Designing functional biological sequences requires navigating vast discrete spaces under strict evolutionary and biophysical constraints. Discrete Flow Matching (DFM) offers a generative framework over such spaces, but existing approaches rely on biologically uninformative couplings and offer limited flexibility for variable-length sequence generation and fine-grained control. We propose a structured coupling that encodes domain-specific preferences among sequence elements, biasing the source distribution toward plausible regions without modifying the flow objective or training procedure. Building on this, we introduce a latent edit-based rate parameterization that models variable-length generation via edit operations conditioned on a shared global latent, akin to a latent variable model, while remaining tractable. We further introduce a latent classifier-free guidance mechanism that steers generation coherently in continuous latent space, along with Dirichlet-prior temperature scaling for test-time control over edit operations. Our method achieves state-of-the-art performance across diverse biological sequence tasks, including density estimation, unconditional and conditional DNA sequence generation, and peptide sequence generation. Our code is available at https://github.com/Aalto-QuML/FlexFlow.
Victim search in wilderness areas is time-critical. Drones enable rapid area search, but scheduling them is challenging because victims may continue moving, detection results are uncertain, and effective decisions must combine structured mission information with unstructured contextual clues and domain knowledge. To address this, we propose an agent-aided drone scheduling system. The system employs an agent at the ground control station (GCS) to coordinate multiple drones over successive search cycles. In each cycle, the agent applies a scheduling policy that combines a large language model (LLM) for reasoning about where to search with an optimizer for determining how to dispatch the drones under energy constraints. Simulation results show that the proposed system consistently completes the victim-search mission, whereas none of the six non-LLM baselines does so and direct LLM scheduling is considerably less reliable. The source code is publicly available at https://github.com/1643204431/llm-uav-victim-search.
Biological intelligence emerges from neural activity that connects sensation, memory and action. Synaptic plasticity, neuronal assemblies and oscillations provide principles for biomimetic computing, but translating them into reusable operations requires a defined computational unit. We introduce Oscillatory Processing Units (OPUs), groups of synthetic neurons sharing a rhythm. Their directed interactions transform incoming activity and prior receiving state into an Oscillatory Occupancy Space (OOS), which becomes available to subsequent computation. We show that source--receiver conditions and oscillatory organization shape these transformations, including their geometry, persistence and historical accessibility. An Atlas of measured relationships supports prospective composition with predicted response classes. Configured circuits perform logical and numerical operations, express associative recall and generate closed-loop decisions with fixed synaptic weights. OPUs and OOS establish a computing substrate for biomimetic intelligence in which experience remains part of the state through which new input becomes action.
This paper introduces a chatbot developed to assist in diagnosing common diseases in tomato plants through natural language interaction. Users can describe symptoms in text. Then, the chatbot responds with the most likely disease, and basic treatment suggestions and related image-based reasoning. The system leverages pre-trained Mistral and Llama 3.2 large language models. The system is also built using LangChain and Streamlit to provide a lightweight and offline-friendly interface. This work targets to support tomato growers a fast and accessible tomato plant health guidance when the technical expertise is not available nearby, especially in the remote areas. The experimental results are promising and encourage further improvement of the chatbot for future real-world deployment.
The digital substrate - data, algorithms, infrastructure, platforms, applications - is being governed without adequate conceptual foundations. The ability and legitimacy required to govern this substrate, and to govern with it, are simultaneously misaligned, contested, and structurally absent. We introduce digital statecraft as the organising concept for this emerging field, arguing that 'digital' reconstitutes the statecraft question rather than merely extending its domain. The concept operates on two dimensions - statecraft over digital systems, concerning the authority and capacity of the state in relation to the digital substrate itself, and statecraft with digital systems, concerning the deployment of algorithmic tools as instruments of governing authority. And it rests on two foundational requirements, technical coherence and legitimate authority, that are genuinely in tension. We derive ten principles of digital statecraft from these foundations, each naming a condition whose absence produces an identifiable and structural governance failure: public interest first, human-machine complementarity, governability by design, systemic coherence, hybrid institutions, adaptive governance, human centricity and civic agency, accountable and traceable authority, judgment across time, and the non-delegable core. This article takes the state as the starting point, the institutional form that developed historically in response to the problem of effective and legitimate public governance, and the only current candidate for which the full set of legitimacy conditions is institutionally available. But the digital statecraft programme holds open a deeper question than just whether states can reform themselves: governing well in the algorithmic age may require rethinking the boundaries, scale, and affiliative basis of statehood itself.
