arXiv:2508.10029v3 Announce Type: replace
Abstract: Safety-aligned large language models can still be manipulated through white-box interventions that modify their internal representations. We introduce Latent Fusion Jailbreak (LFJ), which works by pairing a harmful query with a structurally similar but benign counterpart, then interpolating their hidden states at carefully selected layers and token positions. Refusal-loss gradients determine exactly where to intervene, and we optimise layer-wise mixing coefficients using token-normalised compliance and refusal-suppression objectives. The edited prompt states propagate sequentially through the remaining transformer blocks. Across four safety benchmarks and five open-weight target models, LFJ reaches a macro-averaged attack success rate (ASR) of 94.13% under the white-box protocol we describe. Because LFJ directly accesses internal states, comparisons with prompt-only attacks serve as a descriptive reference rather than a matched evaluation. Dropping rejection sampling lowers ASR to 86.72%, whereas replacing the structured harmful-benign pairing with random pairing causes it to fall to 27.45%. We also design an LFJ-specific latent adversarial training procedure that, when the attack is re-optimised against the defended model, reduces ASR from 94.13% to 12.37%. This defence evaluation does not cover transfer to other attack types or preservation of benign utility.
Science Journals
arXiv:2607.15707v1 Announce Type: new
Abstract: AI-based labor-market systems or platforms can affect access to job opportunities prior to organizational candidate rankings or hiring decisions. Such applications warrant caution, as biases in skill extraction, profile formation, and candidate-job matching may contribute to unfair treatment of candidates. In this paper, we propose a two-stage framework for detecting and governing bias in skills-based job matching. Stage 1, skill extraction and profile formation, addresses how candidates provide skills and preferences to the system, how the system extracts and structures this information, and the bias risks this entails, with a focus on chatbot-based elicitation. Stage 2, multistakeholder candidate-job recommendation, would embed this information in a recommender system in which candidate, company, and regulatory objectives are represented by separate agents, each producing an independent candidate-job ranking; these rankings would be combined through social choice-based aggregation into a single, auditable recommendation. The two stages are connected by a shared distinction between hard constraints, which require correction before processing continues, and soft constraints, which are logged to inform later decisions. Following an AI Act-aligned assessment methodology (based on the Fraunhofer AI Assessment Catalog), we propose using distributional auditing and counterfactual testing to produce a Stage 1 bias inventory sorted into hard and soft constraints, with the latter informing fairness thresholds for Stage 2. The same logic would apply to Stage 2: fairness metrics crossing predefined thresholds would trigger an adapted recommendation process, while smaller deviations would be logged as bias reports and persistent fairness states.
arXiv:2201.08501v2 Announce Type: replace
Abstract: In this article, we investigate the behaviour of TMs with time limit and tape space limit. This problem is in P when the time limit is unary coded. If both limits go to infinity, it is undecidable which limit is exceeded first. Thus logspace-incomplete sets in P can be constructed. This implies L $\not=$ P.
arXiv:2607.15976v1 Announce Type: new
Abstract: Hybrid quantum systems offer a route to combining the complementary strengths of distinct quantum platforms while mitigating their limitations. A particularly promising architecture combines neutral atoms and polar molecules: atoms provide fast, controllable interactions through excitation to Rydberg states, while molecules possess long-lived rotational states that are attractive for quantum memories and qudits. Although dipolar interactions between atoms and molecules have been observed in gas-phase and beam experiments, they have not previously been explored in a scalable optical tweezer platform that enables the controlled coherent interactions needed for quantum state transfer and entanglement. Here, we realise this goal, demonstrating coherent dipolar interactions between an individual Rydberg atom and an individual polar molecule. The separation of the particles is controlled using species-specific optical tweezers and their dipolar interactions are made strongly state-dependent by tuning two atom-molecule pair states into resonance. We exploit these interactions to demonstrate atom-mediated state readout of a molecular qubit, observe coherent spin exchange between the particles, and generate entanglement using a blockade-based controlled-NOT operation. Together, these results establish a coherent atom-molecule interface in which long-lived molecular quantum information can be rapidly mapped onto internal states of a Rydberg atom for readout or onward coherent transfer. This platform can be scaled to realise hybrid quantum processors utilising atom-mediated readout and entanglement of molecular qubits and mixed-species quantum simulators of dipolar systems.
