Papers

Dowker homology is a topological tool that may be used to analyze the relative position of two point clouds living in a common space. We investigate whether Dowker homology captures sentence similarity information by treating the embeddings of the tokens that constitute a sentence pair as a pair of point clouds in the latent space of a transformer model, using both models that have and have not been fine-tuned for sentence similarity. We find that Dowker homology captures sentence similarity information, as measured by regressing Dowker homology features onto ground-truth similarity scores, and that it can be used for visual inspection of similarity data and models. In an attempt to make Dowker homology readily applicable, we derive from it single-number summaries that we expect to capture sentence similarity directly. These turn out to work reasonably well, but without outperforming standard sentence similarity measures based on established pooling methods.

Spoken Language Models (SLMs) generate textual responses directly from speech, offering an alternative to cascaded systems. Despite recent advances, existing SLMs still exhibit weaker instruction-following behavior and limited generalization across diverse tasks compared to text-based language models. Our analysis shows that speech and text representations in current SLMs remain weakly aligned despite strong downstream performance, indicating that structural differences between continuous, temporally varying speech and discrete text remain insufficiently addressed. To address this, we propose a simple framework that decouples length mismatch from semantic alignment and encourages closer correspondence between speech and text representations. Experiments across multiple benchmarks demonstrate competitive performance against strong baselines, underscoring the importance of explicitly addressing structural differences between speech and text in SLM training. Our code is publicly available at https://github.com/jaykim9870/Do_SLMs_Hear_Speech_as_They_Read_Text.

Recent monocular 3D Gaussian Splatting (3DGS) streaming reconstruction methods have achieved impressive performance by balancing reconstruction quality and efficiency. However, extending these frameworks to underwater scenes remains challenging due to severe visual degradation, such as light attenuation and scattering, which degrades camera pose tracking and distorts scene geometry. To address these challenges, we propose AquaFlow, a monocular Gaussian Splatting streaming reconstruction framework for efficient and high-fidelity underwater reconstruction. Specifically, AquaFlow fine-tunes a 3D vision foundation model on large-scale underwater data for robust pose and pointmap estimation, and introduces a medium-guided incremental Gaussian initialization strategy for streaming mapping. Furthermore, we develop a streaming-compatible hybrid scene representation that integrates structured, distance-conditioned neural Gaussians with a physics-inspired optical model to compensate for underwater image formation effects, enabling accurate scene reconstruction. We evaluate AquaFlow on a comprehensive dataset of 62 diverse underwater trajectories, collected from both public benchmarks and in-the-wild web videos across various scales. Extensive experiments demonstrate that AquaFlow achieves state-of-the-art tracking and rendering performance, reducing average localization error by 13.2% and improving PSNR by 4.74 dB compared to WaterSplat-SLAM.

This paper presents computational results of cryptanalysis of AES using the Local Inversion by Black Box computations of the forward encryption and utilizes these results to develop a practically feasible approach for the key recovery of the full scale AES128 under Known Plaintext Attack (KPA). It is shown that complete recovery of unknown key bits is possible upto $80$ bits in a practically feasible time and memory in random KPA situation by sequential computation when remaining $48$ bits are known. The results of key recovery in $64$, $72$ and $80$ bit unknown cases are extrapolated to predict the period of the iterative sequence generated in the local inversion approach for the full $128$ bit unknown key case and a strategy is proposed to search the actual period by brute force parallel search of the sequence period with $10$ free bits defining the search space. Then it is shown that the actual key can be verified in polynomial time by fast powering of the forward encryption map. Hence this strategy shows that the key recovery problem for AES128 under KPA has a chance of success in practically feasible time for a majority of the plaintexts. Local inversion approach to cryptanalysis using black box computations is a universal method applicable to a vast variety of key recovery and map inversion problems. Hence the results presented in this paper are representative of estimates of cryptanalysis of other ciphers which can be considered almost as strong as AES128 as encryption functions.

