arXiv:2607.09156v1 Announce Type: new
Abstract: Additive activation steering (injecting a scaled residual-stream direction during generation) is calibrated almost entirely in single-turn chat, yet the models it targets are increasingly deployed as tool-using ReAct agents. We present the first systematic chat-to-agent transfer study of additive steering, coupling behavioral measurement with a representation read-out in a matched-information design: the same items rendered as plain chat or as a ReAct tool-use episode, with matched-norm random-direction controls and the transcript re-encoded every turn to exclude KV-cache contamination. Transfer is real but rescaled, and the right description is a dissociation: the injected direction reaches the late layers at near-full strength in every setting and model tested (install-site agent-over-chat ratios 0.83-1.16 across three families), while the behavioral coupling is reset per model and context. On Qwen2.5-7B a refusal bypass vector amplifies in the agent (T = 1.45, CI [1.20, 1.78], N = 300); across a powered uniform-protocol distribution the coupling spans amplification (Gemma-2-9B T = 2.00) to attenuation (Yi-1.5-9B T = 0.43, CI [0.29, 0.60]), with no universal constant and a single clean attenuator against a universal sign. Directional ablation of the same axis does not amplify (T = 0.93, CI including 1) while additive injection amplifies (T = 1.50), a 20.1-point gain difference (CI [13.4, 26.8]) that identifies an additive-specific mechanism. Two pre-registered instruments converge to localize the rescaling to the ReAct format scaffold, before any tool observation, rather than to the observation boundary where a dilution account would predict it. The safety implication is immediate and unpredictable: agentic deployment amplifies steering-based refusal bypass by up to 2.00x on some models while others attenuate, so a deployment cannot assume a given model is safe under additive steering.
Science Journals
arXiv:2605.13922v2 Announce Type: replace
Abstract: During thDuring the last few years, the term Mechanistic Interpretability, a specific area, under the umbrella of explainable artificial intelligence (XAI), has been introduced, to explain the decisions made by complex machine learning (ML) models in critical systems like UAV intrusion detection systems (UAVIDS). In this paper, we apply best-practices for data pre-processing and examine a wide range of tree-ensembles, deep neural networks, hybrid stacking models and the latest ensemble neural networks to detect intrusions in UAV, with stratified 10-fold cross validation. With our top-performing model, XGBoost, we proceed to Shapley Additive explanations (SHAP), to analyze the global and local feature importances and understand which features, each attack targets, to mimic normal traffic and where the misclassifications occur. Furthermore a distribution analysis follows, by visually comparing violin plots and the curves of kernel density estimations. With the Westfall-Young permutation test for multiple comparisons, the Bandwidth optimization of the KDEs and the selection of Jensen-Shannon Distance for the test, we discover the true causes of false predictions, observed in Wormhole and Blackhole attacks in UAVIDS-2025. The findings provide robust, reliable and explainable models for UAV intrusion detection, along with statistical insights, which capture and clarify the masked nature of the attacks, regarding the challenge of Density Support Intersection, between these attacks, in this dataset.
arXiv:2607.09153v1 Announce Type: new
Abstract: Process Reward Models (PRMs) have been proven to be highly effective in guiding test-time scaling (TTS) methods, which significantly boost the capabilities of LLM-based multi-agent systems. However, existing PRMs are text-based: they re-encode the entire trajectory text from scratch. In long multi-agent rollouts, the scoring cost, growing quadratically with respect to sequence length L, creates a severe computational bottleneck, severely limiting PRMs' application in long-context scenarios. To resolve this, we introduce KV-PRM, a highly efficient process reward model that eliminates the heavy text re-encoding by directly reading the KV cache produced naturally during the LLM's generation phase. By processing a single "verify token" against the pre-existing KV cache, KV-PRM reduces the scoring cost from O(L^2) to O(L). We formally prove that the KV cache contains strictly greater information capacity than text, and is more efficient for downstream reward modeling. Empirically, across the MATH, GSM8K, and AIME benchmarks, KV-PRM matches or strictly outperforms text-PRMs under various TTS methods such as Beam Search, MCTS, and Weighted Voting, with up to a 5,000x reduction in scoring FLOPs, a 37x reduction in latency, and a 34x reduction in per-sequence memory footprint compared to text-based PRMs.