arXiv:2607.15982v1 Announce Type: new
Abstract: Learned policies trained end-to-end on large datasets often remain brittle in high-precision tasks and struggle with generalization. We find that these limitations largely stem from a lack of structure and focus in data collection. Our key insight is to leverage dense data collection only for the critical segment of contact-rich tasks and to rely on traditional planning during simple free-space motion. We propose an automated data-collection scheme in combination with offline deep reinforcement learning for the critical segment of the task, eliminating reliance on a teleoperator's skill and on online policy updates. Across four challenging real-world tasks, using only 2 to 2.5 hours of autonomous data collection, we achieve an average success rate of 96%, compared to the strongest baseline at 55%. Notably, performance remains high in out-of-distribution scenarios where end-to-end approaches struggle. Our results pave the way for targeted data collection for contact-rich tasks and for high success rates in precision applications.
arXiv:2607.15680v1 Announce Type: new
Abstract: Grid-forming (GFM) energy storage system (ESS) is a key enabler for stabilizing future power systems with high penetration of converter-based resources (CBRs). To get a better overview of the state-of-the-art and challenges for implementing and deploying GFM-ESS, a global survey has been initiated by Cigre Working Group B4.101 - industrial implementation and application of grid forming energy storage systems. Feedback was collected from universities, transmission system operators (TSOs), power plant developers, original equipment manufacturers (OEMs), research institutes, as well as consultants. It is interesting to note that while many common understandings have been established in practice, certain gaps persist among different stakeholders. This article intends to bridge this gap by presenting a summary of the survey, including the questionnaire, responses from various stakeholders, and in-depth analysis of the survey results. The key challenges faced by different stakeholders in deploying GFM-ESS are identified, shedding light on future research in this direction.
arXiv:2607.15322v1 Announce Type: new
Abstract: A large-scale microgrid typically consists of several cross-regional subgrids aggregated by a virtual power plant (VPP). However, current consensus based schemes can-not guarantee the feature of differential demand between subgrids. Thus, distributed cluster consensus control induced by communication weights is investigated in this paper to solve the ED problem of a large-scale microgrid, which can achieve the expected cluster via well-designed communication weights. A communication weight matrix design method for a directed and connected graph based on eigenvector centrality is designed, which enables the adjacency matrix of the communication network to have a given leading eigenvector and allows agents in each cluster to have the same eigenvector center value. Based on this, a distributed cluster ED scheme, namely a leader-follower cluster consensus controller, is designed to drive marginal cost (MC) to achieve multiconsensus, thus allocating power among DGs. In addition, the power deficit of each subgrid collected by a VPP can be allocated to utility grids according to predetermined ratios, thus maintaining power supply-demand balance of each subgrid. For this scheme, it should be emphasized that the weighted network used is directed and connected; meanwhile, leader information only can be accessed by a few clusters. Correspondingly, relevant simulations are attached to verify the effectiveness of the designed scheme.
arXiv:2607.15690v1 Announce Type: new
Abstract: Urban Air Mobility (UAM) with electric Vertical TakeOff and Landing (eVTOL) vehicles can help address ground traffic congestion. The design of an eVTOL trajectory that is safe and reduces travel time is key for UAM adoption. Existing works on trajectory design either may not adequately incorporate dense obstacles in urban environments, complex eVTOL flight dynamics, or one or more flight phases. Not considering these factors can result in low-quality, or worse infeasible, trajectories. We develop a hybrid framework that can integrate building obstacles data, wind data, eVTOL flight dynamics, and other real-world operational constraints to estimate a four-dimensional eVTOL flight trajectory in ascent, cruise, and descent that aims to minimize travel time. Our framework first fills the obstacle-free regions with intersecting convex polygons, then identifies potentially low-travel time candidate sequences of these polygons using a Graph of Convex Sets-based path planner, and then uses an Optimal Control Program to give the final trajectory that passes through the polygons in a sequence identified before. We evaluate our framework on routes within New York City. Our framework can design trajectories respecting the above constraints in the presence of as many as 250 building obstacles. We show that not including the above constraints can underestimate the flight time by as much as 20\%.
