Papers

Nanodrones require accurate, real-time state estimation under severe sensing and computational constraints. We present TinyCVIO, a visual-inertial odometry system that co-designs miniature sensing, visual processing, and estimation for a commodity dual-core microcontroller with 520 kB SRAM. Lightweight LED constellations provide known geometry without surveyed positions or yaw angles, assuming placement on a common level plane. A streaming visual frontend tracks LED observations from a millimeter-scale camera at 29.2 FPS, while a rigid-board measurement model retains inter-LED constraints and streaming QR bounds estimation workspace for a fixed filter-state size. Across 19 hand-held hardware-in-the-loop datasets, the rigid-board model reduces mean absolute trajectory error by 27% relative to planar points. The complete system runs onboard a Crazyflie across nine flights at three speeds, achieving 3.5-3.7 cm mean absolute trajectory error and 0.50-0.60% relative pose error over 10 m segments, with mean estimate latency of 15.7-16.3 ms.

Cities differ in built form, land cover and development history, complicating comparison across places and time. Satellite foundation models map Earth's surface onto common numerical representations. Yet the tasks and targets used to shape them typically do not focus on cities: globally consistent labels for urban function do not exist, and many datasets - especially land cover and land use classifications - collapse the built environment into few classes. Here we audit the representation, focusing on AlphaEarth but with broader applicability to other Earth embeddings, by probing the geometry and geography of embeddings for 1,000 urban areas in 162 countries. We find that cities occupy a shifted but overlapping region on the hypersphere, 62.7{\deg} from the global mean direction, and continent and climate predict 24.3% of variation among the mean directions of urban centres in excluded countries. Inside cities, degrees of urbanisation carry 8.9% of the variation, and what they leave holds shared directions whose local orientation varies, not one universal axis of urbanisation. Retained variation is itself unequal: dispersion within urban centres is 14.1% greater per standard deviation of national development, even after adjusting for population, land area and continent. Further controls suggest cities in developing countries present less contrast in vegetation and texture, and dispersion follows that contrast: full adjustment for it leaves at most 6.4% of the gradient. Annually, a city's representation moves nearly eight times more than redrawing its own pixels explains, and contracts where the 2022 loss of Sentinel-1B removed a pass direction. AlphaEarth's representations therefore support comparison across regions, while the differences between its annual layers are not yet validated for comparison over time.

We give a polynomial-time algorithm that constructs, in every connected $n$-vertex graph of minimum degree at least $7$, a spanning tree with at least $\frac{25200}{46189}n>0.5455\,n$ leaves. No bound specific to minimum degree $7$ was known: the best bound available for this class was $\frac{11}{21}n\approx0.5238\,n$, inherited from Simarova's theorem for minimum degree~$6$. The algorithm and its analysis are carried out for an arbitrary minimum degree $\delta$, and yield a recursion that gives an explicit lower bound on the number of leaves for every $\delta$. The resulting bounds improve all previously known ones for every $\delta\ge7$; for $\delta=8,9,10$ they are $0.5850\,n$, $0.6151\,n$ and $0.6413\,n$, and they are tabulated for $\delta\le25$ at the end of the paper.

This volume contains the proceedings of ICE'26, the 19th Interaction and Concurrency Experience, which was held in Urbino, Italy, as a satellite event of DisCoTec'26. The ICE workshop series features a distinguishing review and selection procedure: PC members are encouraged to interact, anonymously, with authors. The 2026 edition of ICE received 8 submissions, each reviewed by three PC members, and about 60 comments were exchanged during the review process, witnessing very lively discussions. Five papers were accepted for publication plus two oral communications, which were accepted for presentation at the workshop. We were proud to host one invited talk, by Emilio Tuosto. The abstract of his talk is included in this volume, together with the final versions of the research papers, which take into account the discussion at the workshop and during the review process.

