid loco-manipulation, where errors during approach, grasping, transport, or placement can invalidate the remainder of a long-horizon plan. We present FRAMES, a failure-aware supervisory framework for the Unitree G1 humanoid that operates above the CEER whole-body controller. A Planner Agent selects subtasks through parameterized mid-level skills, while a vision-language-model-based Monitor Agent evaluates each skill using temporal multi-view observations and structured robot and contact evidence. Detected failures stop the active skill and provide grounded feedback to a Recovery Agent. The framework further includes a Memory Module for reusing prior skill experience, and geometric grounding via depth and segmentation. We independently evaluate the monitoring module of the framework in MuJoCo using 100 trials comprising 50 failed and 50 successful executions across five tasks. The monitor detects 48 of 50 failures, correctly accepts 46 of 50 successful executions, and achieves 94.0% overall accuracy. These results provide initial evidence for the monitoring component, while end-to-end evaluation of the complete recovery loop remains ongoing.

Enterprise AI assistants must produce responses that are verifiable and traceable to source evidence. However, retrieval augmented generation (RAG) over heterogeneous enterprise data can suffer from citation drift, unsupported content, and weak source traceability. We present EvidenT (T = Trust + Transparency + Traceability), a lightweight pipeline that verifies extracted evidence against retrieved documents before answer generation, without model retraining. EvidenT combines structured passage extraction with deterministic lexical alignment to filter unsupported content, correct citation drift, and preserve source-span traceability. On approximately 500 real enterprise queries, EvidenT improves gold-source hit rate by an average of 29% over prompting baselines, produces no citations to nonretrieved urls, and achieves near-saturated answer-to-source lexical coverage.

Bangla, the seventh most spoken language in the world, exhibits significant regional dialectal diversity, with dialects such as Barishal, Chattogram, Sylhet, Noakhali, and Mymensingh differing in lexical, morphological, and syntactic characteristics. These variations pose substantial challenges for Named Entity Recognition (NER), limiting the generalization of models trained on Standard Bangla or a single regional dialect. This paper presents a cross-dialect Bangla NER framework using the publicly available ANCHOLIK-NER dataset, comprising 17,405 annotated sentences and 101,817 tokens across five major Bangla regional dialects. A Leave-One-Dialect-Out Cross-Validation (LODOCV) strategy is adopted, training models on four dialects and evaluating on the remaining unseen dialect. Eight pretrained transformer-based models, including BanglaBERT, MuRIL, XLM-RoBERTa, and Multilingual-E5, are evaluated under identical experimental settings. Multilingual-E5 Large achieves the highest F1-score in every fold, peaking at 97.26% on Mymensingh and reaching its lowest, 82.38%, on Chattogram, the most challenging target dialect. To improve interpretability, Local Interpretable Model-agnostic Explanations (LIME) are applied to word-level predictions, revealing that the model's decisions are driven primarily by the surface form of the target entity word itself rather than by surrounding sentence context. These findings establish a benchmark for cross-dialect Bangla NER and demonstrate the effectiveness of transformer-based transfer learning for low-resource regional dialects, while highlighting the surface-form dependence of current models as a direction for future work.

Strategic classification studies how a decision maker should choose a classifier when the agents being classified can adjust their features in response to it. In existing models, the classifier is the only instrument available to the decision maker, and therefore a feature that is predictive but easy to fake can only be down-weighted or discarded. However, in many settings the decision maker can also verify: lenders verify income, admissions offices check documents, and tax authorities audit returns. In this paper, we propose a model of strategic classification in which the firm jointly designs a linear classifier and an audit profile, which assigns to each fakeable feature a probability of detection and a penalty when caught. We show that under linear costs, the classifier affects the audit problem only through the distribution of gaming rents it induces, so that the joint design problem decomposes into the choice of a score rule and an audit allocation problem. We use this decomposition to characterize the optimal audit allocation, to identify when the allocation problem is tractable and when it is NP-hard (namely, when agents can game through overlapping features under an inspection cap), and to bound the regret of a firm that has to learn the rents by auditing. We further show that audit intensity is a quantity to be tuned rather than maximized: welfare is single-peaked in it, and a firm and a social planner disagree on the mix of detection and penalty that delivers a given level of deterrence. Notably, two populations with identical costs and causal structure can game in one domain and improve in the other, a difference that a cost-only model cannot account for. Together, our findings highlight that whether a feature is “gameable” depends on the institution's verification policy as much as on the feature itself.