arXiv:2607.15812v1 Announce Type: new
Abstract: Cone-beam computed tomography (CBCT) is fundamentally challenged by scatter and beam hardening artifacts, which originate from X-ray scattering and the polychromatic nature of the X-ray spectrum, respectively. These two types of artifacts are intricately coupled in reconstructed images and manifest with similar streaking and cupping features, severely compromising high-precision CBCT imaging. This paper proposes a physics-driven iterative framework rooted in the polychromatic Polyquant attenuation model, which decouples these artifacts by establishing an optimization loop between scatter estimation and relative electron density (RED) reconstruction. We develop a hybrid strategy for scatter estimation, in which the first-order scattering component is analytically derived based on a polychromatic physical model to preserve high-frequency structural information, whereas the smoother multiple scattering component is efficiently estimated via an object-adaptive convolution module. Subsequently, for beam-hardening correction, we introduce a voxel-adaptive update mechanism that solves linearized, scatter-corrected polychromatic equations to derive optimal weights, enabling direct RED refinement without manual parameter tuning. The proposed method was validated through comprehensive studies on biomedical phantoms, utilizing both Monte Carlo simulations and physical experiments. Representative results demonstrate that the proposed method outperforms state-of-the-art techniques, with the mean relative error decreased from 11.96\% to 1.27\% for the anthropomorphic head phantom and from 12.55\% to 5.46\% for the physical Yin-Yang phantom.
arXiv:2607.15325v1 Announce Type: new
Abstract: Multisensory integration, particularly through visual and tactile feedback, plays a crucial role in enhancing audience engagement with artworks. Although recent research has increasingly explored tactile experiences in art, existing systems often lack real-time variable stiffness modulation and depend on bulky mechanical infrastructures. In this work, we propose a novel tangible display based on a magnetic jamming mechanism, enabling real-time, low-noise, and low-voltage stiffness modulation integrated into traditional sculptural artworks. Our system combines visual motion and dynamic tactile feedback within a compact standalone module, allowing audiences to interactively experience variations in the rigidity and form of features such as those found in the traditional Korean mask Hahoetal. This approach offers a new paradigm for interactive art, enabling more immersive, multisensory engagement through the fusion of cultural artifacts and modern technology. Our project page is available at https://cold-young.github.io/jamming_tangible/.
arXiv:2607.15865v1 Announce Type: new
Abstract: Large Language Models (LLMs) have become the dominant workload on modern AI accelerators, yet deploying them on specialized hardware still faces two core challenges: how to import a trained model into a compiler-friendly intermediate representation, and how to efficiently schedule the autoregressive inference loop under limited on-chip memory. This paper presents an MLIR (Multi-Level Intermediate Representation) based compilation method for large language models, illustrated using two dialects of operators, TopOp and TpuOp. TopOp serves as a high-level graph dialect that is independent of both the source framework and the target chip, and is responsible for expressing model semantics; TpuOp serves as the target hardware dialect, carrying chip-related decisions such as quantization, layer groups, and memory layout. A model is first represented as TopOp, then lowered layer by layer to TpuOp, and finally a deployable binary is generated. In addition, each Transformer layer is split into three stages for static compilation: prefill, prefill_kv (prefill with historical key-value cache), and decode, so as to accommodate the different computational characteristics of prompt-parallel processing and per-token generation. The method has been implemented in the TPU-MLIR compiler{https://github.com/sophgo/tpu-mlir} and the LLM-TPU deployment project\footnote{https://github.com/sophgo/LLM-TPU}, supporting a variety of generative models including the Qwen, Llama, InternVL, and MiniCPM-V series, as well as multiple quantization and deployment forms such as GPTQ, AWQ, and AutoRound.