Self-consistency has become a popular technique for enhancing the reasoning abilities of large language models by generating multiple reasoning paths and selecting the final answer through a majority vote. However, because model providers typically charge users in proportion to the number of reasoning paths generated, they have a financial incentive to artificially increase the path count. In this work, we show that an unfaithful provider can exploit this incentive using a simple, efficient algorithm while avoiding detection by an auditor: by generating and strategically reordering additional reasoning paths, the algorithm makes every path appear necessary to reach the majority. To validate our algorithm, we conduct experiments with multiple instruct models from the Llama and Qwen families, as well as reasoning models distilled from DeepSeek-R1, on benchmark datasets spanning mathematics, science, and question answering. Our results suggest that the distribution of additional reasoning paths generated by our algorithm is heavy-tailed and that substantial capacity to overcharge remains even under the best possible audit designed to keep the false-positive rate below $\alpha = 0.1$.

Context: Code recommender systems (CRSs) are specialized software systems operating on source artifacts to provide automatically generated recommendations to software developers in all phases of development. The goal is to improve software quality while enhancing developers' efficiency, effectiveness, and experience. Problem: Despite the growing importance, little is known about the state of evaluating these systems in a human-centric manner. Research Approach: We conducted a systematic literature review to identify and to synthesize primary studies evaluating CRSs (2017-2024). Ninety-two publications were included and subjected to a systematic content analysis. Results: Our study confirms that Offline Evaluations are the most common, system-centric evaluation type, whereas user-centric evaluations (Online Evaluations, User Studies) are rarely reported. Evaluations are concentrated on the Software Construction phase. Most studies explicitly report threats to validity, with External threats to study Materials being the most common. Conclusion: The state of evaluations on CRSs is narrowly focused on single or a few system qualities. There is an imbalance between a majority of system-centric evaluations and a minority of user-centric evaluations. Combined, multi-type evaluations are rarely reported.

The code quality in the game domain is perceived as less structured and more convoluted than in other domains, which results in lower maintainability and higher bug frequency. In this work, we conduct a systematic literature review to address four main research questions: (1) how the incidence of code smells and technical debt differs in game development compared to other software domains; (2) what academic solutions are proposed to mitigate these code quality issues; (3) whether researchers collaborate effectively with developers to implement these solutions; and (4) if the proposed solutions are sufficient to tackle the diagnosed problems. Thirty-four works were mapped and analyzed. We found that the game domain is affected by a short-term mentality, continuous requirements changes, reduced code reuse, and minimal to no automated testing. As potential solutions, we identified tools and models to detect code smells, along with testing practices and product lines adapted for game development. Although these solutions exist, they are not widely adopted by practitioners, highlighting the need to investigate barriers to their integration.

Gaussian processes constitute a cornerstone of probabilistic machine learning, yet scaling them to large datasets typically forces a trade-off between computational efficiency and model fidelity. This work bridges this gap by presenting an adaptive multi-resolution Gaussian process framework that is both scalable and exact. Our key innovation is constructing a naturally data-sparse covariance matrix with adaptive multi-resolution basis functions. These basis functions are directly anchored to samples, eliminating the need for auxiliary points. By shrinking the support domains of multi-resolution basis, the matrix block sizes are limited, guaranteeing sparsity. The inverse of the data-sparse covariance matrix is computed exactly and efficiently via the sparse Cholesky inverse algorithm. To further improve predictive uncertainties, we construct an augmented basis function. Theoretical analysis and numerical experiments demonstrate that our model achieves exact inference with $\mathcal{O}(n \log^2 n)$ training cost and $\mathcal{O}(\log^d n)$ prediction cost, establishing a principled framework for scalable and high-fidelity Gaussian process regression.