Dashboards, feeds, and leaderboards determine whom agents believe they are playing against. We study how such information architectures shape strategic behavior, and how a designer who controls them can use visibility as a lever. We propose a behavioral model in which a transparency parameter governs how selectively agents are exposed to opponents of nearby reasoning depth, and embed it in a Cognitive Hierarchy environment, which provides a tractable language for tying agents' beliefs about the sophistication of others to what they see. We show that because the exposure operator is an exponential tilt of the type distribution, the effect of transparency on aggregate behavior has a closed form: the derivative of the aggregate action with respect to $p$ is proportional to the population-average variance of perceived sophistication. This identity signs the effect in games of strategic complements (opacity raises aggregate effort) and in coordination environments (opacity lowers the mean action), and an ordinal version holds for any response rule that is monotone in the belief. Notably, under a Poisson prior, opacity acts as a sophistication multiplier: every reasoner plays as if the population were $\mathrm{Poisson}(\tau/p)$. We use this structure to characterize the disagreement-versus-level tradeoff faced by a coordination-minded designer, to identify the reasoning depths that respond most to visibility, and to show that opacity imposes a “sophistication tax” under which the deepest reasoners can lose to moderate ones. Simulations in which agents see only a noisy leaderboard of published actions reproduce both the reversal and the tax. Together, our findings highlight that who an agent sees, and not only how deeply it reasons, shapes strategic behavior, and point to visibility as a design lever.

International law provides the normative framework through which states coordinate action, regulate armed conflict, and protect human rights, yet its texts remain without token-level named entity recognition (NER) resources. We introduce IntLawNER, a NER dataset and benchmark for codified sources of international law, covering 2,987 gold-annotated sentences and 8,094 entity spans from International Court of Justice (ICJ) decisions, UN Security Council resolutions, and European Court of Human Rights (ECtHR) judgments, annotated with seven institution-specific entity types. We construct IntLawNER with a cost-effective hybrid algorithmic-agentic pipeline that reduces 468k source sentences to a compact annotation set through candidate retrieval, LLM-based vetting, and human review, with 89.6% of gold spans accepted unchanged from the silver layer. However, the silver-to-gold analysis reveals that human-machine aggregate agreement metrics can be misleading in domain-specific NER: Cohen's kappa=0.964 on boundary-matched spans masks a macro-F1 of 0.753 when missing entities, boundary errors, and label corrections are included. The benchmark shows that zero-shot span-based GLiNER collapses on entity types dependent on institutional function rather than surface form (0.243 micro-F1), while fine-tuned transformers struggle on rare labels. Carefully selected few-shot examples that demonstrate label contrasts improve every LLM over zero-shot prompting, with Claude Opus 4.6 reaching the best score of 0.873 micro-F1. We release IntLawNER as a benchmark and reusable resource for extracting references in international legal texts.

The Precision Time Protocol (PTP) provides the time and phase synchronization required by disaggregated Open Radio Access Networks (O-RAN). Yet, in current open fronthaul deployments, PTP traffic lacks mandatory authentication and integrity protection, leaving synchronization vulnerable to spoofing, replay, and delay manipulation attacks that can degrade radio access performance. Existing protections are poorly suited to this setting: they either add excessive latency, do not support multicast dissemination efficiently, or fail to contain key exposure under partially trusted RUs. This paper analyzes the security risks of unprotected O-RAN PTP and develops a threat model for open fronthaul deployments. We then introduce PRTESLA-C, a lightweight synchronization protection mechanism that combines per-round delayed key disclosure with ASCON-based message authentication. PRTESLA-C uses an apply-then-verify-and-correct paradigm: timing samples are applied immediately to preserve real-time control, verified after key disclosure, and removed from persistent synchronization state if authentication fails. This design maintains sub-microsecond synchronization accuracy, provides strong protection against spoofing and replay, and bounds the impact of delay manipulation with minimal computational and latency overhead.