Three-dimensional time-harmonic Maxwell simulations generate massive complex indefinite systems whose mesh coarsening is strictly limited by phase accuracy. Although matrix-free finite element kernels utilize GPU throughput efficiently, standard multilevel solvers are ultimately bottlenecked by the memory and communication costs of exact coarse-grid factorizations. We present a fully matrix-free, factorization-free three-grid preconditioner for curl-conforming Nédelec discretizations with perfectly matched layers (PML) and optimally blended quadrature. The method employs an outer FGMRES to solve the unshifted fine-grid equation, while an intermediate-grid correction is computed by a fixed-work FGMRES preconditioned with a complex-shifted $2h$–$4h$ cycle. This strategically confines the complex shift to an auxiliary preconditioner, preserving the physical Maxwell operator. A local Fourier analysis derives the blended Maxwell branches and compatible edge transfers, identifying robust shift and Jacobi damping parameters. Validated against the analytical Maxwell Green tensor, our approach demonstrates extreme scalability: using a single solver configuration, both homogeneous and highly heterogeneous systems with approximately 10.89 billion complex edge unknowns are solved in 42.0–72.0 seconds on just 64 NVIDIA A100 GPUs.

A wireless sensor must alert a remote monitor before a monitored process crosses a safety threshold; an alarm arriving afterward may be too late. The sensor can react to its current estimate or predict ahead and trigger earlier, but the value of such lookahead is not obvious. In some systems it creates an early-alarm opportunity unavailable to the current test, while in others it cannot cross the alarm boundary. This letter gives a practical three-stage rule for deciding when to predict. First, an algebraic spectral test decides at design time whether lookahead is structurally useful: it is redundant exactly when the threshold direction is a left-eigenvector of the dynamics with a non-negative eigenvalue. Second, a closed-form channel decomposition shows that deeper prediction becomes more valuable as the channel degrades, because longer lead windows permit more pre-crossing transmission attempts. Third, simulations show that large gains also require retained prediction magnitude; oscillatory dynamics amplify the benefit through rotation, and a two-sensor setting reveals a sensing-channel tradeoff.

Unlike static medical question answering, long-horizon diagnosis captures the sequential nature of clinical practice: evidence is progressively acquired, integrated, and evaluated over multiple rounds of interaction before reaching a final diagnosis. However, existing doctor agents often fail when critical evidence is either not acquired or not adequately incorporated into diagnostic reasoning. Recent agentic approaches attempt to address these failures by reusing historical trajectories or distilled memories. But their diagnostic gains remain constrained because such experience may contain noisy or incidental information and is typically reused without validating which evidence actually drives diagnostic decisions. To address this limitation, we introduce CDEG, a graph-based framework that learns reusable decision-critical evidence from historical diagnostic trajectories. CDEG contrasts successful and failed trajectories from the same case to identify candidate evidence, validates their diagnostic impact through controlled counterfactual interventions, and organizes the resulting diagnosis–evidence–action relations into a structured graph. During inference, CDEG tracks the evolving patient evidence state to retrieve relevant diagnostic relations and selectively guide missing evidence acquisition or overlooked evidence reappraisal. Across in-domain and out-of-distribution benchmarks with multiple doctor agent backbones, CDEG consistently improves diagnostic performance, achieving up to an 11.5% accuracy gain over vanilla agents. These results demonstrate that reliable long-horizon diagnosis requires moving beyond trajectory-level experience reuse toward evidence-level learning of the factors that truly shape clinical decisions.

Diffusion language models (DLMs) alleviate the inherent latency bottleneck of autoregressive (AR) large language models (LLMs), but their degraded generation quality limits practical applicability. Although knowledge distillation (KD) can be a promising direction for improving performance, we empirically find that naively applying conventional KD yields only marginal gains, or even degrades generation quality. Based on these observations, we propose a novel self-distillation framework for DLMs, namely SelFusion. To enable effective KD without an external teacher model, SelFusion performs two forward passes with different masking levels, defining the hard mode with a larger masking probability and the easy mode with a smaller masking probability. However, the easy mode is not always more accurate than the hard mode and can be overconfident on incorrect tokens. Thus, we introduce bidirectional KD between the two modes, which can dynamically determine the distillation direction based on token-level correctness. Experimental results on instruction-following tasks show that the proposed self-distillation substantially outperforms other KD methods with external LLM and DLM teachers. In many configurations, the student trained with SelFusion even surpasses the performance of the LLM teacher, providing a practical path toward improving DLM generation quality. Source code can be found at https://github.com/scai-research/SelFusion_official

