Papers

World Action Models (WAMs) jointly predict robot actions and future observations. However, their visual prediction objectives do not explicitly distinguish task-relevant transitions from incidental scene changes. For whole-body policies, this distinction matters because task progress depends on coordinated changes in body pose, contact, and object state. We introduce LANGUAGE-FORCING, a training strategy that uses language to emphasize task-relevant changes in the future visual representations of a whole-body WAM. Given paired current and ground- truth short-horizon future frames, a frozen vision-language model assesses task progress and predicts plausible subsequent state changes. We encode these descriptions with a frozen text encoder and align their embeddings with intermediate, action-conditioned future-visual features, complementing the original visual and action prediction objectives. All auxiliary components are used only during training, introducing no additional inference overhead. On HumanoidArena, LANGUAGE-FORCING achieves an average success rate of 79.17%, exceeding the strongest competing policy by 9.46 percentage points and the same WAM without language alignment by 10.77 percentage points. Across five real-world tasks, language alignment raises average success from 53.33% to 71.33% over the same WAM baseline.

Transformer compression often couples two distinct quantities: retained attention capacity and the number of sequential depths at which that capacity is executed. We study whether useful attention functions in a hierarchical vision transformer are locally mobile across depth. Using SegFormer-B5 for human segmentation, we find that some task-relevant attention heads can be relocated to nearby depths, but only when their computational context is respected. Head-only relocation is insufficient, while joint adaptation of the relocated head and surrounding representation boundary improves functional preservation; replacing donor-specific normalization and spatial-reduction preprocessing causes large degradation. We use this path-conditioned mobility to reorganize retained attention from depth into width. Starting from a fixed parent with 120 retained Stage-3 heads across 40 attention execution depths, we consolidate the same 120-head budget into 20 execution hubs. Under a layout-matched three-seed recovery control, K20 reaches 0.960686 +/- 0.000059 mIoU versus 0.960400 +/- 0.000074 for the corresponding State120 control. The paired point difference is +0.000285, while a paired image-bootstrap confidence interval includes zero; we therefore interpret the result as no measurable quality penalty rather than statistical superiority. For deployment, FOLDED_VEC preserves donor-specific paths while vectorizing their execution. Relative to the same 120-head State120 parent, it reduces end-to-end NVIDIA L4 latency by 5.28%, 3.15%, and 3.58% at batch sizes 1, 4, and 8. These results suggest that attention redundancy includes not only removable capacity but also execution-depth redundancy, provided that the computational path defining a relocated function is preserved.

Long video understanding requires reasoning over hours of content that far exceed the context window of multimodal large language models (MLLMs). Existing approaches compress video into fixed-length clips, yet events are fragmented and unrelated content is merged, yielding noisy retrieval. To overcome this challenge, we propose EMMA, an Event-driven Multimodal Memory system with Multi-Agent retrieval, which organizes memory around semantic events in three stages: (1) Event segmentation: a dual-stream detector adaptively locates event boundaries by fusing visual semantic drift with spoken transitions, rather than fixed windows. (2) Memory construction: built on the detected events, a three-level hierarchy of raw video, clip-level memory, and event-level memory stores content at progressively higher semantic abstraction. (3) Memory retrieval: a multi-agent framework collaboratively queries this hierarchy through query decomposition, tiered search, evidence validation, and answer generation, narrowing evidence from coarse events to precise clips. Extensive experiments on three challenging benchmarks demonstrate substantial improvements, showing the effectiveness of our event-aligned memory system. Code, prompt templates, and reproduction scripts are available at https://github.com/BAI-LAB/EMMA.

A robotic agent operating in an unknown environment must gather sufficient task-relevant information before acting. However, determining when the available information is sufficient remains a challenge. We propose a conceptual hybrid neuro-symbolic framework for task-sufficient active perception in unknown environments. After interpreting a requested task, the agent generates an initial plan and iteratively reevaluates it as new information becomes available. At each iteration, it can acquire a more informative viewpoint, inspect or manipulate candidate objects, ask the user for clarification, or perform the task. Neural perception modules update an object-centric belief state, while a symbolic decision layer determines whether the remaining uncertainty justifies further information gathering. The framework will be implemented on a mobile manipulator comprising a humanoid torso and a mobile base, with complementary evaluation in simulation. The proposed approach aims to enable robots to determine not only how to gather additional information, but also when further perception is no longer necessary.