Cloud-security investigation is dominated by population tasks: which identities can read a data store, how many resources fail a control, what is reachable from another account. These resolve against a complete inventory, not a named object, so a partial answer to one is not a partial result but a different one. Coding agents can now be given read-only cloud credentials and asked to investigate directly, which raises the question of what a purpose-built security context layer still contributes. We evaluate the Sola Security Brain, a security intelligence layer whose relational substrate is resolved offline and whose security logic is evaluated against it at query time, against two coding agents operating the same live AWS environment through a read-only CLI, Claude Code and OpenAI Codex, over 28 investigation tasks. All three are scored by a blinded, tier-weighted, grounding-gated relative recall over one joint claim pool, so the scores share a denominator. The Sola Security Brain reaches 0.549 +/- 0.012 coverage against 0.340 +/- 0.011 for Claude Code and 0.281 +/- 0.006 for Codex, with the ordering identical in every grading draw. It leads 24 of 28 tasks from the cheapest model tier, at 17.8x and 20.8x lower cost per task than Claude Code and Codex. Beyond the aggregate, we describe an answer-level pattern we term sample-and-generalise: both agents enumerate a fraction of a large population, assert an unhedged universal negative, and disclose the sample size only in answer metadata rather than in the answer. In one task Claude Code reported that no bucket policies exist after checking four bucket families, in a sweep that sampled 40 of roughly 5,000 buckets, in an account where 65 buckets carry a wildcard-principal read grant.

Modern sensing records the motion of physical systems, but often leaves the forces and mechanical response governing that motion unobserved. Inferring these quantities from discretely sampled trajectories is especially difficult at coarse time scales, when mechanical response evolves between observations and interactions propagate across the system. Here we introduce Newmark-e̱ta}-DGN, a graph neural network-based framework that combines two structures inspired by computational mechanics. First, a semi-implicit update inspired by the Newmark-e̱ta} method uses learned momentum fluxes and matrix-valued response operators to advance the state over each observed interval. Second, an operator-weighted virtual hub provides system-wide coupling through a sparse set of connections. The learned quantities thus determine the predicted motion and remain accessible for mechanical analysis. Across a deformable beam, human motion and protein dynamics, Newmark-e̱ta}-DGN supports long-horizon prediction at time steps for which explicit learned simulators deteriorate. Without force, moment or constitutive relation supervision, forces inferred from walking kinematics track independently derived hip and knee joint moments, while response operators learned on the beam recover the relative spatial and directional structure of its finite-element stiffness tangent. Newmark-e̱ta}-DGN therefore links coarse-step prediction to the inference of mechanical quantities that were never observed during training.

iKFAD is a recently proposed optimiser that replaces adaptive learning rates with adaptive friction in the momentum dynamics, yet performs as well as Adam. Its limitation is that the full friction tensor $\xi\in\mathbb{R}^{m\times n}$ carries the same $\mathcal{O}(mn)$ memory overhead per layer as Adam's second-moment buffer. Here we replace iKFAD's friction tensor $\xi$ with a rank-1 outer-product factorisation built from row and column momentum statistics, resulting in Rank-1 iKFAD (R-iKFAD). This reduces the friction memory footprint from $\mathcal{O}(mn)$ to $\mathcal{O}(m+n)$ per layer, which approximately halves iKFAD's total optimiser state. Despite this reduction, R-iKFAD maintains parity in performance with iKFAD: experiments on GPT2-Nano, TinyViT, DistilBERT and GPT2-S confirm that it matches or exceeds iKFAD while nearly halving the memory footprint and remaining comparably robust to hyperparameters. We analyse the continuous-time dynamics in two damping regimes. For linear damping ($\gamma>0$) we prove exponential convergence under strong convexity. For $\gamma=0$, the preferred option in our experiments, the friction is generated entirely from past momentum and switches off as the momentum vanishes, so geometric convergence cannot be shown. We nonetheless prove convergence to the minimiser, together with matching upper and lower bounds on the energy: of order $t^{-1}$ when the regularisation scale $\epsilon_{\mathrm{stab}}$ is zero, and of order $t^{-1/2}$ when it is positive. To our knowledge this is the first convergence rate for a rank-1 factored optimiser in continuous time, and the first such result that does not require positive damping.

Data Spaces enable sovereign and governed data sharing across organizational boundaries, but their integration with AI agents remains challenging due to mismatches between probabilistic language model interactions and policy-driven data infrastructures. This article presents an architectural mediation approach based on the Model Context Protocol (MCP), implemented through the Eunomia Agent, to enable controlled interaction between large language model (LLM) agents and data space services. The proposed mediation layer translates data space capabilities into structured, schema-driven tools that AI agents can discover and invoke while preserving governance constraints. A prototype implementation validates end-to-end interaction across catalog discovery, metadata retrieval, and data service invocation without modifying existing data space components. Results demonstrate that protocol-based mediation enables interoperable and standards-aligned integration of AI agents into data space ecosystems. The approach provides practical guidance for organizations seeking to introduce AI-driven automation into governed data-sharing environments while maintaining compliance, interoperability, and architectural separation of concerns.