arXiv:2607.15425v1 Announce Type: new
Abstract: We present a quantum-classical algorithm for electronic structure calculations that dramatically reduces the quantum measurement cost of variational quantum eigensolver (VQE) approaches. While conventional VQE methods require measurements scaling as O(M^4) with system size M, the proposed occupation-number VQE (ON-VQE) reduces this cost to O(M/2) by avoiding reduced density matrix (RDM) measurements and relying exclusively on ONs. The method exploits only the diagonal elements of the one-particle RDM in the natural orbital representation, where occupations are obtained directly from computational-basis measurement outcomes. By restricting the variational ansatz to double excitations within orbital subspaces associated with electron pairs, the required measurements can be grouped into a small number of qubit-wise commuting observables, yielding an efficient and scalable measurement strategy. The approach is validated through simulations and executions on quantum hardware for the cubic H$_8$ cluster, demonstrating the feasibility of extracting accurate ONs from quantum measurements and evaluating electronic energies within the natural orbital functional (NOF) framework. Across representative molecular systems, the extracted ONs enable accurate energy evaluation with state-of-the-art NOFs while maintaining a dramatically reduced measurement cost. These results establish a scalable route toward quantum simulation of strongly correlated electronic systems, demonstrating that accurate electronic energies can be obtained from quantum measurements of ONs alone.
arXiv:2607.15714v1 Announce Type: new
Abstract: Vision-Language-Action (VLA) models excel at end-to-end robotic manipulation but struggle with out-of-distribution (OOD) generalization when familiar sub-tasks are recombined in unseen configurations. We identify two mutually reinforcing failure modes: \emph{trajectory overfitting}, where models overfit to holistic trajectory patterns rather than compositional sub-skill semantics; and \emph{perceptual shortcut}, where action tokens over-rely on wrist-view textures at the expense of global spatial grounding. To address both, we introduce \textbf{AC-VLA}, a plug-and-play Action Compositional learning framework comprising two architecture-agnostic components: \textbf{(i)} a compositional learning module that uses an LLM-driven instruction decomposer and a proprioceptive trajectory aligner to generate dense sub-task supervision, followed by mixed training on complete demonstrations and decomposed data to endow the model with compositional generalization; and \textbf{(ii)} a state-conditioned asymmetric masking strategy that suppresses wrist-view inputs during closed-gripper phases, enforcing global semantic grounding. All components are architectural modification-free and directly integrable into any VLA backbone. Instantiated on $\pi_{0.5}$ and evaluated on LIBERO and LIBERO-OOD benchmarks, AC-VLA achieves a ~28% absolute improvement on compositional OOD tasks while maintaining near-perfect in-distribution performance.
Statistical State Dynamics Eigenmodes and Equilibria Form the Structural Basis of Couette Turbulence
arXiv:2607.15842v1 Announce Type: new
Abstract: Wide channel Couette (WCC) turbulence consists primarily of steady roll streak structures (RSS) maintained by a self-sustaining process, yet the Navier Stokes equations in velocity variables admit no linear RSS instability or stable equilibrium, leaving the WCC turbulent state's analytical basis obscure. We show that in a second order statistical state dynamics (SSD) a fixed-point state of Couette turbulence arises from a modal RSS instability and equilibrates as an exact, attracting RSS at spanwise wavenumber 3. This fixed point comprises a rank-1 streamwise mean flow and a conjugate pair of neutral eigenmodes, regularized by roll advection rather than viscosity, forming a rank-2 fluctuation covariance. WCC turbulence arises as a spanwise tiling by this fixed-point RSS unit cell.
arXiv:2607.15715v1 Announce Type: new
Abstract: Large language model (LLM) agents are increasingly used for complex information-extraction tasks, yet it remains unclear whether agentic components such as reflection and memory lead to observable and controllable improvements over fixed LLM workflows. We study this question through conference-paper dataset extraction, where a system must identify datasets mentioned in scholarly PDFs and produce structured records. We compare a fixed workflow baseline with reflective agent variants and specify an optimized agent condition (S2) that extends the same task with richer PDF tools and dynamic tool selection. Our evaluation emphasizes process-level behavior--including tool execution, retries, reflection, memory use, runtime, and failure recovery--while treating extraction coverage and field completeness as secondary outcome measures. The paper characterizes when agentic mechanisms change system behavior, whether these changes improve task completion, and how the observed failure modes motivate an optimized agent design under the same evaluation harness.