Long-term personalized agents are required to reason over distributed history across modalities and evolving dialogue states. While existing flat cross-modal methods compress visual evidence prematurely and struggle with temporally stale observations, they neglect the necessity to unify multimodal and cross-session evidence. Crucially, raw multimodal memories are inherently entangled and heterogeneous, making fine-grained visual grounding and cross-modal reasoning computationally prohibitive. To this end, we present latent multimodal memory reasoning, i.e., LMMR, a novel neuro-symbolic framework that shifts long-term multimodal memory organization into a continuous latent space. Specifically, (i) instead of converting images to text and relying on static retrieval, LMMR maps both textual hidden states and visual patch features into a shared Sparse Autoencoder concept space, establishing sparse activations as common semantic addresses to dynamically construct a turn-level multimodal memory graph. (ii) A graph encoder then aggregates cross-modal and temporal dependencies across the task-specific latent subgraph, injecting the graph readout as a continuous residual rollout into the generation backbone while retaining native visual tokens to preserve fine-grained details. Extensive experiments on nine Mem-Gallery tasks and twelve MemEye settings demonstrate that LMMR significantly outperforms state-of-the-art baselines in capturing complex multimodal dependencies while enabling faithful and grounded responses.

Locally recoverable codes (LRCs) permit reconstruction of a lost coordinate from a small subset of the remaining coordinates, a property that is a fundamental coding technique for distributed storage and, more recently, for fault-tolerant quantum memories and distributed storage systems. Existing constructions of quantum locally recoverable codes (qLRCs) and their entanglement-assisted generalizations (EA-qLRCs) impose a single locality parameter $r$ governing recovery from both bit-flip ($X$-type) and phase-flip ($Z$-type) errors, even though physical noise is rarely symmetric between the two. We introduce entanglement-assisted quantum locally recoverable codes (EA-qLRC) with asymmetric locality $\{r_X,r_Z\}$, in which $X$-type and $Z$-type errors are corrected from recovery sets of size at most $r_X+1$ and $r_Z+1$, respectively. We extend a stabilizer sufficient condition for this property and, in particular, the CSS framework, where asymmetric locality corresponds to separate locality conditions on the two classical codes $\mathcal C_1$ and $\mathcal C_2$ underlying the construction. From this characterization we derive Singleton-, Griesmer-, Plotkin-, sphere-packing-, and Cadambe\textendash Mazumdar-like bounds for CSS-type asymmetric EA-qLRCs (AEA-qLRCs) built from Proposition~\ref{CSS-EA-qLRC-Asym}, and we characterize exactly when a pure CSS-type AEA-qLRC built from a pair of classical LRCs attains the Singleton-like bound with equality.

Multi-object grasping is a fundamental yet challenging problem, constrained by data collection costs, varying spatial layouts, and the need for broad generalization. To address these challenges, we propose MoDex, a data-efficient framework that provides the first unified solution for universal multi-object dexterous grasping across diverse spatial layouts. MoDex accomplishes this by formulating the task as a single-step decision problem of learning residual corrections to a reference trajectory, and introducing a spatially adaptive dual-strategy that performs single-shot grasping for proximal objects and sequential interaction for separated objects. Extensive evaluations show that MoDex significantly outperforms baselines in proximal configurations, achieving 61.24% success on unseen objects, and achieves high success rates in separated layouts. Further experiments demonstrate its generalization across robot embodiments, scalability and zero-shot sim-to-real deployment through a vision-based policy trained solely on simulation data. Code and pretrained models are available at https://github.com/Aurora-Xiong/MoDex.

Vision-language models are increasingly used for driving-scene understanding, yet their defensive reasoning under weakly structured Global South traffic remains unclear. We introduce BD-HazardVLM, a diagnostic benchmark of 500 Bangladesh road-scene images, comprising 375 from RSUD20K and 125 from TFP-BD, annotated for hazard presence, primary hazard category, risk timing, severity, defensive actions, visible evidence, and hallucination- and Western-road-bias traps. We evaluate Qwen2.5-VL-7B, InternVL3-8B, and MiniCPM-V-4.6 using a shared zero-shot task definition and structured output schema. Binary hazard-presence accuracy ranges from 88.0 percent to 89.2 percent, essentially matching the 89.0 percent majority-class baseline. Fine-grained reasoning is substantially weaker: category accuracy remains below 40 percent, risk-timing accuracy ranges from 1.6 percent to 12.4 percent, and exact defensive-action matching remains below 2 percent. Although InternVL3-8B achieves the highest severity accuracy at 61.2 percent, it has zero recall on the 58 high-severity cases. Qwen exhibits severe category mode collapse, while all models show weak latent-hazard anticipation and near-zero self-reported trap recognition. For the evaluated models, these findings show that coarse hazard recognition does not imply reliable fine-grained defensive-driving reasoning and motivate geographically diverse safety evaluation.