Models of opinion dynamics on networks, provide a framework to study how a network of agents aggregates dispersed opinions into a consensus. However, existing models assume that agents truthfully report their opinions, and do not account for environments in which a platform aggregates agents' reports across multiple networks, and agents can strategically report different opinions on each. In this paper, we propose a model of strategic opinion manipulation on multiplex networks, in which a platform merges L network layers with attention weights, and each agent, holding a private opinion, chooses (potentially different) reports on each layer, at a layer-specific misreporting cost, so as to pull the merged consensus toward their own opinion. We show that this game has a unique Nash equilibrium in closed form, that the resulting consensus is the truthful consensus under a centrality tilted toward a manipulability index of each agent, and that the resulting distortion is the (centrality-weighted) covariance of agents' manipulability and opinions. We further show that agents' reports on every layer are more extreme than their opinions, so that the polarization observed by the platform overestimates the true polarization. Notably, we highlight that merging layers is a double-edged sword: as manipulability depends on the square of an agent's influence on each layer, spreading attention across layers dilutes manipulation, while heterogeneous misreporting costs across layers and shifts in agents' centralities can amplify it. Finally, we characterize the platform's optimal choice of attention weights, in closed form when the layers share a stationary distribution, and through an exact gradient and a marginal test otherwise. Together, our findings shed light on when aggregating opinions across networks is (not) robust to strategic manipulation, and point out potential interventions to alleviate it.

Cosine similarity is widely used to analyze transformer representations, implicitly assuming that similarity reflects task-relevant structure. We study when this assumption fails in dialogue-conditioned large language models. Across three 7-8B chat-tuned models, ambient cosine similarity substantially underestimates linearly decodable persona structure on the same hidden states; numerically, linear probe AUC is in the 0.73-0.97 range while cosine kNN is in the 0.56-0.77 range on a 30-class task. A low-dimensional supervised subspace recovers much of this gap, whereas a matched-rank PCA subspace does not and in some cases degrades performance. This mismatch is regime-dependent: it is absent in single-sentence sentiment classification (SST-5), and a matched-cardinality control rules out attribute cardinality as a confound. The gap does not systematically increase across dialogue turns, and the task-aligned subspace remains stable over time. However, two of three models violate a pre-registered within-subspace separability invariance criterion (|Delta AUC| <= 0.03), and one model violates a pre-registered turn-invariance criterion (|Delta L| <= 0.05). These results show that cosine similarity can fail to reflect task-aligned structure in dialogue representations even when that structure is linearly accessible.

The performance of learned robot visuomotor policies depends heavily on the size and quality of their training data, yet collecting high-quality demonstrations remains costly for robots in the real world. Although large-scale robot and human datasets are increasingly available, embodiment gaps and mismatched action spaces make them difficult to leverage directly. Cross-embodiment transfer, reusing experience from other embodiments to improve learning on a target embodiment, is therefore crucial for scaling robot learning beyond per-robot data collection. In this work, we find that efficient transfer can be achieved by learning from what is shared across embodiments, the visual dynamics of how the world responds to motion, and by effectively exploiting the scarce target-embodiment data at test time. The proposed framework, called Latent Policy Steering (LPS), implements an embodiment-agnostic pretraining phase, which trains an image-based World Model (WM) with optical flow across diverse embodiments. The resulting WM is finetuned on the target embodiment with robot actions. It then steers the base policy toward better actions by searching in the WM's latent space for plans that stay close to the finetuning data. LPS is a policy-agnostic framework: it can flexibly accommodate different policies without having to retrain them. In Robomimic and real-world evaluations, LPS improves the average performance of Diffusion Policy relatively by 16% and 62%, and Pi0.5 by 8% and 14%, with only 50 demonstrations on an unseen target embodiment.

Recent advances in generative artificial intelligence have enabled the synthesis of complex human motion with unprecedented fidelity. However, current text-to-motion systems rely strictly on linguistic semantics: if an input reads "I put my hands up", the model searches for a pose with raised hands, and every non-semantic property of the text is discarded as noise. In this work, we treat that discarded structure as the signal. We present an embodied visualization instrument that amplifies not what a text means, but how it is built. Our method first quantizes dance kinematics into a motion codebook of 256 stylistic "regions" using Principal Component Analysis and K-Means clustering, and orders those regions along the dominant axis of movement. We then map the raw byte representation of any input text directly onto this codebook, producing a deterministic sequence of regions that we call the text's "choreographic genome". A precomputed plausibility graph and a set of physics smoothing routines turn this genome into fluid, full-body movement, so that the dancing body becomes a display surface for the byte-level structure that semantic systems ignore. Through a series of artistic case studies, including a Shakespeare sonnet, a machine error log, source code, an abolitionist's question, and Indigenous and Devanagari scripts, we show that each text produces a visibly distinct dance, and that scripts marginalized by ASCII-centric computing are amplified into close to three times as much movement per character. We frame this not as a motion-synthesis benchmark, but as a critical and poetic visualization that asks what we choose to count as signal, and what we allow to go unheard.