Robotic navigation in human environments requires a spatio-temporal semantic representation that can rec- oncile open-vocabulary perception with long-term environmental changes. While foundation models provide strong zero-shot recognition, their predictions are intermittent and view-dependent, and naively integrating them into mapping pipelines leads to identity drift and stale semantics over time. We present SuperMap, a 4D spatio-temporal mapping framework for language-guided navigation that integrates high-frequency geometric SLAM with asynchronous open-vocabulary perception. Our core contribution is a consistency-driven mapping engine that combines 3D-aware instance association/re-activation with a principled existence-and-label confidence update to maintain stable object identities and prune outdated map content under occlusions and scene changes. SuperMap produces a queryable 4D scene-graph representation that interfaces naturally with Vision-Language Models by supporting compositional queries over object semantics, relations, We demonstrate SuperMap on benchmarks and real robots, including dynamic scenes with appearance/disappearance and relocation, and provide ablations and runtime analysis. We release the full system as open-source to provide the community with a deployable baseline for open-vocabulary spatio-temporal mapping. Project website: superodometry.com/supermap.

Arabic harmful-language detection has received considerable attention, yet Arabic text detoxification remains underexplored. We introduce AraDetox, a multi-dialect Arabic detoxification dataset comprising 10,500 harmful social-media posts and 84,000 detoxified rewrites generated using GPT-5 and Gemini 2.5 Flash across Modern Standard Arabic, Gulf, Levantine, and Egyptian Arabic. The generated outputs were assessed through human evaluation and automatic analyses of lexical change, semantic preservation, sentiment, and dialectal style. Results show that detoxification is primarily a meaning-preserving rewriting task: substantial lexical and structural reformulation is accompanied by consistently high semantic similarity. Human evaluation confirms successful harmful-language removal while largely preserving the original meaning. Dialectal analyses further indicate that the generated variants exhibit measurable stylistic alignment with reference Arabic dialect corpora. Comparison with existing resources highlights two complementary approaches to detoxification: minimal-edit lexical substitution and meaning-preserving reformulation. Our findings demonstrate that large-scale Arabic detoxification resources can be constructed through LLM-assisted generation and human verification. The dataset is publicly available at https://github.com/ArabicNLP-UK/AraDetox to support future research on Arabic detoxification, safe text generation, and multi-dialect Arabic NLP.

Quantum states with conserved total spins, or spin eigenfunctions, are important for studying quantum chemistry and quantum manybody physics problems. A typical class of spin eigenfunctions are Dicke states, which attain maximal spins. While we already have many efficient quantum algorithms to prepare Dicke states, it is not yet clear if we could do so for arbitrary spin eigenfunctions deterministically. Generalizing Bärtschi and Eidenbenz's elegant algorithms for Dicke state preparation, we successfully prepare arbitrary spin eigenfunctions characterized by branching paths and binary spin trees. As a byproduct, we also develop the corresponding classical algorithms to reconstruct all these spin states.

5G networks employ multiple radio channels to meet growing demands for bandwidth and high-resolution video streaming for emerging applications. However, existing multipath video systems are largely designed around monolithic codecs, which require sufficiently complete chunk delivery, or layered codecs, which depend on timely base-layer delivery. Under fast-varying 5G conditions with blockage, handovers, and heterogeneous path capacities, we observe that decoding dependencies in existing codecs make multipath delivery fragile: transient under-delivery of critical video data can directly trigger stalls and degrade QoE. This paper proposes NeuralMDC, a neural multiple-description video codec co-designed with multipath streaming for dynamic 5G networks. NeuralMDC encodes each video chunk into independently decodable and mutually refinable description streams, each spanning the full chunk. This design changes the multipath delivery unit from dependent packets or layers to independent chunk-level streams, so missing streams primarily reduce quality rather than making the chunk undecodable. Built on NeuralMDC, we develop a user-space multipath streaming system that maps description streams to heterogeneous 5G paths with simple yet effective scheduling logic. Across trace-driven emulation and operational 5G experiments, NeuralMDC improves QoE by 26%-44% over existing monolithic, layered, and neural streaming systems, improves video quality by up to 41.8%, and keeps stall ratios below 0.32%.