We study the problem of learning diffusion-based world models that jointly capture scene geometry and visual appearance in complex first-person environments. Existing approaches typically train a single action-conditioned model that predicts future RGB frames, but they either ignore geometry or require action supervision for all modalities, limiting scalability to large unlabeled video datasets. In this work, we introduce a two-stage diffusion world model that separates geometric and appearance prediction while maintaining joint temporal coherence. Our key idea is a staggered RGB-depth formulation: depth at time $t+1$ is predicted first using an action-conditioned diffusion transformer, and the RGB frame at the same timestep is then generated conditioned on this predicted depth using a second diffusion model without action inputs. We show that this decoupled design faithfully approximates a hypothetical joint RGBD diffusion model trained with diffusion forcing, enabling the second stage to be trained or fine-tuned on additional RGB-depth data without requiring action labels. Experiments on the CS:GO Dust 2 dataset demonstrate that our approach improves both depth accuracy and RGB fidelity compared to fully joint models and RGB-only baselines, while significantly enhancing scalability to augmented or unlabeled datasets. Our results highlight the effectiveness of geometry-guided diffusion for world modeling and open a path toward large-scale action-free video model training. Code to be released upon acceptance of paper.

This study investigates generalized boundary representations and transfer learning for physics-informed neural-network (PINN) estimation of spatially varying convective heat transfer coefficients (HTCs) in jet impingement cooling. Piecewise-linear (PWL) and neural-network boundary estimators are inferred from transient interior-temperature measurements and evaluated against CFD-CHT reference HTC profiles at Reynolds numbers of 5000, 10000 and 15000. A discrete Adjoint-FEM formulation with gradient regularization based on the $H^1$-seminorm provides an independent inverse-method comparison. Transfer learning is investigated under two source-target conditions: transfer between noise levels at the same Reynolds number, and transfer from a controlled Gaussian-profile problem to CFD-CHT-derived target problems at different Reynolds numbers. For the representative Re = 5000 target with 5% measurement noise, quadratic, PWL, and neural-network PINNs achieve relative HTC-profile errors of 5.08, 5.31 and 5.19%, respectively, while generalized representations reduce the error for the Re = 15000 target from 16.76% for the quadratic PINN to 14.41% and 14.46%. Transfer learning reduces target-training time by 57.86-81.71% for fixed-Reynolds-number transfer and by 50.59-84.94% for Gaussian-to-CHT transfer. Its effect on reconstruction accuracy is target dependent: fixed-profile PWL transfer remains within 0.30 percentage points of scratch training, whereas Gaussian-to-CHT transfer produces differences of 0.10-1.52 percentage points across targets and representations. These results demonstrate that generalized PINN representations can accommodate spatially varying HTC profiles without a prescribed quadratic form, while transfer learning can substantially reduce target-training cost for repeated inverse estimation, with the resulting reconstruction accuracy depending on source-target compatibility.

A familiar line of reasoning runs as follows: language models are flexible approximators trained under a strictly proper loss, so scale brings them closer to Solomonoff induction, and their predictive success is evidence of that approach. We apply known results from algorithmic information theory, universal prediction and Bayesian consistency. In its uniform form the conclusion holds for no computable model: each has cheap environments it never learns, and every uniformly computable family of total predictors shares one on which each member fails beyond a horizon growing with its index, so “closer” can hold only pointwise, one environment at a time. Nor could the offered routes certify even that: finite evaluation cannot certify agreement on rarely reached contexts, perfect corpus fit selects the corpus law as prior, and asymptotic criteria cannot see the constants that carry Solomonoff's guarantee. Our lens is the Kraft inequality read for predictors. Every model has at most the same budget of predictive success, so Solomonoff's predictor, which leaves part of it unspent, differs from a language model not in having more of it but in where it spends it; on mutually singular families this is exact, and only full-mass Bayes mixtures reach the frontier. We then discuss what is inheritable, including in-context learning as amortised Bayesian inference; what survives within a context window; which Solomonoff-like properties a deployed system should have or avoid; and what served conditionals determine. A closing table states what common practices certify and what to report. Tests are developed in a companion paper.