Note on v2 (Nov 2026): errors found in the evaluation code mean the empirical results reported in v1 are not supported; the theoretical corollary is unaffected. Details are in the notice on page 1 and in KNOWN_ISSUES.md in the linked repository. We compare five inference-time safety paradigms (no defense, static steering, CAST, AlphaSteer, probe-gated steering) across seven instruction-tuned models (7-31B) and five attack types (GCG, AutoDAN, DeepInception, prefilling, intent laundering). We derive a corollary: any defense that gates intervention on a single layer's activation alignment with a benign reference (cone, subspace, or null-space) is blind to attacks that craft activations to lie inside that reference, whether checked at prompt time or per token. For prefilling attacks we propose response-time probing, a linear probe on the hidden state at the first generated tokens combined with a halt, and compose this halt with AlphaSteer's null-space steering. We also examine whether MMLU captures steering's utility cost and how the diversity of negative training sets affects probe false-positive rates. Code, attacks, and per-sample results are released.
Note on v2 (Nov 2026): attack success now requires a response judged non-refusing and harmful (v1 counted every non-refusal), so all rates are lower; v1's figures are reported as the non-refusal rate, and with 3 seeds no significance claims are made. Safety alignment in LLMs need not improve monotonically across model generations. Studying four generations of Google's Gemma family (7B-31B) with quality-diversity evolution (MAP-Elites) as an automated red-teaming probe, we find that Gemma 3 (12B) has the highest attack success rate (ASR; response judged non-refusing and harmful): 33.7% +/- 1.3% (mean +/- std, 3 seeds), against 19.0% +/- 10.6% for Gemma 2 and 12.3% +/- 4.2% for Gemma 4. Every Gemma 3 seed exceeds every Gemma 2 and Gemma 4 seed, which we report as directional evidence. Harmful successes rarely transfer between generations (1.4-7.9% ASR when archives are replayed on other generations), lowest on Gemma 4 (1.4-2.8%). Cybercrime is the most exploited category in Gemma 1-3 (47-83%) but falls to 2% in Gemma 4, while misinformation rises from 17% (Gemma 2) to 80% (Gemma 3) and is still 53% in Gemma 4. A second-judge audit (Llama-3.1-70B) confirms 74% of the 8B judge's harmful labels. The non-refusal rate shows the same ordering at 2-3x higher levels and overstates harmful transfer by 6-16x. Code and artifacts: https://github.com/bassrehab/red-queen
Self-attention in Transformers generates dynamic operands that force conventional Compute-in-Memory (CIM) accelerators into costly non-volatile memory (NVM) reprogramming cycles, degrading throughput and stressing device endurance. Existing solutions either reduce but retain NVM writes through matrix decomposition or sparsity, or move attention computation to digital CMOS at the expense of NVM density. We present TrilinearCIM, a Double-Gate FeFET (DG-FeFET)-based architecture that uses back-gate modulation to realize a three-operand multiply-accumulate primitive for in-memory attention computation without dynamic ferroelectric reprogramming. Evaluated on BERT-base (GLUE) and ViT-base (ImageNet and CIFAR), TrilinearCIM outperforms conventional CIM on seven of nine GLUE tasks while achieving up to 46.6\% energy reduction and 20.4\% latency improvement over conventional FeFET CIM at 37.3\% area overhead. To our knowledge, this is the first architecture to perform complete Transformer attention computation exclusively in NVM cores without runtime reprogramming.
Artificial intelligence for power electronics is evolving toward a deeper comprehension of system intrinsic behaviors and dynamics. As a novel integration of neural computation and electric circuits, spike-based neuromorphic computing offers promising potential in high energy efficiency and adaptability for data-driven applications. Building on the latency-coding principles, this article proposes a framework for data-light condition monitoring that relies solely on three-phase AC current measurements. The AC current signals are encoded as membrane potentials of RC-based spiking neurons, which are inherently compatible with spiking neural networks that can effectively facilitate the estimation of device parameters. Case studies are presented to validate the performances, while constraints and further optimizations are also evaluated and discussed, and future extensions of the methodology are provided.
E-commerce supply chains generate large volumes of customer feedback and operational data that remain underutilized in inventory management. Traditional inventory classification techniques such as ABC analysis rely primarily on annual dollar usage, while multi-criteria inventory classification (MCIC) methods extend this but still depend on structured quantitative data and do not utilize unstructured information available in modern supply chains. This paper proposes an LLM-based feature extraction approach that transforms unstructured sources such as customer reviews into quantitative features using vector embeddings and LLM-generated evaluation criteria. The enriched feature set is incorporated into an MCIC framework based on the Hadi-Vencheh model. Experimental results show measurable reclassification effects when customer sentiment signals are included, demonstrating that integrating unstructured data into MCIC can enhance inventory prioritization.