Air-quality sensor outages often create contiguous missing blocks, where side information useful for isolated missingness may be less reliable. We study a block-specific, GAUDI-aligned conditional diffusion imputer that retains temporal and feature processing, visible-value and mask conditioning, variable identity, and diffusion-step information, while suppressing absolute time-position side embeddings. On ItalyAir (13 variables, length-32 windows, nominal 50% block missingness; three archived seeds), this feature-side configuration achieves RMSE 0.340, versus 0.355 for full context and 0.355 for local CSDI. The experiment isolates a geometry-aware conditioning effect under block missingness.

We investigate the least squares linear regression problem with random partial Discrete Fourier Transform (DFT) matrices, providing a rigorous analysis of the model's generalization error. By leveraging tools from random matrix theory, we derive exact non-asymptotic bounds for the risk of the Moore-Penrose estimator, which hold for finite-dimensional problems and reveal the precise dependence on key parameters such as the sample size, dimension, and noise variance. Then we obtain a characterization of the double descent phenomenon in the linear regression context, demonstrating how the risk evolves when the number of parameters $p$ and the number of samples $n$ tend to infinity, with $p/n$ fixed. The analysis relies on applications of the Stieltjes transform for random Fourier matrices, enabling a precise description of the spectral properties of these matrices and their impact on regression performance. To validate our theoretical findings, we present several numerical examples that illustrate the double descent curves. These simulations align closely with our derived bounds, confirming their predictive power in both under-parameterized and over-parameterized regimes.

Retrieval-Augmented Generation (RAG) mitigates knowledge obsolescence and factual hallucination in large language models by introducing external context. However, when retrieved knowledge conflicts with the model's internal parametric knowledge, the model may either blindly follow misleading context or incorrectly rely on parametric knowledge, leading to unreliable responses. To address this issue, this paper proposes TRACE (Debate-TRace and Answer-Completeness rEgularized fine-tuning), a robust fine-tuning framework for RAG under knowledge conflicts. First, we propose a fine-tuning method that leverages multi-agent debate traces to extract correct candidates, incorrect candidates, and answer-shift patterns, providing fine-grained supervision for reliable knowledge-source selection. In addition, we design an answer completeness regularization mechanism to alleviate empty, overly short, and prematurely terminated responses via answer-tail token reinforcement and premature termination suppression. The fine-tuning objective combines correct-answer supervision, incorrect-candidate suppression, answer-tail token reinforcement, and premature termination suppression, enabling the model to use reliable external context, resist misleading or irrelevant retrieved content, and fall back to parametric knowledge when retrieved evidence is unreliable. Experiments across multiple knowledge-conflict scenarios and datasets show that TRACE improves robustness against misleading retrieved knowledge and reduces incomplete answers. These results demonstrate that multi-agent debate traces and answer completeness regularization jointly enhance knowledge-source selection, conflict robustness, and answer quality in RAG models. Our code is available at https://github.com/PHD-lanyu/TRACE.

AI development agents are increasingly used to support and partially automate software architecture tasks. To explore how practitioners perceive this shift, specifically what changes, what remains, and what new responsibilities emerge, we conducted a focus group at the 31st European Conference on Pattern Languages of Programs, People, and Practices (EuroPLoP 2026). Twenty-two participants from industry and academia discussed current practices, trust and validation strategies, the boundaries of AI autonomy, governance challenges, and implications for education. Among others, we found broad consensus that architectural decision-making, accountability, and the authoring of architectural guardrails remain fundamentally human tasks. A central emergent concept was harness engineering: the discipline of building the system that governs AI-assisted system creation, comprising validation mechanisms, knowledge lay- ers, and company-specific standards. The participants agreed that criticality, understood as the combination of uncertainty and cost of change, serves as the universal criterion for calibrating human oversight. A further concern was cognitive debt: the progressive erosion of human understanding of the system when AI-assisted decisions are accepted without full intellectual engagement. In this report, we present the findings of the focus group and describe directions for future work.