Poor clinical communication can delay care, contribute to errors, and harm patients, yet opportunities for repeated practice with feedback remain limited. Our prior randomized trial showed that practice with the SOPHIE AI patient platform improved serious illness communication, but the system addressed a single clinical context and required human effort for delivery and assessment. We developed SOPHIE 2.0, a browser-based, self-service platform integrating embodied AI-patient interactions, personalized feedback, and automated assessment across 24 clinical scenarios. An automated large language model assessor evaluated three communication skills—Empower, Be Explicit, and Empathize—with agreement comparable to individual human raters ($r=0.759$; ICC$=0.746$). In a study of 59 clinicians and students, participants completed two AI-patient encounters with personalized feedback; 92% found the platform engaging, 86% easy to use, and 83% clinically relevant. Scores were higher in the second encounter, though the uncontrolled design precludes attributing this change specifically to training.

Consensus among LLM judges is often taken as strong evidence that a decision is correct. This assumes that judges make their errors independently. In practice, LLM judges are often trained and evaluated in similar ways, so they can make the same mistakes. We study how this dependency affects the reliability of consensus. We find substantial error correlation across both open-weight and frontier LLM judges. In our main bank of ten judges, the average pairwise correlation between judge errors is 0.21. As a result, the ten judges only provide roughly as much statistical information as 3.5 independent judges. The dependency is even stronger among the high-accuracy frontier judges we evaluate, including judges from different providers. In up to 28% of our comparisons, ignoring shared errors leads to the conclusion that one system is significantly better, while accounting for them does not. We also find that the pattern of errors matters. Errors shared by most judges and errors concentrated among a smaller group affect consensus differently and favor different voting methods. Measuring the overall amount of correlation alone is therefore insufficient. Our results suggest a simple approach: use a small set of trusted examples to estimate judge accuracy and identify shared mistakes. These shared errors should then be considered when analyzing the results, and the voting method should be chosen using trusted examples before it is applied to new data.

Identity cues are irrelevant to a fixed programming specification, but raw counterfactual differences can arise from unequal samples and prompt wording. We audit 30.73 million executed Python generations from seven checkpoints on HumanEval+ and MBPP+, supplemented by an exploratory 550B slice, under model-assigned gender, country, and occupation personas. Within-task randomization and false-discovery-rate control identify occupation as the most consistent structural axis: CodeBLEU dispersion exceeds its exchangeability null in 10/14 model–benchmark cells, remains significant in 8/12 full-coverage cells, and exceeds the country ratio in every paired cell, although the median excess is only 0.141 points. In six high-pass-rate cells, occupation dispersion replicates across token similarity, length, comments, reference similarity, and complexity, while pass-rate dispersion is significant in none. Country leads raw dispersion in 12/14 cells but has a median calibrated ratio of 1.00. Gender-associated variation cannot be separated from persona wording in this design. Thus, the evidence supports small, reproducible occupation-conditioned changes in code form, not stable disadvantage to named identities or demonstrated downstream harm. We use this finding to motivate a broader commentary on the potential impacts of bias in code generation, including the possibility that models may condition their outputs on identity information available from prior conversational context.