Attacks against Internet-connected IoT devices continue to increase; however, transforming observed attack traffic into deployable intrusion detection system (IDS) rules remains largely a manual process. Recent studies have explored using large language models (LLMs) to generate IDS rules; nonetheless, existing approaches often require auxiliary information beyond observed traffic or generate rules without validating their detection logic against benign traffic. This study presents a verification-guided specification synthesis framework for generating Suricata rules directly from HTTP request traces. Instead of having an LLM generate IDS rules in a single step, an LLM first identifies a vulnerable parameter and synthesizes a semantic detection specification. These specifications are iteratively refined through counterexample-guided inductive synthesis (CEGIS), in which benign traffic samples serve as counterexamples during synthesis and verification. Verified specifications are then deterministically compiled into Suricata rules. Experiments on 281 real-world CVEs and benign traffic collected from real IoT devices show that the proposed method achieves a detection rate of 81.5% while maintaining a false positive rate of 0.0%. An ablation study also demonstrates that CEGIS-based verification improves detection performance while maintaining a low false positive rate.

Reconstructing photorealistic scenes in unconstrained underwater environments remains challenging due to severe media-induced light scattering and unpredictable dynamic objects. Recent feed-forward visual foundation models have demonstrated remarkable capabilities in generalized novel view synthesis and tracking. However, when directly applied to aquatic videos, optical attenuation and motion interference fatally corrupt their feature aggregation, leading to severe tracking and reconstruction failures. To overcome these limitations, we present NemoSplat, the first feed-forward 4D Gaussian Splatting framework tailored for media-aware dynamic reconstruction directly from uncalibrated marine videos. Beyond providing robust estimations of camera poses and dense scene depth, we devise a Promptable Dynamic Disentangler that utilizes a confidence-aware fusion strategy of learned dynamic probabilities and optional semantic text priors, effectively isolating massive transient entities. Furthermore, to counteract visual degradation, a Media-Aware Gaussian Predictor is formulated to jointly estimate intrinsic 3D Gaussian attributes alongside physical media parameters, rendering pristine scene appearance in a single forward pass. Additionally, we introduce a large-scale underwater dataset with massive dynamic elements to facilitate training and evaluation. Extensive experiments on our dataset demonstrate that NemoSplat achieves state-of-the-art tracking accuracy and high-fidelity rendering. Homepage: https://nemosplat.hkustvgd.com

Large language models are increasingly used to judge people, and their answers can depend on inputs correlated with a protected attribute such as race, even when that attribute is never shown. Interpreting this reliance requires knowing how much predictive information those inputs add, but audits often report only whether a decision changes. We introduce an evidence-referenced audit that compares proxy reliance with two references, the known outcome rule and a Bayesian linear regression fitted to the same labeled examples, while measuring ranking accuracy separately. We apply it to four models in a simulated patient-ranking task with known field weights. We find four results. First, at zero predictive value, the models still relied on fields that carried no information, while the same-example learner did not. Second, as proxies became predictive, reliance grew far less than the reference. Third, display format mattered. Ordered labels drew more reliance than arbitrary labels, a direction that held in all four models. Fourth, ranking accuracy did not consistently track these changes in reliance. Based on these findings, we argue that decision changes and accuracy cannot show whether proxy reliance is warranted by the evidence. Audits should report reliance against an explicit predictive reference, alongside accuracy. Ordered fields whose values move with group membership, such as scores or income bands, are a useful starting point. Together, these results make it possible to measure how far reliance departs from the evidence, rather than infer it from observed choices alone.