Since compact, interpretable classifiers correspond to simple programs, it is natural to ask if LLM-based coding agents can be used to learn interpretable models from data. In this paper we present a simple harness that repeatedly prompts a coding LLM to produce the next rule in a rule set, given a sample of labeled examples. We show that this approach is competitive with, and often superior to, state-of-the-art rule-learning methods on tabular data, especially with small training sets, and wins by large margins on text, including against another LLM-based rule learner; that this approach outperforms standard harnesses, and that the approach immediately leads to new kinds of interpretable models, since coding agents are not restricted to search a highly restricted hypothesis space. In addition to the advances made in learning interpretable models, these results are of interest because they introduce a new class of coding benchmarks - construction of interpretable classifiers - which includes quantitative measurements of output quality (i.e., test-set error) and existing mature “specialist coding algorithms” (i.e., existing learning methods).

Masked diffusion language models (MDLMs) decode by repeatedly committing tokens to masked positions, but these commitments are usually irreversible. A token chosen under sparse, partial context is kept fixed, even when later context no longer supports it. Existing samplers mainly decide when to commit a token, but rarely check whether an already committed token should still be kept, allowing early mistakes to propagate. We trace this issue to confidence drift, where the model's confidence in a committed token drops from its sparse commit-time context to the denser context available later. Based on this signal, we propose CoDR (Confidence Drift Remasking), a training-free and sampler-agnostic refinement pass. CoDR estimates drift for all committed positions in only k forward passes via k-partition probing, then remasks and regenerates only the tokens the model no longer endorses. Across two backbones, four reasoning and coding tasks, and three base samplers, CoDR improves average accuracy across all evaluated model-sampler configurations and improves most individual task settings with modest overhead. Controlled experiments show that the gains come from targeted confidence-drift remasking rather than extra compute alone, and that CoDR uses far fewer forward passes than prior remasking methods. Code is available at https://github.com/YueWu0301/CoDR.

Localizing latent structures in the activation space of language models (LMs) is central to understanding and controlling their behavior. Yet, localized structures can differ substantially in their causal influence, raising the question of what makes a structure actionable. We tackle this question by casting causal influence as a product of three factors and showing empirically that they act as interpretable, distinct constraints: capacity, measuring the sensitivity of the model's output to movement along the structure, responsiveness, capturing how promotable the concept is given the current context, and alignment, reflecting how well the structure aligns with the context-specific representation of the concept. Across 4 LM families and 50 concepts, we observe that causal effectiveness requires all factors to be high; low capacity and responsiveness reduce it by 84% and 95%, respectively, while low alignment can reverse it, suppressing concept expression. Moreover, we find that causality is context-dependent rather than an intrinsic property of the structure, with causally effective directions forming a low-dimensional subspace that varies across contexts. By restricting the training of linear probes to this subspace, we introduce causal probes that achieve 17%-118% improvement in steering across models, with only 3% reduction in concept detection.

Frontier multimodal large language models (MLLMs) are increasingly positioned as general purpose visual reasoners as part of the quest for artificial general intelligence. A key test of this generality is whether they can perform novel visual judgments that humans can make reliably from visual evidence and task instructions, without task-specific parameter optimisation. We investigate this question through the task of medical image alignment assessment, where the goal is to establish whether there is anatomical correspondence between two images. Human visual assessment of image alignment is still the gold standard and most common approach; however, it requires trained operators and is impractical to scale for large datasets. We evaluate recent generations of MLLMs on two exemplar medical image alignment tasks, varying both prompting strategies and image-presentation methods. We compare against a locally fine-tuned MLLM and a task-specific CNN to examine the trade-off between frontier general purpose models and smaller models that require specific task optimisation but can be used locally. We show that are reaching an inflection point, where frontier MLLMs can now perform effective visual assessment of medical image alignment. Models released only a few months ago generalise poorly and, in some settings, perform barely above chance, whereas GPT-6 achieves over 85% across almost all scenarios tested. Fine-tuned local models can match or exceed frontier-model performance on the tasks on which they are trained, but transfer substantially less effectively to unseen settings. These findings identify medical image alignment as a useful test bed for generalist visual reasoning and suggest that frontier multimodal models are beginning to acquire capabilities that could support a common quality-control mechanism across heterogeneous medical-imaging pipelines.