Asynchronous monitoring, incident investigations, and compliance audits primarily rely on agent traces to reconstruct what happened. These analyses assume that LLM agents cannot tamper with their own execution traces. We show that local LLM agents such as Claude Code, Codex, Antigravity, Open Code and Grok Build fail to enforce this boundary. All tested harnesses, except Muse Code, allowed agents to delete their traces when asked, without triggering monitor guardrails. We also validate that external attackers can exploit this gap to induce trace deletion. Finally, we show that trace tampering behavior emerges naturally in frontier models, when agents try to improve their rewards. We advise practitioners to ensure trace logging happens through an independent interception mechanism outside of the agent's control, preserving trace integrity even in cases of full host compromise. Overall, our findings identify a concrete failure of trace integrity in agent infrastructure which can be used to conceal misaligned behaviors like scheming or sabotage.

Latent world models are typically trained to predict factual transitions, whereas model predictive control (MPC) must compare alternative actions from the same state. A model can therefore achieve low factual prediction error yet poorly distinguish candidate actions. We introduce AD-WM, an action-discriminative joint-embedding world model for counterfactual MPC. AD-WM combines residual latent dynamics with predictor-level action-recovery regularization, using inverse dynamics and a normalized recovery objective motivated by conditional mutual information. Both objectives encourage planning transitions to preserve action information; their auxiliary heads are discarded at test time, leaving MPC unchanged. On OGBench-Cube, AD-WM improves hard-start success from 3.7% to 52.0% over a matched LeWM baseline and improves mean success over the reproduced baseline in four of five simulation environments. Planning diagnostics show that factual prediction error and whole-bank action ranking do not follow the closed-loop success ordering, whereas CEM-aligned elite regret tracks success more closely. With a frozen V-JEPA 2 encoder and matched DROID post-training, AD-WM also improves zero-shot transfer to our Franka setup, increasing basic pick-and-place success from 42.2% to 71.1% without lab-specific adaptation. These results suggest that world models for planning should preserve action-dependent differences needed for counterfactual selection, rather than optimize factual prediction accuracy alone. More videos and code are available at https://ad-wm.github.io/.

Distributed learning in embodied reinforcement-learning agents offers a degree of privacy by retaining raw sensor data on-device and transmitting only policy gradients to the server. Yet temporal structure can amplify this leakage beyond single-frame attacks. We introduce Temporal Reconstruction Attack on Consecutive Encodings (TRACE), an amortized temporal gradient-inversion attack that autoregressively reconstructs the sequence of private observation-action trajectories from per-step policy-learning gradients. The attack exploits two structural signals ignored by prior single-frame methods: (i) cross-time correlation between successive embodied gradients, which we formalize via a conditional mutual-information bound, and (ii) closed-form action recovery from policy-head gradient structure, which we prove exact when standard entropy regularization is sufficiently small. On held-out embodied scenes, TRACE reaches $18.8$ dB PSNR with near-perfect action recovery at $3$-$4.5$ ms per reconstructed frame, dominating the learning-based baseline across all reconstruction metrics and exceeding optimization attacks while running orders of magnitude faster. Further evaluation demonstrates TRACE's broader applicability across recurrent, residual, and compact transformer victim architectures, multi-modal inputs, and larger discrete action spaces. Defense experiments suggest that protecting temporal gradient streams may require sequence-aware privacy mechanisms.