arXiv:2607.09194v1 Announce Type: new
Abstract: Modern control systems increasingly rely on heterogeneous sensors operating at different sampling rates, where intermittently missing outputs pose fundamental challenges for system identification. This paper proposes a non-iterative, control-oriented identification method for multirate systems based on cyclic reformulation. The method transforms multirate data into an expanded time-invariant representation and yields M parameter sets from a single input-output dataset, where M is the least common multiple of the sensor sampling periods. These parameter sets are used in two complementary ways: their centroid serves as a noise-reduced nominal model, while their convex hull gives a polytopic uncertainty model compatible with vertex-based LMI robust control design. Building on the noise-free structural recovery theorem of the authors' preceding work, which is restated here in the notation of the present paper, the present paper newly introduces the centroid and polytopic models derived from the M parameter sets; finite-noise behavior is treated as an empirical observation and is evaluated numerically. Numerical simulations support both models: an illustrative SISO example shows that the centroid attains higher validation FIT than the best individual vertex and substantially outperforms an interpolation-based baseline, while a MIMO multirate sensing example confirms, in line with the LTI counterpart, that the constructed polytope contains models whose validation FIT exceeds 95\% on average even at the highest tested noise level. The polytope is interpreted cautiously, with finite-noise behavior assessed through output-level validation statistics rather than realization-dependent matrix-coordinate distances. The proposed framework therefore links multirate system identification with robust-control-oriented uncertainty modeling without iterative EM-type optimization.
arXiv:2607.08808v1 Announce Type: new
Abstract: Recent advances in 3D Gaussian Splatting (3DGS) have enabled high-quality, render-ready scene representations for novel-view synthesis. However, most existing 3DGS pipelines rely on multi-view observations (or non-causal access to future frames) to achieve sufficient coverage, which is often unavailable in on-device robotics and AR settings where sensing is restricted to a single stereo rig. Recovering a high-quality 3DGS scene from one stereo observation, therefore, remains challenging due to occlusions, limited field of view, and missing geometry. We present StereoSplat+, a diffusion-enhanced feed-forward framework that enables causal reconstruction from a single stereo pair. Our method builds on two key components. First, we propose StereoSplat, an input-invariant feed-forward 3D Gaussian estimator that takes a variable number of posed stereo pairs as input and predicts high-quality 3D Gaussians. StereoSplat fuses complementary geometry cues via a cost-volume branch and a triplane-based 3D volume branch and leverages continuous pose encoding to generalize across view counts and camera configurations. Second, since multiple posed stereo pairs are typically unavailable at inference time, we introduce a diffusion-enhanced one-shot progressive inference scheme called StereoSplat+: starting from one stereo pair, we render novel stereo views from the predicted 3DGS, refine them with a one-step diffusion enhancer, and feed them back as additional inputs to update the 3DGS. Experiments on the KITTI-360 dataset show that StereoSplat+ improves novel-view rendering quality and geometry accuracy, especially in occluded regions and under strong view extrapolation, outperforming recent feed-forward 3DGS baselines.
arXiv:2607.09217v1 Announce Type: new
Abstract: In this system paper, we present OpenProver, an open-source system for LLM-driven automated theorem proving (ATP) with integrated Lean 4 formal verification. OpenProver integrates a Planner-Worker-Verifier architecture inspired by recent ATP agentic systems such as Aletheia. A Planner agent maintains a compact Whiteboard scratchpad and an unbounded Repository of intermediate findings, and decomposes mathematical work into parallel Workers.
OpenProver is fully open-source, offers reproducible evaluation through automatic formal verification of generated proofs, and provides an interactive terminal interface for human-guided proof search. In interactive mode, OpenProver allows the human operator to monitor and steer the proof search process, motivated by the established human-AI synergy in interactive code generation.
To showcase the potential for quantitative ablation experiments enabled by automatic formal verification, we evaluate OpenProver on ProofNet and compare it with a simple baseline. OpenProver is publicly available at https://github.com/kripner/OpenProver.