A central question in the prediction of stochastic processes is the extent to which past information can improve the probability of correctly predicting the next state. We introduce the Memory Prediction Excess (MPE) to address this question quantitatively. The MPE measures the average improvement in prediction accuracy obtained by using the entire observed history relative to using only the static marginal distribution, in discrete-time finite-state processes. It is defined as the difference between the expected optimal conditional prediction accuracy and the optimal static prediction accuracy. Its basic properties are examined: the MPE is always non-negative; it admits an upper bound depending on the static accuracy, attained if and only if the future is almost surely a deterministic function of the past; and degenerate cases in which the MPE vanishes are characterized. A normalized version, taking values in the unit interval, is introduced as a dimensionless measure of predictive efficiency. A lower bound is derived by comparing predictions based on histories of different lengths, showing that the expected optimal prediction accuracy is monotone with respect to the history length. The framework is extended to finite-length histories, where the finite-history MPE (FH-MPE) measures the predictive gain attainable when only the most recent observations are retained. This leads to the notion of a minimal memory length required to achieve the same predictive performance as the full history. For finite-order Markov chains, this minimal memory length is shown to be bounded by the Markov order. The MPE and its variants are formulated in terms of conditional probabilities and prediction accuracies, offering a probabilistic perspective on the predictive utility of memory that is complementary to classical information-theoretic approaches.

Discrete motion tokenizers encode motion as atomic units and are widely used for co-speech gesture generation. It remains unclear which motion properties, especially those relevant to gesture semantics, are recoverable from these codebooks. We probe a reconstruction-trained codebook using 19 co-speech gesture descriptors spanning from raw motion to abstract communicative function. Results show that geometry and handedness are readily decodable from token embeddings, while motion category is only weakly decoded despite showing systematic differences in discrete code usage. This gap between reconstruction quality and descriptor decodability suggests that reconstruction objectives alone do not guarantee that gesture semantics are captured, and that evaluating codebooks on such properties can guide the design of more semantic motion tokenizers.

GPU compression needs both dense complete frames and inexpensive checked recovery. We compare three causal representations under the same 512-row model budget, select before encoding one payload, and align entropy states with causal byte banks. Same-lane and residual banks decode concurrently; high-first banks decode in causal order. On ten reused structured/electrophysiology inputs, split-bank execution lowers aggregate median encode-plus-recovery time 13.10% for 0.18% more frame bytes. A separate twelve-configuration GPU comparison places it on the measured median frontier: 5.75% and 7.11% smaller frames than transformed nvCOMP Zstd at 35.14/19.03% lower paired cost for 64/8 KiB requests. A matched native-CPU model control isolates a 44.16% reduction in complete encoding cost; a CUDA-free CPU receiver restores the same frames. Three whole EEG recordings retain a separate Zstd-1–16 curve, while fixed ECG fixtures expose transformed-Zstd counterexamples. Model overhead, concentrated gains, technical ranges and adverse outcomes bound these operating points. ANS interleaving and transforms are established; reused inputs, unequal resources and fixed controls do not establish codec optima or population superiority.

Long-context decoding is increasingly constrained by key–value (KV) cache memory and bandwidth. Existing fixed-budget compression methods typically separate retention from compensation, while a retention ranking specifies neither discarded attention mass nor the direction of induced output error. We start from an exact factorization: eviction error equals evicted attention mass times the directional gap between the evicted centroid and retained output, highlighting the importance of set-level coverage in retention and mass-preserving memory writing. We introduce CORE COverage Calibration and Evicted-Mass REdistribution for KV Cache, which distills an offline allocation combining query utility and log-determinant coverage into a lightweight cache-aware indexer. At inference, one calibrated distribution drives both channels: its Top-$B$ ordering retains complementary KV states, while its excluded allocation mass and conditional weights parameterize latent-memory writes without a separate write-weight predictor or online log-determinant evaluation. Our analysis provides a four-term pre-compensation error certificate, characterizes non-additive coverage interactions, and establishes mass-independent write stability with hierarchical bounds through recurrent and query-adaptive normalization. Across three backbones, CORE exceeds the strongest RULER baseline by up to 3.78 points at 90% compression; LongBench and repeated-eviction evaluations further demonstrate strong effectiveness and decoding efficiency.