arXiv:2607.16080v1 Announce Type: new
Abstract: Precipitation nowcasting over the immediate 10-90 min period is important for flood management and real-time decision-making in urban regions. Conventional short-range forecasting with high-resolution numerical weather prediction requires frequent data assimilation, model initialization, and spin-up, introducing computational latency. Machine learning provides an alternative by learning storm evolution directly from high-frequency observations and producing forecasts quickly after training. This is particularly relevant for Mumbai, India, where monsoon convection, land-sea interactions, and localized intense rainfall make short-term prediction difficult. Here, we develop a compact radar-only nowcasting framework that combines multi-elevation reflectivity, Doppler radial velocity, and radial-velocity-gradient proxy features within an encoder-decoder U-Net. Using the most recent radar volume scan, the model predicts 12 future composite reflectivity fields at 7.5-min intervals up to 90 min lead time. The derived velocity magnitude, divergence-like, directional-shear, and vorticity-like channels represent kinematic signatures associated with convergence and boundary interactions without requiring full wind-field retrieval. A high-reflectivity attention module improves sensitivity to convective cores, and physics-guided attribution examines whether the learned sensitivities are meteorologically meaningful. The model is trained using Mumbai Doppler radar observations from May to August 2023 and evaluated on temporally independent events. At 90 min lead time, Critical Success Index values are 0.437, 0.332, and 0.193 for $\geq$10, $\geq$20, and $\geq$30 dBZ thresholds, respectively. Compared with persistence, the model gives lower RMSE and higher spatial correlation at longer lead times. Once trained, it runs on a standard computer, generating nowcasts within seconds for real-time use.
arXiv:2607.15737v1 Announce Type: new
Abstract: We study the Connected Components Reconfiguration problem (CCR), in which connected components on a graph are transformed according to a specified reconfiguration rule. CCR generalizes Independent Set Reconfiguration by treating tokens not as individual vertices but as connected components of prescribed sizes. Among the variants of CCR, we focus on the component-jumping model, denoted by \CCRCJ. Nakahata.\ introduced this problem and showed that the decision problem for \CCRCJ~can be solved in $O(n^2)$ time on path graphs for arbitrary component sizes, and in polynomial time on chordal graphs when all connected components have the same size. However, the complexity on chordal graphs under a multiset size constraint remained open.
In this paper, we study this multiset version of \CCRCJ~from both complexity-theoretic and algorithmic viewpoints. First, we prove that \CCRCJ~is NP-hard even on caterpillar graphs, which is a very restricted subclass of trees and chordal graphs minimally above path graphs. This result immediately implies NP-hardness for chordal graphs under a multiset size constraint, thereby resolving Nakahata's open problem on chordal graphs under multiset size constraints. Second, we revisit \CCRCJ~on path graphs. We improve the previous $O(n^2)$-time algorithm for the decision problem by giving an $O(n\log n)$-time decision algorithm. Moreover, when the instance has sufficiently large empty space, we show that there exists a reconfiguration sequence of length $O(n\log n)$, and such a sequence can be output efficiently.