As LLM applications integrate with external tools, they are increasingly exposed to indirect prompt injection (IPI), where adversarial instructions are embedded in retrieved content. Conventional IPIs fire on contact: the moment an agent ingests the content, it carries out the instruction. We introduce the explosive prompt, a conditional payload that stays dormant until an attacker-chosen trigger is met, in effect a training-free, inference-time backdoor planted in a single piece of retrieved content. This temporal separation reaches where ordinary IPI cannot. On frontier models that refuse the bare imperative almost entirely, rephrasing the same goal as a dormant conditional drives real, state-changing tool execution against a live agent backend (a paired mean of 16.5% vs. 2.4% for the imperative, reaching 34.2% on a proprietary model). In trials on nine production agents (OpenAI Codex, Google Gemini CLI, Anthropic Claude Code CLI, Cursor CLI, GitHub Copilot, Devin AI CLI, Amazon Kiro CLI, Qwen Code, Google Assistant; n=30 each), explosive prompts succeed in 43-83% of cases versus at most 3% for an imperative baseline, and they slip past deployed defenses: off-the-shelf injection classifiers are miscalibrated on them, and a preference-optimized model that closes imperative injection entirely still executes 11.8% of explosive prompts, every one at the trigger turn. The durable defensive lever is ingestion-time detection of the conditional structure, once detectors are trained on explosive-prompt data, which no prior benchmark supplied and our generator does. Retraining cuts live tool-execution attack success from an undefended 34.3% to 7.5-8.1% for the encoder baselines. Our detector, DeFuse, reaches 3.0% at a calibrated 5% false-positive budget with the best detection quality of any method tested (AUC 0.9994) and 25x lower latency, though it needs length-aware thresholds.

Generative Adversarial Networks (GANs) have demonstrated strong performance in generating high-quality synthetic data. However, they are limited by no formal guarantees regarding convergence and the effectiveness of the learning process. In practice, this leads to training instability, mode collapse, and sensitivity to hyperparameters. To address this, we propose EmbeddGAN, a novel adversarial training framework based on a dependence-based objective. Instead of relying on a discriminator that classifies samples as real or fake, EmbeddGAN introduces an embedding network that learns a representation in which statistical dependence between samples and their real/fake labels is maximized, while the generator is trained to minimize this dependence. This objective is implemented using the Gini distance correlation (gCor), which equals zero if and only if the embeddings are statistically independent of the real/fake label. Minimizing this objective therefore encourages real and generated samples to become statistically indistinguishable in the learned embedding space. The embedding network projects both real and generated data into a shared low-dimensional space, where distributional discrepancies can be measured directly through pairwise distances. We adopt a minimax training strategy: the embedding network maximizes the Gini distance correlation (maximizing dependence), while the generator minimizes it (minimizing dependence). Experiments on the MNIST, CIFAR-10, and CelebA datasets demonstrate that EmbeddGAN achieves competitive performance relative to established baselines while exhibiting notably stable training dynamics on the evaluated datasets.

Vision Transformers devote most of their parameters to MLPs for channel mixing, but still rely on quadratic multi-head self-attention for token interactions. While linear attention fixes the complexity problem, bringing it down to O(N), it is usually just paired with the same fixed-activation MLP as before. Kolmogorov-Arnold Networks take a different approach, placing learnable univariate functions on the edges instead. However, existing vision KANs either retain standard attention or remove attention entirely, so the two ideas have not been effectively combined. We introduce LKAT (Linear Kolmogorov-Arnold Transformer) to close this gap: an isotropic ViT-style encoder that couples chunk-wise Gated Linear Attention with a two-layer KAN feed-forward block, backed by an I/O-aware fused RBF-KAN kernel to make radial-basis grid functions efficient in practice. Under a shared DeiT-style training recipe, LKAT-B outperforms ViT-B/16, ViT-5-B, and Mixer-B/16 on ImageNet-100, while Tiny, Small, and Base variants scale consistently on CIFAR-10/100. ImageNet-100 pretraining also transfers effectively to CIFAR fine-tuning, suggesting that gated linear attention and KAN-based radial basis functions provide complementary inductive biases for mid-scale visual representation learning. Code: https://github.com/mehizelali/linear-kan-transformer