Physical AI models such as Vision-Language-Action (VLA) architectures enable generalist robotic policies through large-scale transformer backbones and diffusion-based action decoders. While edge GPU platforms excel at parallelizing the compute-intensive vision-transformer workloads, they exhibit fundamental limitations for the Diffusion Action Expert (DAE) module: the iterative denoising process requires repeated parameter loading from DRAM across multiple steps, resulting in memory-bound performance where the GPU's massive computational throughput remains underutilized. This mismatch between DAE's I/O-intensive characteristics and GPU's compute-centric architecture motivates a heterogeneous acceleration approach. This paper presents EdgeDAE, a heterogeneous FPGA-GPU system that strategically partitions workloads based on computational characteristics. We offload the perception-heavy vision-transformer to GPU while accelerating DAE inference on FPGA through complete on-chip parameter storage in BRAM/URAM. This architecture eliminates the memory bottleneck by co-designing quantization strategies, fixed-point arithmetic, and hardware-efficient random number generation for the FPGA fabric. Compared to an edge GPU baseline, EdgeDAE reduces end-to-end inference latency by 52.5% for Octo-Small and 38.7% for Octo-Base, with up to $2.10\times$ higher throughput; compared to a consumer GPU (RTX~4090), it achieves ${\sim}17\times$ higher energy efficiency.

In the designer-port routing-labeling problem, every vertex of a rooted tree receives a binary label and the child edges receive distinct port numbers. Given only the labels of a source and a destination, a decoder must return the first port on their path. Gawrychowski, Janczewski, and Lopuszanski gave labels of length $\log_2 n+O((\log_2\log_2 n)^2)$, whereas the previous lower bound was $\log_2 n+\Omega(\log_2\log_2 n)$. We prove that every scheme for all $n$-vertex trees needs a label of length $\log_2 n+\Omega((\log_2\log_2 n)^2)$ for every sufficiently large $n$. The result allows arbitrary port assignments and imposes no computational restriction on either the encoder or the decoder. Thus the second-order term in the optimal worst-case label length is determined up to constant factors.

Large language models (LLMs) are increasingly deployed in privacy-critical domains (e.g., healthcare, finance, and government), but their propensity to memorize and disclose personally identifiable information (PII) poses serious security and compliance risks. Existing defenses typically force a trade-off between model utility, privacy protection, and access to fine-tuned private knowledge. We propose LoRA-Oriented Control via Keyed Entry Tokens (Locket), a practical framework that embeds fine-grained, policy-driven access control directly into LLM generation. Locket trains a set of lightweight LoRA (Low-Rank Adaptation) adapters, each encoding a distinct access policy (e.g., full reveal, partial redaction via PII masking, or reveal under a specified differential privacy level). A compact gating module is trained to associate a learned keyed entry token with exactly one LoRA adapter via sequence-level hard routing; the presence of a valid token acts as an authorization key that unlocks corresponding private knowledge, while an invalid or absent token triggers a privacy-preserving adapter that redacts or sanitizes sensitive content. This design ensures Locket remains fully compatible with off-the-shelf LLMs, supporting scalable deployment while satisfying regulatory and privacy requirements. We evaluate Locket across multiple datasets (Enron, ECHR, Yelp) and a diverse set of state-of-the-art LLMs, including Qwen3 (1.7B and 8B), Meta's Llama-3.2 (1B and 3B), and Google's Gemma-2-2B. Our extensive experiments demonstrate that, when the correct token is provided, Locket preserves perplexity comparable to fine-tuning on raw data (without any defense). Conversely, when the token is missing or invalid, it substantially reduces PII leakage while maintaining utility and perplexity on par with strong baseline defenses.

Malicious coordination has long been regarded as a principal source of information ecosystem pollution. Here, we focus on crude, text-repetition-based coordination. As the demand for mitigating its dissemination has grown, scholars have studied such coordination, focusing especially on bot detection. Few studies, however, have characterized malicious coordination per se or examined how it elicits reactions from general users. Leveraging a dataset of 734,173 Chinese-language coordinated accounts and around 495 million coordinated posts published between May 2024 and March 2026, this study analyzes the characteristics of coordinated behavior and how general users react to coordinated posts. We report three findings: (1) most coordinated accounts are crude and retain the classic marks of automation, and the same criterion applied to Japanese-language accounts over the same month yields a share six times lower; (2) their content is overwhelmingly non-political; (3) regarding their reach, most reactions within large observable cascades originate from coordinated accounts themselves, while posts classified as potentially harmful or illegal material receive a comparatively high proportion of reactions from outside the Chinese-dominant population. We provide a longitudinal quantitative map of crude Chinese-language coordination appearing in X's Japanese-classified stream.