Referring Expression Segmentation (RES) aims to generate a pixel-level mask for the object specified by a language expression. Recent methods based on multimodal large language models (MLLMs) often rely on one-pass coordinate prediction for visual localization, which serializes continuous spatial locations as discrete text tokens and may lead to localization bias and alignment errors. To address these issues, we propose DRAgent, an MLLM-driven discriminative reasoning (DR) framework for RES. Instead of requiring the MLLM to generate localization coordinates, DRAgent first constructs a detector-generated candidate space and then uses the MLLM as a visual-semantic target discriminator. Specifically, the MLLM performs reliable target selection among potential distractors through a two-stage DR mechanism, which first screens high-recall candidates and then performs instance-wise verification. The selected target box is subsequently used as a spatial prompt for a foundation segmentation model to produce the final pixel-level mask. Furthermore, we construct a self-consistency-filtered reasoning-chain data pipeline for LoRA-based fine-tuning, providing more reliable supervision for enhancing the MLLM's discriminative reasoning capability. Experiments demonstrate that DRAgent achieves competitive performance on RefCOCO, RefCOCO+, and RefCOCOg.

Societies of interacting language-model agents offer a controllable and repeatable way to study collective behaviour at scales that would be difficult to test with people. Their scientific value, however, depends on whether a social mechanism that works in a small group still operates when thousands of agents interact, and testing this directly requires costly large-scale runs. Here we introduce an audit that predicts a mechanism's fate as a population grows. It asks how often the mechanism can act, whether agents use the information it supplies, and whether the measurement itself creates apparent scale effects. Controlled experiments show that a single structural term can decide whether reciprocity, consensus or punishment survives scaling. For gossip, the population at which the mechanism fails is set by the reach and lifetime of its messages. In language-model societies, agents respond not only to social information but to how it is expressed: counts and percentages led to different scale behaviour. Predictions made before execution held on third-party code and a second model family, while a failed prediction exposed the boundary of the finding. The audit provides a prospective way to decide which social mechanisms can be interpreted across population scales.

Multimodal speculative decoding accelerates vision-language models by allowing a lightweight draft model to propose candidate tokens for parallel verification by a larger target model. Existing methods typically condition the drafter on a fixed visual interface, such as a predefined visual-token budget or a static compressed representation. However, our controlled visual-budget analysis shows that visual demand varies substantially across tasks and decoding stages, which means more visual input is not always beneficial. Actually, insufficient evidence may weaken visual grounding, while excessive context adds overhead and may disrupt drafting. We propose FOVEA (Focused On-demand Visual Evidence Adaptation), a cache-friendly approach that builds a reusable visual memory and dynamically retrieves a bounded subset for a draft state. A cumulative-mass rule determines both how many and which entries are selected. The selected entries are aggregated into a visual readout and fused with the current draft hidden state through a lightweight gated residual correction. Rather than inserting visual tokens into the autoregressive context, the correction modifies only the representation passed to the language-model head. Experiments across multiple vision-language backbones and multimodal benchmarks show that FOVEA improves draft acceptance and end-to-end decoding speed, achieving up to $2.13\times$ speedup over autoregressive decoding. These results demonstrate that state-conditioned evidence retrieval is an effective alternative to reusing a fixed visual representation throughout multimodal generation.

The collision operator for the Boltzmann equation is a nonlinear nonlocal operator. When lifted in the extended 2-particle space, it is viewed as the projection of a collisional linear operator. In this paper, we propose a particle method that samples the post-collision relative velocity directly from the Green's function of this operator (the transition probability of the generated time-continuous Markov chain). The normalizing flow amortized sampling is then proposed to reduce the sampling complexity. The resulted method takes $O(N)$ each time where $N$ is the particle number, and conserves momentum and energy exactly. This method does not require the boundedness of the kernel and, more importantly, it allows learning the Green's function directly from the scattering data without selecting the kernel in a specified family.