We introduce Forte, a sensitivity type system for Rust whose soundness rests on ownership. The graded sensitivity type systems, from Fuzz's linear grading to Solo's environment indices, are pure calculi: a claim about a value holds for the value's whole lifetime because nothing can mutate it. The imperative sensitivity analyses admit assignment to first-order variables and no references, so no question of aliasing arises in them. The programs that compute differentially private statistics in deployment are Rust, and they mutate through borrows. Forte closes this gap. Its central rules strongly update a sensitivity environment through an exclusive borrow, at a primitive call and across a checked function boundary; its soundness theorem is metric preservation over an operational semantics with a store, in which the exclusivity of &mut alone licenses framing across a mutating call, and two aliased borrows suffice to refute the theorem without it. Verus mechanizes the theorem, the function rule, and the refutation. Flux checks Forte as an ordinary library, with no fork of the compiler; a machine-checked theorem backs every deterministic primitive signature, and a correspondence theorem transports metric preservation to the programs the checker accepts. We evaluate Forte on mechanism kernels from OpenDP with genuine in-place mutation, matching the library's trusted stability maps with checked constants, covering the constructors that have no proof document, rejecting off-by-one diameters, tightened bounds, miscalibrated releases, and overspent budgets, and deriving one trusted constant as an inferred loop invariant.

Conspiratorial discourse on social media is not always expressed through explicit claims or stable lexical markers. The same surface content may express endorsement, legitimate concerns, criticism, satire, or mockery. The main challenge is therefore not only recognizing conspiracy-related claims, but inferring the speaker's intent – the utterance's illocutionary force. We argue that this can be achieved through the use of relevant social contexts and propose an agentic framework, equipped with a set of tools supporting social queries. We demonstrate the benefits of our approach on a unique dataset of Hebrew tweets, covering 80%–90% of the public Hebrew tweets published over a four-year span (late 2018– early 2023), encompassing several election cycles as well as the COVID pandemic years and related vaccination campaigns. This extensive coverage can be used in recovering different social contexts. Evaluating our framework on a manually-annotated adversarial dataset, we find that context-aware workflows consistently outperform text-only classification and that the agentic framework performs better than other frameworks and settings, with significant gains over a non-agentic model exposed to the same contexts available to the agent. We further provide an analysis of the results, the errors and efficiency (token economy) tradeoffs. These findings support viewing the task of conspiracy detection as a socially embedded interpretation task, in which effective classification depends not only on access to contexts, but also on adaptive reasoning in which the agent uses tools on a per-case basis, asking only for evidence relevant to its current reasoning step.

Coding agents have demonstrated enormous success in solving complex programming problems. To leverage their potential for robot systems, this work introduces Robot Agentic Programming from Demonstrations (RAPID), which automatically generates, verifies, and refines robot programs, given a single visual human demonstration. The iterative agentic loop of code refinement requires several key ingredients: (i) a testable task specification, (ii) action primitives for robot execution, and (iii) an interactive environment for program execution and verification. RAPID infers all three from the demonstration automatically. To make the resulting program reusable beyond the demonstration setting, RAPID uses an object-centric relational program representation that focuses on the underlying structure of the demonstrated strategy rather than the specific motion per se: it expresses the action primitives as trajectory-optimization programs that realize object-level motion effects, while composing them through relational constraints that capture scene-specific geometry at run time. We evaluated RAPID in simulation on eight challenging contact-rich nonprehensile manipulation tasks as well as general prehensile manipulation tasks in the LIBERO-Pro benchmark. We also successfully deployed it on a real Franka arm and evaluated on all eight nonprehensile tasks. In all experiments, RAPID demonstrated strong performance, with generalization over object pose, shape, material, and environment. Website: https://yuyaoliu.me/projects/rapid.

World Action Models (WAMs) couple action generation with future visual prediction for robotic manipulation. However, completing the joint video-action denoising process at each replanning cycle incurs substantial latency, delaying action updates and limiting closed-loop responsiveness. We present Rolling-WAM, a formulation that distributes joint denoising across successive replanning cycles. Our method maintains a sliding window of video-action chunks at staggered noise levels. At each step, a rolling noise schedule fully denoises the imminent action chunk for execution, while partially refining farther-future chunks. As the window advances with new camera observations, the retained future chunks continue their denoising process. This distributes the computational cost over time while carrying an evolving visual-action context across chunk boundaries. Evaluations on LIBERO, RoboTwin, and a real-world Unitree G1 humanoid show that Rolling-WAM achieves competitive manipulation performance. By removing the need to denoise the entire prediction horizon from scratch, it delivers a 4.5x steady-state replanning speedup over standard joint WAMs.