arXiv:2607.09192v1 Announce Type: new
Abstract: Mobile robots are going to share the sidewalks with pedestrians. They must ensure their objective safety and respect the walkers' subjective safety/comfort. Computationally efficient Social Force Models (SFM) present interpretable solutions for real-time robot navigation in dynamic crowds. Recent explorations of Projected Time-to-collision (PTTC) integration into SFM variants, for example, PTTC-based SFM (TSFM), improve safety metrics. But the effect of predictive variants is unclear. We introduce Predictive SFM (PSFM) and Predictive TSFM (PTSFM) by integrating predicted social force vectors over a finite time horizon. The paper implements SFM, TSFM, PSFM, and PTSFM on a nonholonomic mobile robot and performs experimental trials with volunteers attending a facing scenario. We systematically study objective and subjective safety across the variants. Minimum PTTC, average speed, minimum distance, lateral distance, and the maximum trajectory curvature benchmark the objective safety. Likert scale post-interaction surveys assess subjective safety by marking comfort, smoothness, distance appropriateness, and speed suitability. We confirm that PTTC integration improves safety metrics. The prediction contribution is limited and occasionally visible in some of the sub-metrics. Some participants perceive smoother movements and safer speed behavior with predictive methods, but Mann-Whitney tests reveal no significant differences in subjective ratings. Therefore, PTTC-based navigation enhances safety, whereas the formulated prediction offers limited additional benefits in single-pedestrian scenarios.
arXiv:2607.09175v1 Announce Type: new
Abstract: Deployed LLM agents rely on agentic context, the model-external textual control content assembled by an operational harness. In this work, the mutable component of that context is a persistent system-level instruction that is updated from operational experience while the model, tools, and harness remain fixed. Over long evolution horizons, flat-text maintenance makes verification increasingly difficult as accumulated instructions grow and interact. We propose Graph-Regularized Agentic Context Evolution (GRACE), which maintains the persistent instruction component as a typed semantic graph and validates proposed updates within the local typed neighborhoods of modified nodes. Accepted graph updates are reconstructed as incremental edits to the textual instruction checkpoint used at deployment. We evaluate GRACE within a fixed telecom agent harness derived from $\tau^2$-bench under a controlled distribution-shift protocol. Across five independent replications, GRACE improves strict reliability, measured by pass^3, from the Gemini 2.5 Flash zero-shot value of 0.091 to 0.673$\pm$0.136 at the final checkpoint. This exceeds a Gemini 3.1 Pro zero-shot reference of 0.242 on the same held-out set, while the flat-text HCE baseline finishes at 0.191$\pm$0.051. These results identify two requirements for reliable long-horizon context evolution, a structural substrate that makes verification local and a consolidation mechanism that keeps accumulated instruction content usable.
arXiv:2607.09179v1 Announce Type: new
Abstract: Recent LLM-based systems have shown promising capabilities for security-focused code analysis. Malware understanding, however, poses a distinct challenge: analysts must reconstruct high-level malicious behaviors under partial observability from sparse, dispersed evidence intertwined with benign functionality. While static analysis can expose security-relevant signals, the central challenge is not merely identifying suspicious code, but determining whether the evidence sufficiently supports an auditable behavior-level conclusion. We formulate malware understanding as a grounded reasoning problem and argue that reliable behavior reconstruction requires three complementary forms of grounding. Domain grounding constrains how behavior hypotheses are generated and evaluated, semantics grounding localizes and connects supporting program evidence, and knowledge grounding supports behavioral attribution through externally verifiable threat knowledge. To study this hypothesis, we present Malaika, a multi-agent framework that operationalizes the three grounding mechanisms through analyst-inspired reasoning, tool-mediated evidence localization, and retrieval-based behavioral attribution. We instantiate Malaika for Android malware analysis and evaluate it on malware-understanding tasks. Results show that Malaika improves analysis quality over prior LLM-based malware-analysis frameworks and demonstrate that reliability depends not only on model capability but also on the reasoning process. In particular, comparisons against malware-analysis systems and frontier agentic frameworks show that grounding-aware reasoning produces more precise and auditable conclusions. Ablation studies further support the grounding hypothesis. These findings suggest that grounding-aware reasoning provides a principled foundation for reliable malware understanding and, more broadly, for evidence-grounded software analysis.
arXiv:2512.09454v3 Announce Type: replace
Abstract: A glow nanosecond discharge from a plane-to-plane impulse dielectric barrier discharge (iDBD) with ammonia gas has been characterised by employing fast imaging and electrical diagnostics. More precisely, the aim of this study is to investigate the dynamics of the discharge establishment under various conditions of applied voltage, pressure, and gas gap. The comparison between the current measurements and the image analysis exposes a strong correlation between the fast excitation and ionization wave velocity and the rising current velocity. This correlation has been found only for diffuse mode discharge since a front wave could be clearly defined, denoted as the luminous propagation front (LPF). Furthermore, this correlation is supported by a proportionality factor of 1.5 10$^{-3}$ which is systematic over the studied conditions. Further investigations are considered to evaluate the relevance of such a value over more parameters.