arXiv:2607.15890v1 Announce Type: new
Abstract: Perceiving multimodal cues and forecasting fine-grained actions from an egocentric (Ego) perspective is vital for applications like robot manipulation. However, previous studies either rely mainly on under-informed visual inputs to predict coarse human motions or follow the VRM/VLA paradigm, which suffers from insufficient robot data and the gap between human and robot embodiments. We observe that 3D hand pose naturally serves as a unified representation to bridge human-robot actions. Hence, we investigate an under-explored Vision-Language guided Egocentric 3D Hand Pose Forecasting (VL-EHPF) task, which aims to predict future Ego 3D hand poses from visual observations, a language instruction, and pose states. To overcome the limited field-of-view and highly dynamic motions in the Ego view, we propose a framework dubbed Exo2EgoPose, which innovatively leverages holistic and stable exocentric (Exo) demonstrations as guidance to compensate for partial and dynamic Ego-view cues. Specifically, we introduce a Dual-level Exocentric Reconstruction Module (DERM), which incorporates the paired Exo videos as supervision to reconstruct their video-level and chunked frame-level representations, thereby modeling spatial contexts and temporal dynamics. Then, the Global-to-Local Modulation Module (GLMM) utilizes the reconstructed hierarchical Exo representations for progressive feature refinement via attention mechanisms and adaptive modulation, enabling comprehensive Exo guidance for accurate Ego hand pose forecasting. Extensive experiments on \textit{AssemblyHands}, \textit{Ego-Exo4D}, and our newly constructed \textit{EgoMe-pose} benchmarks show the superiority of our method, which outperforms state-of-the-art methods by a large margin. Moreover, it demonstrates an effective human-to-robot transfer capability and yields improvements on the \textit{CALVIN} dataset. Code will be released.
arXiv:2607.15867v1 Announce Type: new
Abstract: Advanced control of heating and cooling systems can substantially reduce energy costs and pollution. However, real-world adoption of popular algorithms among researchers, such as model predictive control (MPC) and reinforcement learning (RL), remains limited due in part to their high deployment and commissioning costs. Here, we develop two nearly commissioning-free controllers tailored to objectives that depend linearly on the controlled thermal load, such as energy costs and pollution. The controllers require at most two thermal parameters. In representative heating simulations, controller performance is robust to large parameter specification errors, suggesting potential for deployment with no tuning. The controllers maintain good occupant comfort while achieving 43 to 98% (depending on the electricity pricing and controller variant) of the performance improvement achieved by an omniscient policy with perfect model information and forecasts. These results suggest that simple, structure-exploiting controllers may capture most of the attainable value of advanced control while avoiding the data, modeling, tuning, and computational burdens that can arise with conventional MPC or RL.
arXiv:2607.15329v1 Announce Type: new
Abstract: Control co-design considers the physical system and its controller together, enabling the strong coupling between system design and control to be uncovered and exploited. This is especially relevant in aeroelastic flight systems, where structural, aerodynamic, and control design choices jointly determine manoeuvrability and efficiency. This paper presents a model-free nested co-design framework for aeroelastic systems using deep reinforcement learning, in which a design-conditioned control policy is trained with proximal policy optimisation while an outer loop updates a distribution over candidate design parameters. The approach is evaluated on three case studies of increasing complexity: a spring-mass-damper system, a pitch-plunge-flap aerofoil, and a highly flexible high-aspect-ratio glider performing a thermal-soaring mission in a stochastic environment. Across these case studies, the framework is shown to progressively concentrate the design search towards high-performing regions and to outperform policies trained on randomly sampled designs. The results also show that reward shaping plays an important role in enabling stable learning in partially observed and stochastic environments. In the final glider case, the method jointly addresses wing design, flight control, and mission-level behaviour in the presence of aeroelastic coupling and atmospheric uncertainty. These results highlight the potential of model-free co-design for complex aeroelastic systems in which design, control, and mission objectives are tightly coupled.
arXiv:2508.11444v4 Announce Type: replace
Abstract: In a recent paper, Francis, Illickan, Jose and Rajendraprasad showed that every $n$-vertex plane graph $G$ has (under some natural restrictions) a vertex-partition into two sets $V_1$ and $V_2$ such that each $V_i$ is \emph{dominating} (every vertex of $G$ contains a vertex of $V_i$ in its closed neighbourhood) and \emph{face-hitting} (every face of $G$ is incident to a vertex of $V_i$). Their proof works by considering a supergraph $G'$ of $G$ that has certain properties, and among all such graphs, taking one that has the fewest edges. As such, their proof is not algorithmic. Their proof also relies on the 4-color theorem, for which a quadratic-time algorithm exists, but it would not be easy to implement.