Large language models (LLMs) have made fluent essay writing, code drafting, and quiz answering instantly available to students at every level, from secondary school through graduate study. Many educators do not object to LLM use \emph{per~se}; what they need to detect is the case in which a student pastes the assignment prompt into a chatbot and submits the model's reply verbatim, without engaging with the work. Existing post-hoc AI-text detectors remain unreliable and have been shown to penalise non-native English writers, while output-side watermarks require cooperation from the model provider. We propose an alternative that the educator controls directly: an input-side watermark in which an invisible instruction is embedded inside the visible assignment prompt itself. An LLM that ingests the prompt verbatim quietly reads the hidden instruction and writes a tell-tale signature into its reply, exposing the copy-and-paste pathway specifically. We describe SteganoPrompt, a single-page, zero-dependency web tool that encodes an arbitrary printable-ASCII payload into the deprecated Unicode Tags block (\texttt{U+E0000}--\texttt{U+E007F}). The encoded string is visually identical to the original, survives common copy-paste channels (Word, Google Docs, PDF, Markdown, Slack, e-mail, the major learning-management systems), and is tokenized as ordinary text by most frontier models. We evaluate compliance across eight LLM families and a representative set of educational content channels. To keep deployment transparent, disclosure is built into the tool itself: every encoded brief also carries a brief, always-on academic-integrity reminder for the student, and we set out concrete guidance recommending for its disclosed, ethical use in the classroom. The tool is released under the MIT licence at https://ezharjan.github.io/SteganoPrompt/.
The rapid expansion of the Internet of Things (IoT) has increased the volume of data generated at the network edge, where many devices operate under severe memory, processing, and energy constraints. In this context, Tiny Machine Learning (TinyML) enables machine learning (ML) tasks to be executed on microcontrollers, smart sensors, and other resource-constrained platforms. However, the literature remains heterogeneous regarding the algorithms evaluated, energy-measurement methods, reported metrics, influencing factors, and optimization strategies. This study presents a systematic mapping of energy-aware TinyML research. It identifies the main models and learning approaches, classifies energy-evaluation methodologies and metrics, examines the technical factors affecting energy consumption, and organizes the reported optimization techniques into a structured taxonomy. The findings highlight the frequent use of convolutional and lightweight architectures, direct measurements and analytical estimations, and factors related to model architecture, hardware, communication, numerical precision, and input characteristics. The optimization strategies were grouped into model compression, hardware-aware design, system and communication optimization, and data and processing optimization. The results also reveal limited standardization in experimental platforms, measurement procedures, and reporting practices, limiting reproducibility and direct comparison across studies. This mapping provides an updated overview of the field and guidance for developing and evaluating more energy-efficient TinyML solutions.
Automated feedback on classroom discourse improves instruction in professional and online teaching contexts, and recent work suggests that delivering similar feedback to learners may strengthen these effects by sharing responsibility for participation. Whether this holds in small-group volunteer tutoring has not been tested. We present results from a randomized controlled trial across two four-week SAT mathematics bootcamps on the Schoolhouse.world platform, involving 1,030 volunteer tutors, 7,443 learners and 6,508 recorded sessions. Tutors were randomly assigned to receive no feedback, automated feedback on their own discourse, or that feedback together with participation feedback delivered to their learners under one of two motivational framings. Feedback to tutors increased eliciting of learner ideas, uptake of learner contributions, and learner reasoning by 8-15%, and the pattern replicated across both bootcamps. However, tutor talk time, which averaged 91.5% in the control group, did not change in any condition, and learner-facing feedback did not further change discourse. Interviews with 17 tutors indicate that they adopted the moves the feedback recommended but were limited by learners who did not respond. Interviews with 13 learners suggest why the learner-facing feedback had little effect: it reported a group-level rather than a learner-level statistic, some learners preferred to listen, and learner-to-learner interaction was rare in these sessions. Automated feedback can change how tutors invite learner thinking, but shifting how much learners talk likely requires changes beyond feedback.