Crowdsourced imagery provides timely, fine-grained, street-level observations for disaster mapping, complementing conventional remote sensing imagery (RSI) during emergency response. However, such imagery is often unstructured, spatially ambiguous, and lacks reliable geographic metadata, making manual geolocalization and interpretation labor-intensive and difficult to scale. This work proposes a multi-task Geospatial Reasoning Disaster mapping framework, namely GRDisaster, to examine the potential of vision-language models (VLMs) in understanding, geolocalizing, and reasoning over crowdsourced disaster imagery. GRDisaster is built on a newly curated benchmark dataset derived from PhotoMappers, comprising 26,340 images organized into human-validated volunteered geographic information (VGI), street-view imagery (SVI), RSI cross-view triplets covering multiple disaster events from 2018 to 2024. The framework combines deterministic and probabilistic cross-view geolocalization with multi-view fusion to associate VGI images with georeferenced SVI and RSI. It introduces two sets of spatial reasoning indicators for cross-view geolocalization validation and disaster damage assessment. These indicators use structural, environmental, and global-scene cues to validate cross-view correspondences and visually observable damage evidence with expert-verified annotations to assess disaster severity, improving the interpretability of VLM outputs. To our knowledge, this study provides the first systematic investigation and unified evaluation framework for examining how VLM-based spatial reasoning can transform crowdsourced disaster imagery into actionable geospatial artificial intelligence (GeoAI) through cross-view geolocalization validation, interpretable spatial reasoning, and damage-aware severity assessment.

Verifying large C codebases requires translating them into formally checkable semantics, but most autoformalization work reports a single aggregate success rate that conflates two different failure modes: code the translator declined to handle and code the translator translated incorrectly. We present a deterministic code-property-graph (CPG)-based exporter that translates C source into a small Lean~4 core semantics under a three-tier correctness discipline: hole-free, call-closed, and dynamic-hole-risk, that keeps these modes distinct. Applied to a re-export of the SQLite source tree (8{,}602 functions, over 1.5~million AST nodes), the exporter translates 5{,}222 functions (60.7%) hole-free, of which only 2{,}134 (24.8%) are call-closed, a $2.4\times$ gap a single rate would hide. Both figures are reported in a full per-construct trust ledger rather than a single score. Our main contribution is methodological: a protocol for separating constructs that are fundamentally unresolvable by whole-program static analysis (such as public API boundaries) from constructs that only look that way. For example, we retracted our own “impossible” classification of function-pointer/vtable dispatch after tracing a concrete counterexample in the target codebase. Separately, a search for “more holes closed” surfaced a latent silent-wrong-answer bug: a translation that succeeded with an incorrect result rather than declining, a failure mode we argue is more dangerous than any hole, and one a hole-count-only evaluation would never surface.

Kubernetes deployments increasingly span multiple clusters for reasons of scale, fault isolation, regulatory boundaries, and heterogeneous hardware placement, including GPU-dense clusters dedicated to AI and HPC workloads. Cilium Cluster Mesh, built on extended Berkeley Packet Filter (eBPF) technology, has emerged as a widely deployed mechanism for connecting such clusters into a single logical network without a dedicated multi-cluster gateway. Despite its adoption, the academic literature contains no systematic, reproducible evaluation of the eBPF-based multi-cluster data plane itself: existing work either benchmarks eBPF against iptables within a single cluster, or applies eBPF to cross-cluster monitoring rather than forwarding and identity propagation. The most detailed scale data available - a report of a clustermesh-apiserver failing under load at approximately 45,000 nodes across 256 clusters - comes from an industry engineering blog rather than a peer-reviewed source. This paper proposes a methodology to close that gap. We define five research questions covering latency, throughput, control-plane limits, churn, and GPU/RDMA sensitivity, and describe a testbed and eBPF-level instrumentation design. Lacking a live multi-cluster testbed for this submission, we report results from a calibrated analytical simulation of control-plane architectures fit to the production data point. These results are presented as falsifiable predictions intended for validation once the testbed is complete. This paper provides the motivation, experimental design, and preliminary simulation results to serve as a template for a full empirical study.