3D Gaussian Splatting (3DGS) enables high-quality novel-view synthesis but requires substantial storage. Existing compression methods often rely on spatial context modeling over irregular 3D representations, increasing the complexity of training and coding. Meanwhile, floating-point context inference can introduce numerical inconsistencies across platforms, causing entropy-decoding failures. To address these practical challenges, we propose COSA-GS, which constructs context without spatial aggregation through anchor-wise causal factorization. Specifically, we use geometry context derived from each anchor's coordinates to model a compact learnable anchor latent. The anchor latent is then fused with the geometry context to form an anchor context for attribute coding. The resulting context model features a simple architecture composed solely of linear transformations and activations. We train COSA-GS using rate–distortion optimization with adaptive Gaussian pruning. Further, we develop quantization-aware training and integer inference for the context model to achieve bit-exact consistency of entropy-decoded symbols across platforms. Experiments demonstrate that COSA-GS achieves state-of-the-art compression performance while retaining fast and consistent cross-platform decoding, providing a simple yet effective framework for practical 3DGS compression. Code is available at https://github.com/pengpeng-yu/COSA-GS.

An ordinary-looking background detail can turn a correct model decision into a confident mistake. We demonstrate this fragility in four decision systems, including Jev, across seven datasets covering knowledge, reasoning, and tool routing. Within 64 accepted target evaluations per decision, we uncover short context additions that redirect 61.4%-73.2% of each system's initially correct decisions toward a wrong option fixed in advance. The additions supply background or procedural information rather than explicit answer-selection instructions, leaving the original question and choices intact. We construct them through probability-guided context optimization, which uses shifts in the option distribution to refine surrounding text under naturalness and answer-preservation constraints. Redirection affects initially confident decisions, often produces high-confidence wrong choices, and transfers across models. In blinded human evaluation, 91.6% of 250 sampled successful contexts are judged natural, answer-preserving, and free of decisive answer-changing evidence by a majority of three independent annotators. These findings expose a weakness in current decision models: context that looks entirely compatible with an input can redirect the choices that agents, routers, and evaluators rely on.

Recent research on Multimodal Sentiment Analysis (MSA) has focused on learning from language, visual, and acoustic modalities with incomplete data to infer human sentiment. Most studies typically compensate for missing information by reconstructing modality features or designing complicated fusion mechanisms. However, these methods still suffer from spurious generation and noisy guidance due to the lack of high-level semantic grounding in partially observed multimodal evidence. To address these issues, we propose SemMSA, a latent semantic-aided framework that constructs rich sentiment-relevant semantics with LLMs, fully integrating with all modalities via anchor-free spectral alignment. It mainly consists of Cross-modal Semantic Refinement (CSR) and Cross-modal Spectral Alignment (CSA). Specifically, CSR first adaptively extracts visual and acoustic representations by corresponding adapters to form a unified multimodal prefix with language in the frozen LLM embedding space. It then iteratively produces continuous discriminative semantic states through a token-efficient latent refinement process without decoding explicit text. Next, CSA simultaneously aligns the refined semantics with all modalities by enhancing the dominant spectral component of their kernel Gram matrix. This captures global nonlinear dependencies among all representations without relying on a predefined anchor modality. In addition, an instance-level spectral separation constraint preserves cross-sample discriminability and mitigates representation collapse. Extensive experiments on SIMS, MOSI, and MOSEI benchmarks demonstrate that SemMSA achieves state-of-the-art performance.

Automated generation of production-ready 3D garment assets from a single image is a central challenge in digital content creation. While recent generative models have significantly advanced 3D geometry reconstruction, synthesizing high-quality textures remains a bottleneck. Existing methods often bake environmental illumination and shadows directly into the texture map, or they fail to maintain global structural coherence, making the resulting assets unusable for physical simulation and relighting. In this work, we introduce OmniFabric, a novel approach that synthesizes globally coherent texture maps directly within the 2D sewing pattern space. Given a single reference image, our pipeline utilizes an estimated 3D mesh and generative priors of powerful Vision-Language Models (VLM) to establish a complete but coarse texture initialization across the unwrapped sewing patterns. We then leverage a specialized diffusion transformer, trained via an automated synthetic data engine and conditioned on 3D positional features, to refine this initialization directly in the canonical UV domain. This effectively removes distortion and baked-in artifacts to extract a clean and normalized texture map that preserves the original garment design. Extensive experiments demonstrate that OmniFabric significantly outperforms state-of-the-art baselines, yielding photorealistic 3D garments with high-quality textures.