Task-specific lightweight models for surgical phase recognition excel at capturing temporal dynamics but generalize poorly under domain shift. Conversely, surgical foundation models (FMs) offer superior transferability via large-scale pretraining, yet their lack of explicit temporal modeling often yields temporally inconsistent predictions, leading to degraded performance. To exploit the complementary strengths of both paradigms, we propose Large-Small Temporal adaptation (LaST), a novel large-small collaborative framework that enables zero-shot adaptation to unseen clinical domains. In LaST, the FM initiates the pipeline by generating frame-level phase priors that serve as initial weak supervision. To effectively utilize these noisy phase priors, we introduce an iterative temporal refinement scheme that integrates dynamic quality control to filter reliable predictions and dual-model cross-learning to mitigate confirmation bias. Simultaneously, the lightweight model leverages its intrinsic temporal modeling ability to progressively correct inconsistent predictions and enhance overall accuracy across iterations. At the end, a cycle replay strategy is employed to close the loop: the refined, more accurate predictions are utilized as upgraded supervision signals for the subsequent iterations, fostering a self-reinforcing evolution of both label quality and model capability. Extensive experiments demonstrate that LaST achieves robust adaptation to unseen domains for zero-shot surgical phase recognition, outperforming the baseline (PeskaVLP) by 24.85%-43.17% in accuracy and even surpassing fully supervised linear probing and several state-of-the-art few-shot approaches. Codes will be released at https://github.com/YIYIZH/LaST.

Hybrid sequence models must satisfy prefix invariance: representations at position t must not depend on future inputs, yet this is rarely verified. We formalize prefix invariance and give a lightweight audit, two forward passes, no training or gradients, yielding a per-layer score localizing where causality breaks. Attention-mask inspection, the field's default check, is incomplete: causality is a graph-level property, and leaks can occur via scans, aggregations, or normalization despite correct masks. Across 192 injected-fault trials on eight checkpoints, mask inspection detected none, while our audit localized all 192/192 to the exact layer. Static/dynamic analysis of chunked-scan code in transformers found the same defect in Zamba2 and Nemotron-H, an inter-chunk axis error fixed via the reference implementation. The method fits on one page and runs in seconds.

Neurobiological studies and representation learning have observed that representations of objects belonging to the same category in high-dimensional neural spaces exhibit low-dimensional object manifold characteristics, and different object manifolds are linearly separable in these neural spaces. However, these experimentally observed phenomena lack rigorous theoretical validation to date. This paper proposes a new stochastic separability theorem for embedding manifolds of two different object categories. First, we establish a projection measure concentration theorem for embedding manifolds under general conditions. We develop a new two-layer measure concentration analysis technique, which unifies two estimation bounds via the law of total expectation to derive measure concentration inequalities. Based on the measure concentration theorem, we further prove a stochastic separability theorem for embedding manifolds of two different object categories. If two datasets have distinct means and bounded total variances, their samples become linearly separable with high probability, provided that the projection direction satisfies a non-singularity condition. The main contributions of this paper are twofold: 1. We prove the projection concentration properties of embedding manifolds in high-dimensional spaces by using two-lawyer tail-bound inequalities. 2. We identify a non-singularity condition for the stochastic separability between embedding manifolds, and rigorously prove the stochastic projection separability theorem. The theorem not only uncovers geometric and statistical properties of the object embedding manifolds, but also provides a novel mechanism for representation learning in deep networks.

Speech-based applications pass spoken queries through automatic speech recognition (ASR) before any retrieval module, so ASR errors enter the pipeline as a fixed upstream constraint. We empirically test whether two extensions to standard retrieval-augmented generation (RAG), entity-graph linking and iterative reformulation, absorb or amplify these errors. Using four English accents synthesized through neural TTS, we evaluate four RAG configurations on three multi-hop QA benchmarks (HotpotQA, 2WikiMultiHopQA and MuSiQue) against a clean-text oracle. Although the structurally richer configurations generally retain higher absolute F1 under ASR input, both extensions amplify the error: the F1 gap from clean text to the highest-WER accent is 36-67% larger under their combination than under naive dense retrieval, on all three benchmarks. The dominant failure mode is corruption of one or more query entities, accounting for 87-96% of degradation cases on 2WikiMultiHopQA across all four methods. Two lightweight surface-form mitigations leave most of the gap intact, indicating that downstream retrieval structure amplifies remaining entity errors. We release code and data at https://github.com/Continuum-AI-Corp/spoken-multihop-rag .