Locally recoverable codes (LRCs) combine global error correction with the ability to repair a small number of erased coordinates by accessing only a limited number of other coordinates. Motivated by their quantum counterparts, we construct several families of optimal cyclic $(r,\delta)$-LRCs that are Euclidean or Hermitian dual-containing. In the Euclidean case, we obtain four families of optimal dual-containing cyclic $(r,\delta)$-LRCs over $\mathbb{F}_q$ with a range of minimum distances extending beyond the local distance $\delta$. In the Hermitian case, we derive analogous families over $\mathbb{F}_{q^2}$ that are Hermitian dual-containing, when $(r+\delta-1)\mid(q^2-1)$. In addition, we develop a distinct construction for the case $(r+\delta-1)\mid(q^2+1)$ using symmetric defining sets and odd $\delta$, which yields optimal codes with minimum distances $\ell+2$, $\delta+2$, $2\delta-2$, and $2\delta$. The Euclidean Calderbank-Shor-Steane (CSS) and Hermitian stabilizer constructions then give corresponding quantum cyclic $(r,\delta)$-LRCs over~$\mathbb{F}_q$. Further, the resulting stabilizer codes are pure; they meet the relevant quantum Singleton-type bound, and are therefore optimal. We also provide a comparison with previous works, highlighting the parameter regimes and minimum distance ranges covered by our results that are not attained by existing constructions. Explicit examples illustrate the constructions and verify the dual-containment and purity conditions.

Autonomous navigation requires precise and efficient semantic segmentation, yet existing frame-based approaches remain limited by motion blur, glare, latency, and the low temporal resolution (20-30 FPS) of conventional cameras, which leads to information loss between frames. Event cameras have emerged as an alternative sensing modality, capturing intensity changes asynchronously with high temporal resolution, high dynamic range, and sparse outputs. However, event-based algorithms still fall short of frame-based ones in accuracy, as most segmentation methods are designed for dense frame data. To overcome these limitations, we propose a hybrid vision architecture that combines conventional frame-based and event-based cameras. The system integrates two complementary components: (1) a compact Spiking Neural Network (SNN) with 42k parameters for motion estimation, and (2) a lightweight event-driven SNN with 0.84M parameters for frame-based semantic segmentation, which interpolates motion between frames to refine segmentation results. By predicting inter-frame segmentations, the framework achieves segmentation rates of up to 500 Hz with an energy consumption below 1.87 mJ per inference, while maintaining real-time GPU execution at frequencies up to 200 Hz. Additionally, our approach compensates for information loss in frames affected by blur or overexposure, enabling more robust perception in challenging conditions.

The aggregation of many lay estimates often outperforms individual expert judgment, a phenomenon known as the wisdom of crowds. While this is usually attributed to the independence of estimates, an even stronger effect arises through deliberation: averaging the consensus estimates of small deliberating groups outperforms the classical wisdom of crowds, with individual judgments themselves also becoming more accurate after deliberation. Whether these improvements transfer to large language models deliberating amongst themselves is unknown. Here we adapt a three-stage deliberation paradigm previously used with human participants for use with large language models from three different families, and test it across four domains of increasing real-world stakes: visual numerical estimation (Study 1), peer review of machine-learning papers (Study 2), detection of hidden malicious behavior by an artificial intelligence agent (Study 3), and sports forecasting against a real prediction market (Study 4). Across domains, deliberation reduced collective error beyond passive aggregation of independent responses, and post-deliberation individual judgments retained this collective gain. Notably, the advantage required model diversity: groups composed of clones of a single model did not benefit from deliberating. These results establish machine deliberation as a general-purpose aggregation mechanism, and point to diversity as an active ingredient.

In recent years, there has been a surge of interest in Cultural NLP, with substantial efforts to create globally inclusive NLP systems. The rapid growth of literature in this field makes it difficult to track trends in methods and data resources. To address this, we analyze over 375 papers to answer three complementary questions: (1) What Cultural Capabilities (CCs) are being targeted in NLP systems? (2) How are cultural data resources being created? and (3) What methods are being used to improve the CCs of those systems? We discuss trends observed across the three questions, and identify relevant research gaps. To facilitate further research in this field, we release our full list of analyzed papers in the form of an interactive web interface, which includes a feature to allow researchers to add their work; we hope this facilitates future research and proves to be a valuable resource for the Cultural NLP community.

This paper presents a stacked two-layer force-sensing resistor (FSR) array designed for robotic fingertips that combines high-resolution pressure mapping with shear-force estimation. A compliant lattice elastomer spacer converts shear loading into a measurable inter-layer displacement, producing relative center-of-pressure (CoP) shifts between layers. A physics-based moment balance links inter-layer CoP displacement to shear force, while an end-to-end CNN–GRU model captures nonlinear effects from load-dependent compression and contact redistribution. This model, with both layers as input, achieves coefficients of determination $R^2 = 0.914$ for $F_x$ and $R^2 = 0.944$ for $F_y$, consistently outperforming single-layer baselines for shear-force estimation. Robotic manipulation experiments show that, for contact-motion tracking, the deep layer tracks the translation and rotation imposed by the robot arm, whereas the superficial layer tracks the slip at the contact surface. Transient changes in the difference between the total pressure responses of the two layers provide the best slip-event detection performance among the tested cues. These results demonstrate that two-layer FSR arrays can provide three-axis force estimation, contact-motion tracking, and slip-event detection beyond conventional normal-force sensing.