In this paper, we give a new proof that every $n$-vertex plane graph $G$ has (under the same restrictions) a vertex-partition into two dominating face-hitting sets. Our proof is constructive, and requires nothing more complicated than splitting a graph into 2-connected components, finding an ear decomposition, and computing a perfect matching in a 3-regular plane graph. For all these problems, linear-time algorithms are known and so we can find the vertex-partition in linear time.
arXiv:2607.16048v1 Announce Type: new
Abstract: Atomic beams for collinear laser spectroscopy are typically produced via charge exchange reactions in an in-beam vapor cell. This process is accompanied by the formation of a considerable amount of longer-lived excited state population, which is not accessible for spectroscopy and also induces fluorescence background to photon detectors. We present an alternative method to produce an atomic beam, consisting exclusively of ground-state atoms. Negative ions are neutralized in-flight by photodetachment and are subsequently used for fluorescence spectroscopy. As a test candidate, a negative boron ion beam was produced in a cesium sputtering source and superimposed with a co-propagating high-power pulsed infrared laser. The neutral atoms were subsequently excited along the $2s^2 2p\,{}^{2\!}P_{1/2,3/2} \rightarrow 2s^2 3s\,{}^{2\!}S_{1/2}$ transitions by a continuous-wave laser at about 250 nm. The statistics and efficiency were limited by the combination of a continuous ion beam with a low-repetition-rate laser, but a clear route for the improvement combining existing techniques is presented, which can then be applied to other elements as well as negative molecular ions.
arXiv:2607.16058v1 Announce Type: new
Abstract: This work presents an open-source tool for automatically generating efficient hardware parsers from high-level specifications. It uses a parsing intermediate representation (PIR) that decouples application-specific frontends from a common register-transfer level (RTL) generation backend. The backend produces optimized, human-readable SystemVerilog, handling FSM generation, byte-alignment, and multi-cycle field straddling for arbitrary datapath widths. The tool also extends pattern matching beyond simple equality checks by introducing custom symbolic tokens to support operations that existing parser generators cannot express, such as range validation, negation, and comparisons against external ports. We demonstrate two end-to-end flows using a P4 frontend for Ethernet protocol parsing and a Snort frontend for network intrusion detection, both using the same unmodified backend. The generated Ethernet parsers achieve up to 226% higher operating frequency and up to 97% fewer FPGA logic resources than prior work. A controlled synthetic study further shows that the tool's hierarchical pattern decomposition yields up to 8x resource utilization reduction over monolithic designs. Our open-source framework enables designers to rapidly implement high-performance, resource-efficient, vendor-agnostic hardware parsers for diverse applications.
arXiv:2508.12620v2 Announce Type: replace
Abstract: Large language models (LLMs) have recently shown impressive results on diverse code-related tasks, benefiting from large-scale training and instruction tuning. However, studies reveal that their grasp of fundamental programming concepts, such as data flow and control flow, remains shallow, leading to fragile performance when code requires deeper reasoning. This limitation restricts the practical adoption of LLMs in real-world software development. To address this issue, this work introduces a counterfactual code augmentation framework combined with concept-aware tuning, designed to guide LLMs toward stronger conceptual understanding. Comprehensive evaluation across multiple models and benchmarks demonstrates the effectiveness of the proposed approach.
arXiv:2607.15430v1 Announce Type: new
Abstract: In this paper, we propose to design, develop, and implement the new DMAIC methodology for Six-Sigma quality management of AM. First, we define the specific quality challenges arising from AM layer-wise fabrication and mass customization (even one-of-a-kind production). Second, we present a review of AM metrology and sensing techniques, from materials through design, process, environment, to post-build inspection. Third, we contextualize a framework for realizing the full potential of data from AM systems, and emphasize the need for analytical methods and tools. We propose and delineate the utility of new data-driven analytical methods, including deep learning, machine learning, and network science, to characterize and model the interrelationships between engineering design, machine setting, process variability and final build quality. Fourth, we present the methodologies of ontology analytics, design of experiments (DOE) and simulation analysis for AM system improvements. In closing, new process control approaches are discussed to optimize the action plans, once an anomaly is detected, with specific consideration of lead time and energy consumption.