A graph $G$ is apex if $G$ has a vertex $v$ such that $G-v$ is planar. We prove that every $2$-connected apex cubic graph is three-edge-colorable. This result gives the final piece of the proof for the well-known Tutte's three-edge-coloring conjecture from 1966 \cite{tutte}. The proof, as well as the result, generalizes that of the Four Color Theorem, which requires computer checks. As in the previous proof of the Four Color Theorem, the proof is constructive. More precisely, given a $2$-connected apex cubic graph $G$ on $n$ vertices, our reducibility and discharging procedure yields a three-edge-coloring of $G$ in $O(n^2)$ time. As an additional reproducibility check for our computer checks, independent implementations reconstructed from the detailed pseudocode (given in the appendix) using generative AI systems reproduced the required computational results. These reconstructions are not part of the mathematical justification of the theorem, but provide additional evidence for the reproducibility of the computations.

While Vision-Language-Action (VLA) models have leveraged internet-scale pretraining and task-specific finetuning to achieve strong performance on long-horizon tasks, they often struggle with non-Markovian tasks that require memory. Existing approaches to memory typically involve additional Vision-Language-Models (VLMs) for long-term memory management, introducing a memory bottleneck and a fractured training pipeline. Conditioning on multiple historical frames can provide the VLA with access to more descriptive features of past scenes, but can degrade performance if frames are chosen at arbitrary, fixed intervals. To address these limitations, we present UniMem, a framework that unifies high-level visual and textual memory and low-level control under one backbone. UniMem employs an event classifier to initiate textual and visual memory updates, a keyframe encoder for dense visual memory, and a keyframe caching technique to minimize overhead during policy rollouts. We evaluate UniMem across five simulation and four hardware tasks targeting sequential and spatial memory, demonstrating that our unified, single-model system outperforms fixed-interval image sampling baselines (93.4% vs. 68.2%) in simulation and hierarchical VLM-VLA baselines (80.0% vs. 43.5%) in hardware, while offering faster inference and a simple training pipeline for easy adoption. Project website: https://losterberg3.github.io/unimem-vla/

Applications that deploy LLM agents keep revising the permissions their agents use. When authority is consumable, even widening a permission can break a legitimate workflow, because an early call may use up a grant that a later call needs. Checking individual requests, or watching one run succeed, misses this. We present Agenra, a policy language with a Lean enforcement core that tracks the values each call uses, resources discovered at run time, and authority consumed by earlier calls. Agenra treats safety and workflow preservation as separate properties. Our key idea is to certify a revision with a simulation between the old and new permission ledgers. Quota increases, appended grants and independent reorderings admit such simulations by construction, and Lean checks proposed simulations for other finite-quota revisions. The proofs use the deployed matching predicates and extend through the full monitor, covering discovered resources, payloads that are never enumerated, failed calls and approvals. They assume serialized calls and identical service results under both policies, and they trust the Python translation layer. On 51 authored revisions, the checks certify all 21 that preserve their clients and none of the 30 that do not, and every lost workflow is confirmed by replay. They also certify a real quota increase from an open-source agent application. In a report-delivery application, raising the send quota preserves every client yet permits a second approved send, so preservation does not imply safety.

Storing face images under a hard sub-kilobyte budget, as required for identity documents, smart-card biometrics and bandwidth-constrained verification, forces a codec to discard most of the signal while keeping what a face matcher actually reads: identity. Generic codecs optimize pixel fidelity, not the embedding distances that drive verification, so which codec, resolution and setting best preserve identity at 1024 bytes or less, and how that degrades at 512, is unclear. We benchmark ten general and face-specific codecs across resolutions, byte budgets, two datasets (controlled Color FERET, in-the-wild AI-Solutions-KK) and four anchor face matchers, with a fourteen-model ViT and CNN roster confirming the ranking is backbone-invariant. We then train a custom identity-preserving codec that hits the byte budget exactly via binary search over a frozen gain table, and run four studies: resolution, demographic fairness, recompression, and no-box adversarial robustness. Sub-kilobyte identity preservation is feasible, but which codec to deploy depends entirely on the budget. At 1024 bytes and the 112 px working resolution the problem is close to solved: modern codecs hold Color FERET equal-error rate under 0.35 percent on the ArcFace anchor. At 512 bytes the field re-sorts: AVIF, HEIF, JPEG XL and legacy JPEG collapse to 28 to 98 percent false-non-match rate at FMR 1e-4, while WebP, JPEG-AI and our byte-budgeted learned codecs stay out of that band, with 24.3 percent for WebP against 6.9 percent for our accurate variant in the wild. That re-sort, not the 1024-byte ranking, is the operational result: a codec chosen at 1 kB is not the codec to deploy at half that.