We present a candidate counterexample to a conjecture stating that the existence of a Lyapunov-Krasovskii functional with a pointwise dissipation rate is sufficient for the input-to-state stability of time-delay systems. The counterexample has been derived through interactions with large language models.

Matrix-valued optimizer states may contain relational structure that is not captured by treating their entries independently. We study whether relations within matrix-valued optimizer states can be exploited to improve optimization. To this end, we introduce a unit-relation-transform abstraction and instantiate it as COREM, a Cosine-Relation Momentum Reshaping method with stateful writeback. COREM partitions the momentum state into update units, computes cosine relations among them, and uses these relations to reshape the momentum before writing the transformed state back to the optimizer. This stateful mechanism allows the reshaped momentum to affect not only the current update but also future optimization dynamics. We evaluate COREM on CIFAR-10 with an MLP and on enwik8 with a Transformer. Compared with Muon, COREM shows lower early-stage step efficiency but stronger improvement in the mid-to-late stages of training, achieving better final validation performance on CIFAR-10 and comparable final performance on enwik8. Spectral diagnostics on enwik8 show that COREM consistently increases entropy effective rank and reduces the concentration of singular energy in dominant modes, while preserving an anisotropic spectrum. For square matrix updates, COREM requires approximately 13.3% of the transformation FLOPs of Muon with five Newton-Schulz iterations.

Large language models (LLMs) increasingly ground their answers in proprietary data and live web content through both single-shot retrieval-augmented generation (RAG) and multi-turn agentic search. Yet the web path they use was built for human readers: search services return URLs and snippets ranked by generic relevance, and content delivery networks (CDNs) cache URL-addressed objects without knowing which passage an agent needs. LLMs instead consume short, semantically coherent passages - hereafter, chunks - selected for downstream utility rather than similarity alone, retrieve statefully across reasoning turns, and, as uncoordinated streams, repeatedly fetch, clean, chunk, and embed the same content. We argue that semantic chunk retrieval should become a first-class network-delivery abstraction, and propose the Semantics Delivery Network (SemDN): an origin-authorized, hierarchical substrate that indexes, searches, and caches web content at chunk granularity, serving clients and origins alike. SemDN is more than finer-grained caching: it must estimate missing or stale evidence without an explicit miss signal, let clients choose retrieval representations that origins need not see, and share state among clients with the same representation. Preliminary results show that page-first retrieval processes two to three orders of magnitude more content than agents consume, that agentic queries exhibit task-local chunk reuse invisible to URL caches, and that chunk delivery improves answer quality per context token. We organize an open research agenda around shareable retrieval state, hierarchical caching and refresh under coverage risk, and deployability, including incentives, privacy, and verifiability. Our code is available at: https://github.com/sarendis56/SemDN_HotNets26

We investigate whether adding $t$, the positive magnitude of the squared nuclear four-momentum transfer, enables boosted decision trees (BDTs) to improve invisible-dark-boson selection relative to optimized rectangular cuts at the Electron-Ion Collider. We model coherent exclusive scalar and vector production at generator level in electron-gold collisions at 18 GeV by 100 GeV per nucleon. Both methods use identical weighted samples, inputs, preselection, and optimization objectives. Using only electron information, the BDT provided no consistent advantage over optimized cuts for signal selection across 11 masses for each boson type. When both methods also use $t$, the BDT distinguishes signal from background slightly better than optimized cuts at 10 GeV for both boson types. These results motivate further investigation of machine learning in EIC dark-boson searches through exclusive processes where $t$ can be reconstructed.