Streaming spoken agents may produce the correct final action after acting too early. Final-turn scores do not reveal whether each observed speech prefix supports an exposed action. We introduce the Partial Speech Action Contract for Turn Taking in Speech Language Models (PACT-SLM), a controlled test that assigns the first valid action time and evaluates both action identity and timing. In the primary test, 80 paired contrast groups from four held-out semantic families yield 1,600 prefix predictions across clean and 15 dB noise renderings. Using source-utterance semantic targets rather than counterbalanced branch codes, WavLM Base Plus reaches 26.03% pooled post-onset semantic-label accuracy (95% group-bootstrap interval [22.14%, 29.68%]), exposes an action on 18.99% of pre-onset prefixes, and predicts 5.94% of complete trajectories exactly. Its pooled label score is at the 96th percentile of 100 within-prefix label permutations, below the 97.5th-percentile reference (26.73%). It exceeds matched text, scalar-acoustic, and shuffled-representation probes in pooled post-onset label accuracy. Elapsed time is more onset-exact (36.25% versus 23.13%) but less accurate about action identity (9.92% versus 26.03%). These results motivate separate measurement of action identity and onset timing in partial-speech evaluations.

We introduce the two-echelon covering tour vehicle routing problem (2E-CTVRP) for the distribution of relief supplies after a disaster. In the first echelon, a fleet of trucks transports supplies and drones from a central depot to satellites at the periphery of the affected area. In the second echelon, drones launched in parallel from the satellites deliver the supplies to the centroids of victim clusters, which are obtained by clustering the victim locations, and each truck waits at a satellite until its drones have returned. The problem combines the assignment of satellites to trucks, the sequencing of the truck routes, and the assignment of clusters to satellites, and minimizes the sum of the arrival times of the trucks at the satellites and at the depot. We formulate the 2E-CTVRP as a mixed integer linear program and propose a hybrid metaheuristic, GRASP-ILS-PR, which combines greedy randomized construction, a single-trajectory search, and periodic path relinking on the assignment of clusters to satellites. On a new benchmark set of 110 instances, GRASP-ILS-PR finds all known optimal solutions within 30 seconds, matches or improves on the one-hour solutions of a commercial solver on most larger instances, and improves on them by up to 11% for instances with 50 satellites. It clearly outperforms a conventional GRASP with evolutionary path relinking and yields more consistent results than a simulated annealing algorithm that uses the same search components. An analysis of two fleet configurations shows that few large trucks minimize the cumulative arrival time in most instances, whereas many small trucks deliver supplies to the victims earlier and more equitably in every instance.

Imitation learning enables robots to acquire complex skills directly from massive demonstration datasets, but its performance degrades severely when datasets are contaminated with suboptimal or noisy demonstrations. While prior quality-assessment methods attempt to filter or reweight data, they typically rely on manual pre-selection of expert reference data or task-specific heuristics, limiting scalability. To address this challenge, we introduce SynIL (Synergy-based Imitation Learning), a novel framework for automated, label-free demonstration quality assessment in offline reinforcement learning. Grounded in neuroscientific evidence that motor synergy, a low-dimensional coordinated structure in movement, correlates directly with motor proficiency, SynIL algorithmically quantifies synergy manifestation to generate dense, transition-level reward signals via self-supervised reward regression. Comprehensive evaluations on D4RL locomotion benchmarks and multi-human Robomimic manipulation datasets demonstrate that synergy-derived rewards correlate strongly with ground-truth rewards. Furthermore, SynIL substantially outperforms Behavior Cloning (BC) and achieves performance comparable to, and in sparse-reward human teleoperation scenarios, superior to, offline reinforcement learning trained on true environment rewards.

Large language model agents are increasingly granted real privileges (executing commands, modifying files, calling APIs), so an agent that errs has already acted. Existing defences concentrate on the agent's inputs, while the path from a candidate action to privileged side effects remains less directly studied. We argue that this path must be governed outside the model, and study an architecture interposing four roles (planner, policy gate, executor, auditor) between agent and operating system. Two choices are central: actions arrive as structured intents, so adjudication never parses shell syntax; and approval is a one-shot credential bound to the exact bytes that will run. We evaluate on a 313-case benchmark across eight variants, with prompt-only baselines from three hosted LLMs on 150 stratified cases, five repetitions (2,250 attempted calls; 2,249 completed). Effective attack success falls from 98.3% under direct execution to 7.7% deployed. Re-execution against the real implementation yields a similar aggregate rate (7.6% over 66 sandbox-evaluable payloads) but substantial case-level disagreement, at a corrected false-denial rate of 11.1%. A substantial final-stage reduction (from 30.8% to 7.7%) is attributable to the operating-system sandbox, and the benchmark found four implementation defects, none by design review.