Multimodal models often align language, vision, and audio in a single final-layer latent space, implicitly assuming that task-relevant evidence emerges at the same semantic depth across modalities. Using layer-wise CKA analysis, we observe that this assumption leads to semantic granularity mismatch: textual cues usually require deeper contextual abstraction, whereas visual and acoustic cues often provide discriminative perceptual evidence in shallow or middle layers. This mismatch can flatten fine-grained modality-private cues and reduce reliability under noisy, imbalanced, or missing inputs. To address this, we proposed Adaptive Hierarchical Representation Alliance (AHRA), a hierarchical shared–private expert framework. AHRA factorizes each modality into shared and private streams across semantic levels, regularizes them with shared alignment and private decorrelation, routes shared information through a cross-modal expert, and enhances task-relevant private tokens with modality-specific experts guided by a sparsity-controlled soft-gating mechanism (foreground exam). A hierarchical co-fusion module then performs intra-level expert coordination and inter-level semantic selection. Experiments on six benchmarks across image-text classification, multimodal intent recognition, and trimodal sentiment analysis show that AHRA consistently improves over strong baselines and remains robust under noisy and missing-modality settings.

Population optimizers such as CMA-ES, DE, and multi-objective evolutionary algorithms drive search mainly through selection signals that are scalar or rank based: such a signal indicates that one candidate outperforms another, but not the local direction responsible for the improvement. JANUS (Jacobian-Aligned Newton-Unified Search) is a plug-and-play infill module that extracts this missing local geometric signal without replacing the host optimizer. It estimates a local Jacobian from the recent evaluation trace; the same Jacobian yields both a damped Gauss–Newton exploitation candidate and a trace-preserving exploration metric, reserving a fraction of the host's per-generation candidate slots for geometry-guided infill rather than spending evaluations on top of the host's budget. Unlike MetaBBO methods, JANUS needs no offline training or task distribution, estimating this geometry on the fly from the current run alone, while the host keeps full control of selection, survival, covariance adaptation, and step-size control. Under same-protocol comparisons, JANUS improves the CMA-ES host on 11–15/16 BBOB functions across $d\in\{30,100,500\}$. It also attains the best mean error on 13 of the 16 functions at $d{=}500$ in the complete NN-BBO/MetaBBO baseline comparison, with no training cost, and yields a $936\times$ geometric-mean improvement over the host on a $d{=}1000$ BBOB subset. On structured and multi-objective tasks, JANUS gives the best mean cost on 1135-dimensional UAV path planning ($-12.8\%$ vs.\ the strongest baseline), and it improves SMS-EMOA/AGE-MOEA2 hosts on 12/38 multi-objective tasks with zero significant regressions. Code is available at https://github.com/hongyuanyu/JANUS.

State Space Models (SSMs) have surfaced as a promising architecture in Video Frame Interpolation (VFI), as they can capture long-range dependencies with linear computational complexity. However, their predefined scanning order limits their effectiveness in modeling the dynamic motion trajectories inherent in VFI problems. To tackle this challenge, we propose Motion-Guided Mamba for Video Frame Interpolation (MGMVFI), an adaptation of the selective state space model tailored explicitly for VFI. MGMVFI introduces Motion-Guided Serialization (MGS), which leverages optical flow to define a motion-adaptive 1D input order for the SSM. This aligns the causal state updates with semantically related tokens, enabling motion-consistent feature propagation, particularly for large and dynamic motions. Additionally, to mitigate the unreliable feature representations caused by inaccurate optical flow estimates, we introduce contextual synthesis that utilizes the surrounding spatial context for robust inter-frame feature synthesis. These components are seamlessly integrated within our tailored Mamba architecture, which also employs a lightweight refinement block to enhance local detail reconstruction at a reduced computational cost. Extensive experiments on standard VFI benchmarks demonstrate that MGMVFI achievesstate-of-the-artperformance,particularly on complex and dynamic motions, thereby establishing a new direction for sequence modeling in video interpolation.