arXiv:2402.07518v5 Announce Type: replace
Abstract: Large language models (LLMs) have significantly enhanced the usability of AI-generated code, providing effective assistance to programmers. This advancement also raises ethical and legal concerns, such as academic dishonesty and the generation of malicious code. For accountability, it is imperative to detect whether a piece of code is AI-generated. Watermarking is broadly considered a promising solution and has been successfully applied to identify LLM-generated text. However, existing efforts on code are far from ideal, suffering from limited universality and excessive time and memory consumption. In this work, we propose a plug-and-play watermarking approach for AI-generated code detection, named ACW (AI Code Watermarking). ACW is training-free and works by selectively applying a set of carefully-designed, semantic-preserving and idempotent code transformations to LLM code outputs. The presence or absence of the transformations serves as implicit watermarks, enabling the detection of AI-generated code. Our experimental results show that ACW effectively and efficiently detects AI-generated code, preserves code utility, and is resilient against potential code disruptions. Especially, ACW is universal across different LLMs, addressing the limitations of existing approaches.
arXiv:2404.05664v2 Announce Type: replace
Abstract: We find the average time complexity of the breadth-first search (BFS) and the depth-first search (DFS) algorithms, when one searches for a target node selected uniformly at random among all nodes at level $\ell$ in the set of ordered trees with $n$ edges. Intuition suggests that on average BFS must be asymptotically faster than DFS if and only if $\ell$, as a function of $n$, is below a certain threshold. We confirm this intuition by showing that there exists a unique constant $\lambda\approx 0.789004$, such that in expectation BFS is asymptotically faster than DFS if and only if $\ell\leq \lambda\sqrt{n}$. This gives us a practical rule to select between the two algorithms, even when we do not know the exact value of $\ell$, but only an estimate of it. Furthermore, we find the asymptotic average time complexity of BFS in the given setting for an arbitrary class of Galton--Watson trees, which includes ordered trees, binary trees, and other popular classes. We use results on the occupation measure of Brownian excursions, as well as combinatorial identities related to lattice paths. Finally, we consider the simple \textit{truncated DFS} algorithm, which can be shown easily to be asymptotically faster than both BFS and DFS when $\ell$ is known in advance. We show that in fact its asymptotic time complexity is $1/2$ of the asymptotic complexity of BFS, when $\ell = s\sqrt{n}$ for any constant $s$. Several further questions are also raised.
arXiv:2607.09191v1 Announce Type: new
Abstract: Generated videos provide useful visual motion priors for robot manipulation, but their visual plausibility does not imply physical executability. A generated video usually lacks metric geometry, grasp grounding, robot kinematic feasibility, and execution-time feedback, which makes direct trajectory replay unreliable in real-world manipulation. This paper presents GenVid2Robot, a rigid-geometric consistency framework that converts generated video motion into executable real-robot manipulation trajectories. Given an initial RGB-D observation and a task instruction, GenVid2Robot samples task-relevant semantic anchors from the real first frame, tracks these anchors through generated video candidates, and verifies whether the resulting 2D motion can be explained by first-frame RGB-D anchors under a sparse relative $SE(3)$ model. In this way, generated videos are treated as uncertain visual motion hypotheses rather than direct robot demonstrations. Only geometrically consistent motion is transferred to the robot. The accepted relative motion is then applied to the real grasp-time TCP pose selected by mask-constrained grasping, producing a grasp-conditioned execution trajectory that is consistent with both the visual motion prior and the physical grasp configuration. To reduce execution mismatch caused by RGB-D noise, calibration residuals, and small contact-induced displacement, a bounded depth-compensation module corrects local depth-direction errors without assuming full online replanning. Real-robot experiments demonstrate that GenVid2Robot improves the reliability of generated-video-guided manipulation by grounding visual motion priors with sparse metric geometry, grasp constraints, robot feasibility checking, and bounded execution feedback.