Robotic ultrasound (US) calibration is essential for accurately relating US images to the robot coordinate system, but accurate and automated calibration remains challenging because existing methods often require complex phantoms or external 3D trackers. In this work, we develop a tracker-free robotic US calibration framework using a spherical-marker phantom and a highly automated perception pipeline for sphere-center localization. The proposed threshold-free image-processing method localizes the spherical feature based on each US image's intensity distribution, eliminating hand-tuned intensity thresholds and improving robustness across imaging settings without per-system retuning. Multi-pose observations of the spherical fiducial are then used to estimate the US-to-EE transformation without external tracking or prior localization of the sphere center in the robot base frame. We validate the proposed framework on a robotic US platform through repeated sphere scans and further assess the calibrated system using geometrically distinct phantoms with known CAD models. Across 12 cross-validation folds, single-marker calibration achieved sphere-center accuracy and precision of $1.75\pm0.51$ mm and $0.85\pm0.11$ mm, respectively, compared with $1.72\pm0.51$ mm and $0.84\pm0.12$ mm for three-marker calibration. Across the three reconstruction sets, the single-marker calibration yielded pooled post-registration point-to-surface MAE$\pm$SD values of $0.41\pm0.35$ mm for the cone and $0.50\pm0.41$ mm for the triangular prism, closely matching the three-marker results of $0.40\pm0.35$ mm and $0.48\pm0.40$ mm, respectively.

We address the problem of recovering high-speed videos from dynamic scenes under extreme photon sparsity. Existing methods rely on aggregating photon detections in local spatiotemporal windows to improve signal-to-noise ratio; however, this local grouping discards global structure and fails in low-light regimes where photon detections are sparse in space and time. In this work, we show that the information needed to recover both motion and illumination is encoded in correlations over the full space-time pattern of photon arrivals. Building on this insight, we develop a spatiotemporal flux probing theory and an algorithm that estimates the Fourier coefficients of the underlying intensity directly from the photon stream. We demonstrate that our approach (1) recovers fast motion and temporal illumination dynamics with substantially fewer photons than prior methods, (2) enables velocity-selective videography that automatically refocuses video onto specific detected motions, and (3) generalizes across sensing modalities including single-photon, event, and spike cameras.

Behavioural evaluations of hosted language models can vary because the evaluated service, the measurement instrument, or both differ across runs. We separate three validation questions: whether a prior finding recurs on fresh data under its historical configuration (replication), whether the endpoint changes when the evaluation-and-inference configuration is rebuilt under the same identifier (measurement sensitivity), and whether the finding persists across subsequently tested identifiers under one common instrument (persistence). We study these questions in Regent Chess, a sequential environment in which a hidden, mutable state is recorded exactly, allowing stated beliefs to be scored against ground truth at action time; positive endpoint values mean worse performance than a matched-uniform comparator. The previously reported Gemini 3.1 Flash-Lite deficit recurs on fresh games under its historical configuration (+0.0530, 95% CI [+0.0329,+0.0714]). In a back-to-back same-day H/R comparison under the same public identifier, the model-minus-uniform endpoint is 0.0429 lower under the rebuilt configuration (95% CI for the H-minus-R contrast [+0.0182,+0.0667]); all six configuration components vary jointly, so no component is isolated. Under rebuilt R, the prospectively frozen, interleaved same-window 4K comparison reverses sign between Gemini 3.1 and Gemini 3.7, identifiers that differ in release and product tier; additional descriptive and exploratory cells show the same directional pattern. Any additional serving-period contribution remains unresolved (-0.0166, [-0.0483,+0.0157]). Replication, measurement sensitivity, and persistence can therefore yield different conclusions within one evaluation, motivating explicit indexing of hosted-model behavioural claims by tested identifier, serving period, measurement instrument, and inference configuration.

Eye-catching graphics, such as circular figure labels and word-scale visualizations, are increasingly being placed directly within long-form text paragraphs. Some research has claimed that inline visualizations can help readers understand data-rich passages more clearly. However, research in the science of reading calls into question the introduction of images within the flow of text. In this work, we conduct an exploratory study of eye movement in both the presence and absence of visual embellishments. Using a high-resolution eye-tracker (EyeLink Portable Duo), we observed small mean increases in vertical and cross-column saccade rates among six participants, with substantial variation among readers and no detected difference in comprehension accuracy. Mean subjective ratings for the data-light circular-glyph passage indicated it is more distracting than its unembellished comparison passage. These exploratory observations differ from prior findings and motivate a larger, fully crossed study of inline graphics.