Task and motion planning (TAMP) problems remain difficult even with full observability and object-centric states because discrete decisions are tightly coupled to geometric, kinematic, and dynamic constraints. Generalized TAMP addresses this difficulty by exploiting regularities across problem instances to reduce planning effort on new instances. However, existing methods require substantial TAMP-specific engineering. We investigate whether coding agents can automate this process by synthesizing programs that generalize across instances. Given a task description and simulator access, each agent chooses how to interact with the environment while developing a program within a fixed synthesis budget. The program is then frozen and evaluated on unseen instances. We evaluate Claude Code (Opus 5) and Codex (GPT-5.6 Sol and GPT-6 Astra) on 28 simulated environments from KinDER and PDDLStream, with object counts beyond those evaluated in the original benchmark. Across all program synthesis methods, we evaluate 980 generated programs on 100 held-out instances each, 98,000 evaluation episodes in total. Overall, we find that coding agents are surprisingly effective at generalized TAMP: all three agent configurations outperform hand-engineered planners, one-shot generation, and an LLM-based generalized planning baseline in mean success (56% to 95% versus 47% for the planners, on the 16 environments where a planner is available). As object counts grow, the agents' programs maintain higher success than the planner, using an order of magnitude less computation per instance on average. Logs show agents using interaction to calibrate physical models, test edge cases, and refine strategies. We release all code, including the full prompts given to the agents. These findings suggest that coding agents are a strong baseline for generalized TAMP.

Online misinformation increasingly appears in spoken formats such as news clips, podcasts, interviews, political speeches, and social media videos, creating a need for fact-checking systems that can verify claims directly from speech. We introduce VeriSpeak, a probe benchmark for studying speech-based fact verification in Large Audio Language Models (LALMs). VeriSpeak contains 3,879 spoken claims spanning temporal, geographical, and relational facts, with balanced true and false labels. The benchmark is designed to examine whether factual verification ability transfers from text to speech, and whether retrieval-augmented LALMs can use textual evidence to correctly support or refute spoken claims. Our experiments reveal a consistent text-speech modality gap: LALMs that verify written claims reliably often fail on the same claims when spoken. Moreover, retrieval alone provides limited gains because models frequently conflate retrieved evidence with the spoken claim. In contrast, retrieval combined with explicit reasoning improves claim-evidence comparison, with a thinking-tuned LALM reaching 86.1% accuracy. VeriSpeak highlights that effective speech misinformation detection requires not only speech understanding, but also grounded reasoning over retrieved evidence. The dataset is publicly available via Hugging Face at https://huggingface.co/datasets/abhiram4572/VeriSpeak.

Foundation models are post-trained with reinforcement learning (RL) to maximize specific rewards, such as human alignment, correctness, or instruction following. This post-training process is computationally intensive, sometimes unstable, and has to be run from scratch every time the reward model changes or when we want to combine multiple rewards. We hence ask: given a new reward function, is it possible to predict the RL outcomes without actually running RL on it? We answer this in the affirmative by introducing PoEM, a framework to predict the outputs of RL on a new reward function using a set of models already post-trained on other rewards. First, we show that if the new reward function can be written as a linear combination of existing ones, then the new policy in log-space can be written as a linear combination of the existing log-policies. Surprisingly, even in cases where the rewards are not linearly connected, we observe that often log-policies from RL training span an approximately low-rank subspace across rewards. To our benefit, the weighting coefficients for this combination can be estimated using only the reward or basis policy outputs on the samples. We turn these observations into an algorithm that takes post-trained models and a new reward function, and approximates the target RL policy without actually running any additional RL training. We experimentally validate our approach across synthetic and real rewards, spanning both text and image modalities.