arXiv:2602.12873v5 Announce Type: replace
Abstract: Generative social robots (GSRs) powered by large language models enable adaptive, conversational tutoring but also introduce risks such as misinformation, overreliance, and privacy violations. Existing frameworks for educational technologies and responsible AI primarily define desired behaviors, yet they rarely specify the knowledge prerequisites that enable generative agents to express these behaviors reliably. To address this gap, we adopt a knowledge-based design perspective and investigate what information tutoring-oriented GSRs require to function responsibly and effectively in higher education. Based on twelve semistructured interviews with university students and lecturers, we identified twelve design requirements across three knowledge types: self-knowledge (assertive, conscientious, and friendly personality with customizable role), user-knowledge (personalized information about student learning goals, learning progress, motivation type, emotional state, and background), and context-knowledge (learning materials, educational strategies, courserelated information, and physical learning environment). Drawing from these results, this work provides a structured foundation for the design of tutoring GSRs, aligning generative AI capabilities with pedagogical and ethical expectations.
arXiv:2607.09251v1 Announce Type: new
Abstract: Statement-level SQL rewriting can improve query performance and maintainability without changing the DBMS kernel, but existing benchmarks do not evaluate rewrite methods as deployable systems. They typically focus on DBMS performance, rule regression, query equivalence, or dialect translation, while missing the full path from accepting an input query to producing an executable, result-consistent, and operationally useful rewrite. We present SQL-RewriteBench, a benchmark for statement-level SQL rewriting that applies correctness gating and full-denominator accounting. Its metric suite explicitly separates Source Acceptance, Generation Rate, Execution Coverage, Result Consistency, UnsafeRewrite Rate, and speedup distribution. It also defines SCS, a deterministic index of static SQL structure, and CGOQ, a correctness-gated optimization-quality score that gives optimization credit only after the case-specific Checker Contract is satisfied. CGOQ combines runtime improvement with structural simplification through a continuous scoring function, making it suitable for deployment-oriented rewrite assessment. As an artifact, SQL-RewriteBench provides 180 executable Benchmark Instances organized into EQUIV, PERF, ROBUST, and DIALECT pools, each packaged with SQL, schema metadata, provenance, evidence, and rewrite-opportunity documentation. Across seven representative academic and LLM-based methods, every full-benchmark CGOQ is negative. Existing methods often fail before rewriting, fail result checks, or return correct rewrites that are slower or no better than the input. These results show that deployable SQL rewrite requires broader input handling, result validation, and benefit-aware rewrite decisions.
arXiv:2607.09193v1 Announce Type: new
Abstract: Real-world sRGB image denoising remains challenging due to the nonlinear characteristics of sensor noise and the difficulty of acquiring aligned clean-noisy image pairs. Supervised denoisers often overfit to limited paired datasets, while self-supervised methods still depend on sufficiently diverse noisy observations. These limitations motivate scalable noise synthesis methods that can model real-world noise without clean ground truth or camera metadata. We propose YeTI, a real-world sRGB noise generation framework that learns from only two noisy observations of the same scene. YeTI uses a Reconstruction Autoencoder to disentangle scene structure and noise characteristics, and models the latent noise distribution with a one-step Conditional Diffusion Transformer trained using consistency objectives. Given a single noisy input at inference time, YeTI generates realistic, signal-dependent noise while preserving the underlying scene content. Extensive experiments demonstrate the effectiveness of YeTI across real-world benchmarks. We evaluate noise generation on SIDD and further assess generalization on SIDD+, MAI2021, and SID, covering smartphone and diverse consumer-camera sensors. Downstream denoising results on DND further show that denoisers trained with YeTI-synthesized images achieve strong real-world performance, highlighting the practical value of clean-image-free and metadata-free noise generation.
arXiv:2607.09202v1 Announce Type: new
Abstract: Continual learning commonly relies on post-hoc mechanisms such as replay, elastic regularization, or distillation. This work argues that forgetting should instead be modeled directly as interference between tasks. In the frozen-feature regime, forgetting from learning a new task is exactly the interference energy induced on the old task. In deep networks, the same quantity is recovered through path-averaged curvature with minimal additional forward passes.
When task supports are disjoint, forgetting can be eliminated structurally and when task supports overlap in conflicting directions, a non-zero distortion floor is unavoidable. The same geometry optimally merges models through task-aware orthogonalization. From this analysis we derive Interference-Gated Functional Allocation (IGFA), a replay-free, Fisher-free method that shares directions when tasks align and protects them when they conflict. Across benchmarks, IGFA achieves lossless retention when tasks are structurally separable and moves unavoidable cost from irreversible forgetting into deferred but recoverable plasticity when they are not. It matches the strongest replay-free structural baselines on dissimilar-task streams and improves on unconditional projection when similarity makes transfer worth preserving.