Merging companies that each run Salesforce must consolidate their customer data into a single organization while both source systems remain in use. This requires reconciling different schemas and picklists, resolving duplicate customers across and within organizations, preserving relationships between records, and keeping the target current until cutover. This article proposes a reference framework for an autonomous extract-transform-load pipeline for this task. It combines partitioned bulk extraction, declarative mapping templates, rule-based and statistical quality gates, master data management with deterministic and blocked fuzzy matching, constrained loading with per-record error capture, and change data capture guarded by a data-health circuit breaker. Each stage is mapped to Jitterbit Integration Studio constructs. The framework was implemented as an open reference prototype and evaluated on synthetic two-organization scenarios with known ground truth at three scales, from 32,040 to 383,973 source records, over five random seeds. At a preset threshold, hybrid matching reached pairwise F1 scores of 0.918 for accounts and 0.885 for contacts; calibrating the threshold per object would raise these to 0.938 and 0.963. The quality gates prevented all load failures and all silent loss of unmapped picklist values without wrongly flagging any clean record, whereas removing them produced about 1,000 load failures per run. Over five simulated days of hourly incremental synchronization, replayed events were ignored, an injected upstream fault was contained in the batch in which it occurred, and the target agreed with a full rebuild on more than 99.3 percent of field values. Validation on a production deployment remains future work.
Key-value (KV) cache quantization is widely used to reduce LLM inference memory, yet existing evaluations focus only on perplexity and accuracy without assessing the safety impact. In this study, we explore alignment preservation under KV cache quantization. Across eleven instruction-tuned models (3.8B-72B) and five benchmarks (1,894 prompts), we find that low-bit quantization can silently destroy safety alignment: Mistral-7B loses 15.2% of its refusals at only 1.03x perplexity, and no universal safe bit-width exists, with sharp model-specific phase transitions invisible to standard metrics. We identify that the root cause is geometric: safety features occupy a low-dimensional activation subspace $10^2$-$10^3\times$ more vulnerable to quantization noise than the full representation space perplexity averages over. We propose Per-Channel Reduction (PCR), a diagnostic that classifies each model into one of three mechanistic failure modes: outlier-crushes-safety, where safety lives in non-outlier channels collaterally damaged by outlier-driven scale factors; outlier-as-safety, where safety overlaps outlier channels and finer granularity cannot rescue it; and multi-layer dilution, where safety is distributed across many layers and per-layer fixes fail. PCR predicts the correct mitigation direction on all nine primary models and one held-out model from an independent family using 20 calibration prompts. PCR generalizes across unseen prompts, models, and quantizer families, including production quantizers (KIVI: up to 97.2% recovery) and rotation-based quantization (QuaRot-style Hadamard), succeeding where attention-based allocation methods fail. The resulting training-free protocol, requiring ~35 GPU-minutes, recovers up to 97% of lost alignment at under 1% latency and minimal memory overhead, addressing vulnerabilities confirmed in production vLLM serving with FP8 KV cache on NVIDIA GPUs.
We investigate whether open-source LLMs encode a linearly separable truthfulness signal in their hidden states, and at which network depth this signal is strongest. Across three $7$B--$8$B instruction-tuned models (Llama-3.1-8B, Mistral-7B, Qwen2.5-7B) loaded in $4$-bit NF4 quantization, we extract per-layer hidden states on four hallucination benchmarks (TruthfulQA, HaluEval-QA, FEVER, and a controlled synthetic set) and compare four detection approaches: linear and MLP probes, INSIDE EigenScore, self-consistency, and attention entropy. A linear probe on a single mid-network layer achieves $0.904$--$1.000$ AUROC on held-out splits, while sampling-based detectors do not exceed $0.541$ AUROC under the same protocol. The truthfulness signal is approximately linear: MLP probes rarely surpass linear probes by more than $0.01$ AUROC. Peak probing layers fall in a consistent band across model families on natural-language benchmarks -- blocks~$13$--$18$ of~$32$ for Llama and Mistral, and blocks~$19$--$25$ of~$28$ for Qwen. First-block attention entropy provides a complementary signal in knowledge-grounded settings ($0.866$--$0.941$ AUROC on HaluEval-QA) at no additional inference cost. The low discriminability of sampling methods under this protocol reflects a structural mismatch between paired-label evaluation and the information these methods access, rather than an inherent limitation of those methods. Code and data are released for full reproducibility on a single $8$\,GB GPU.