arXiv:2410.21019v2 Announce Type: replace-cross
Abstract: This paper develops a network-based methodology for measuring economic integration in Africa. Conventional indicators such as trade openness and intra-regional trade shares capture the volume of trade but not countries' structural roles within the continental trade system. Using bilateral intra-African trade data for 2000-2019, we construct annual directed trade networks and measure countries' direct connectivity, recursive influence, brokerage role, and core embeddedness using weighted degree, PageRank, betweenness and random-walk betweenness, clustering, and k-core network indicators. We further benchmark observed network positions against a maximum-entropy null model to identify countries whose centrality exceeds that expected from their trade intensity alone. The results show a gradual densification of intra-African trade and the persistence of a core-periphery structure led by a small group of highly connected economies. Dynamic panel estimates indicate that economic development, infrastructure, institutional quality, human capital, regional trade agreements, and FDI are associated with stronger structural integration, whereas trade costs and overlapping regional memberships weaken several dimensions of network position. The findings suggest that African integration depends not only on expanding trade volumes but also on improving countries' positions within continental trade.
arXiv:2607.09219v1 Announce Type: new
Abstract: Doppler backscattering (DBS) measures electron density fluctuations. The measured wavenumber is typically varied by changing the probe beam poloidal launch angle. As most DBS systems are unable to steer during a shot, the shot is repeated and the poloidal angle is changed intershot. An alternative method is to keep the DBS launch angle fixed and move the plasma up and down instead, enabling a range of wavenumbers to be measured within a single shot. We call this the bouncing ball method. We use the Scotty synthetic diagnostic (Hall-Chen, 2022) to evaluate the similarities and differences between these two approaches. Both approaches are capable of measuring a similar range of fluctuation wavenumbers as well as radial locations. However, the vertical-displacement scan has a larger range of measured poloidal locations than the poloidal-angle scan. Using the same synthetic turbulence spectrum as input, we show that the two approaches are expected to have different backscattered powers due to different instrumentation functions. When mismatch attenuation is accounted for via a synthetic diagnostic, the vertical-displacement scan can provide comparable radial and wavenumber coverage while reducing reliance on shot-to-shot repeatability. These results establish vertical-displacement scan as a practical route to single-shot DBS wavenumber spectra.
arXiv:2607.09287v1 Announce Type: new
Abstract: Large language models (LLMs) remain expensive to fine-tune because full-parameter updates require substantial memory, compute, and per-task storage. We study whether saliency signals originally developed for pruning can be reused to choose where a model should adapt. We propose Super, a sparse parameter-efficient fine-tuning (PEFT) method that fixes a small trainable support using a Wanda-style activation-weighted magnitude score [Sun et al., 2023] computed from a calibration pass. We then introduce Supra, a hybrid adapter that combines this sparse update with LoRA while preserving a matched trainable-parameter budget through a simple budget-splitting rule. In single-seed Math17K arithmetic experiments on Llama-3.2-1B and Meta-Llama-3-8B, the best Super/Supra variants achieve the highest average accuracy among the tested schedule-selected adapter configurations. We also include a PaFi-style magnitude-only support as a closest training-free sparse baseline and find that low-score supports under both magnitude and Wanda-style orderings can be effective. These results suggest that simple pruning-inspired orderings can provide useful fixed sparse supports for PEFT, especially when combined with low-rank adapters.
arXiv:2607.09289v1 Announce Type: new
Abstract: We study exact algorithms for Equal-Subset-Sum in the worst-case setting: given a set $S$ of $n$ integers, find two distinct subsets $A,B\subseteq S$ whose sums are equal. We establish a new state-of-the-art bound for this problem by improving the fastest known algorithm, due to Randolph and W\k{e}grzycki (STOC 2026), from $O^*(1.7067^n)$ time and space to an algorithm that runs in $O^*(1.6994^n)$ time and uses $O^*(1.5664^n)$ space. We also improve the best known polynomial-space running time, due to Mucha, Nederlof, Pawlewicz, and W\k{e}grzycki (ESA 2019), from $O^*(2.6817^n)$ to $O^*(2.5430^n)$. Finally, we investigate time-space tradeoffs for this problem and improve the running times achievable under a broad range of exponential-space bounds.