Trainable-frequency quantum models train the prefactors of their encoding gates so that the circuit's frequencies can move onto the target's. From the unit start of prior work, training rarely reaches a target more than one resolution step away. We show that the target-driven part of the prefactor gradient is a weighted sum of slopes of the overlap $\sin u/u$, where $u$ is the model-target frequency distance scaled by the window half-width. This gives each frequency a reach: under gradient flow, an isolated real frequency whose coefficient adapts optimally reaches its target from within half a step, for every target at least one step from zero frequency; with a fixed coefficient, a phase-dependent barrier can stop it. A ternary prefactor grid places a model frequency within half a step of every band frequency, with logarithmically many encoding gates. Single-frequency circuits trained with Adam and jointly trained amplitudes behave as predicted: inside the first side lobe, a frequency starts toward its target exactly when in phase with it in 98% of runs, and only 11 of 768 frequencies starting beyond the lobe reached their target. With several frequencies, the grid start raised successful fits on synthetic targets from 69% to 84% against the best random start. With many target frequencies, least squares on the fixed grid fits near-perfectly, but the circuit does not reach the coefficients; training the prefactors narrows this shortfall, but with 8 and 24 target frequencies, training the fixed grid ten times longer fitted better. On real series, training the frequencies improved held-out $R^2$ at equal budget by 0.010--0.045 on four of seven series, but a classical screen did not separate these from the others. A classical model with independently trainable frequencies shows the same dependence on the start; our results describe how to train such models, not an advantage of the circuit.
The release of SAM 3D Body is a recent development in human mesh recovery, demonstrating improved performance in producing clean, topologically coherent meshes from single images. By leveraging the Momentum Human Rig (MHR), it achieves robustness to occlusion and diverse poses. However, our evaluation reveals a specific and consistent limitation: the model struggles to reconstruct detailed anthropometric deviations, particularly in populations exhibiting distinctive morphological alterations such as geriatric muscle atrophy, scoliosis, or pregnancy, even when these features are prominent in the input image. In this paper, we investigate this phenomenon not as a failure of the model's capacity, but as a byproduct of the "perception-distortion trade-off". We posit that the architectural reliance on the low-dimensional parametric MHR representation, combined with semantic-invariant conditioning (DINOv3) and annotation-based alignment, creates a pervasive "regression to the mean" effect. We analyze these mechanisms to understand why individual biological details are smoothed out. Furthermore, we state our contributions by proposing specific, constructive pathways for future work, such as implicit-explicit hybrid representations and Medical-in-the-Loop alignment, to extend the baseline performance of SAM 3D Body into the high-precision medical domain.
Decoding visual stimuli from the spiking activity of retinal ganglion cells (RGCs) informs both our understanding of the retinal code and the design of visual prostheses. Natural vision is dynamic, yet state-of-the-art decoders mostly reconstruct single frames with artificial neural networks that process dense, continuous-valued signals rather than the retina's sparse, asynchronous spikes. We asked whether dynamic scenes can be reconstructed and tracked entirely within the spike domain, and at what computational cost. We built an end-to-end spiking neural network (SNN) whose neurons integrate the recorded RGC spikes over time. Populations of up to 1836 RGCs were recorded ex vivo from mouse retina on high-density microelectrode arrays while flashed images, moving bars, drifting gratings and randomly moving handwritten digits were presented. For moving digits, one shared spiking encoder fed two readouts, one reconstructing the image and one decoding the digit's position. Compute energy was estimated from counted synaptic operations. The SNN decoded what and where at once. It reconstructed the stimuli with a structural similarity (SSIM) of up to 0.93, refining the reconstruction as spikes accumulated, and tracked a randomly moving target with a mean error of 1.33 pixels on a $32\times32$ grid. Its fidelity nearly matched that of the best of 17 larger spiking and artificial decoders (5.9 million to 1.7 billion parameters, versus 5.4 million), and at a mean firing rate of 14.2% it needed an estimated $\sim$70-fold less compute energy than a dense artificial network of the same architecture. Dynamic visual scenes can thus be decoded from retinal population activity in the spike domain, with a compact network and at a compute cost compatible with implantable and wearable neural interfaces, supporting spike-native decoders for visual prostheses.
Large language model (LLM) assistants increasingly advise backers on which crowdfunding campaigns to support. LLMs have known race and gender problems, and it is unsettled whether they reproduce, erase, or reverse the discrimination documented in entrepreneurial finance. We ask when, and through which decision process, LLM assistants use founders' gender and ethnicity. In a pre-specified choice-based conjoint experiment, seven LLMs chose among five synthetic campaigns with independently randomized quality signals and founder names, under three backer personas and a persona-free control (40,248 responses). By default, founder names barely matter: under budget-conscious and delivery-focused personas, gender and Black-name effects are statistically equivalent to zero. When the backer wants to support underrepresented creators, the models favor Black-signaled (+8.1 percentage points; positive in all seven models) and female-signaled (+5.1) founders. They do so most strongly among near-tied campaigns and for the founder who alone carries the signal, and they almost never name gender or ethnicity in their reasons. The results fit an agency account of algorithmic discrimination: an aligned assistant takes its demographic preferences from its user rather than from its training data.