arXiv:2607.09089v1 Announce Type: new
Abstract: In the task of human mesh recovery (HMR), multi-person scenes are particularly difficult to handle due to the many entities that appear and occlusions between them over time. In particular for video inputs, there is a need to track each entity reliably and consistently. Existing methods rely on pretrained human detection modules, increasing their runtime and limiting the number of tracked entities. We present DETRAM, a unified framework for multi-person HMR and tracking that simultaneously detects, reconstructs, and tracks humans across time, both automatically and via user prompts. DETRAM uses a single transformer decoder with an identity-consistent set of learnable query embeddings that persist across frames: detection queries discover new people, tracking queries maintain pose and shape for existing individuals, and prompt queries follow user-specified identities. Our approach achieves state-of-the-art tracking results on PoseTrack21, 3DPW, BEDLAM, and MuPoTS-3D, and competitive reconstruction accuracy on BEDLAM and 3DPW, while uniquely supporting prompt-based tracking of individuals in multi-person scenes. To our knowledge, this is the first method to unify promptability and multi-person HMR with tracking in an end-to-end trainable framework, enabling user-directed human analysis in videos.
arXiv:2607.09316v1 Announce Type: new
Abstract: Thematic indexing -- the practice of assigning structured conceptual labels to sections of text -- is essential to scholarly access in large-scale literary and historical editions, yet it remains a largely manual, labour-intensive process. This paper explores the application of machine learning to automatic thematic indexing, using two substantial sub-corpora of the Complete Works of Voltaire as a test case: the Essai sur les m\oe urs et l'esprit des nations and the Questions sur l'Encyclop\'edie. The task is framed as a multi-label classification problem, in which a model must assign the set of index entries that a professional indexer would apply to a given page of text. We compare a range of approaches -- from encoder-based models with classification heads to generative large language models (LLMs) fine-tuned via Low-Rank Adaptation (LoRA) -- spanning model sizes from approximately 3 to 120 billion parameters. Our best-performing model, from the Mistral family in a 4-bit quantised configuration, achieves F1 scores of up to 0.67; we argue that these figures represent lower bounds, given the inherent subjectivity of professional indexing and the frequency with which model predictions prove semantically valid despite diverging from the print index. We further evaluate cross-corpus generalisation and conduct a detailed qualitative analysis of model behaviour on literary and rhetorical features of the source texts that prove particularly resistant to automated treatment. Our findings have implications for the broader challenge of providing structured thematic access to large-scale literary and historical corpora.
arXiv:2607.09229v1 Announce Type: new
Abstract: Laser-driven proton acceleration from nanostructured solid targets has been extensively studied, yet its performance under realistic temporal contrast conditions at petawatt-class facilities remains an open question. We present an experimental investigation of proton generation from nanostructured and flat solid targets performed at the ELI-NP facility using femtosecond laser pulses at peak intensities of $\sim 3\times10^{21}$ \wcm. Proton spectra are compared for two contrast regimes: $\sim 10^{-10}$ without plasma mirror and $\sim 10^{-13}$ with single plasma mirror. Importantly, measurable enhancement in the cutoff energy persists for the nanowire targets at both contrast levels, indicating robustness of nanowire targets against moderate pre-pulse intensities. Alongside, study of energy resolved angular distribution reveals that nanowires promote more directional emission with higher flux of high-energy protons along the target normal, while flat targets produce broader angular distributions. The results are well supported and explained by 3D particle-in-cell simulations.
arXiv:2509.13907v3 Announce Type: replace
Abstract: Few-shot 3D Point Cloud Semantic Segmentation (FS-PCS) aims to predict per-point labels for an unlabeled point cloud, given only a few labeled examples. To extract representations from the limited labeled set, existing methods have constructed prototypes with Farthest Point Sampling (FPS). However, we found that this convention results in performance instability due to its sensitivity to FPS-induced variations, while the prototype generation process remains underexplored in the field. This motivates us to investigate deterministic prototype generation method based on attention mechanism. Despite its potential, we found that vanilla attention module suffers from the distributional gap between prototypical tokens and support features. To overcome this, we provide a simple approach, White Aggregation and Restoration Module (WARM), which resolves the misalignment by wrapping cross-attention with whitening and coloring transformations. Specifically, whitening aligns the features to tokens before the attention process, and coloring subsequently restores the original distribution to the attended tokens. This design enables robust attention, thereby generating prototypes that capture the semantic relationships in support features. WARM achieves state-of-the-art performance with a significant margin on the S3DIS dataset, and competitive performance on the ScanNet dataset. Further experiments demonstrate its effectiveness in deterministic prototype generation. Code is publicly available at: https://github.com/JiyunIm00/WARM.git