Computational Engineering
IC-ThermBench: An Open, Progressive Benchmark for Generalizable 2.5D/3D-IC Thermal Learning
David Hang, Wenkai Yang, Kuiye Ding et al. (+2 more)
Standardized
benchmarks are fundamental to reliable progress in AI for EDA, including learning-based thermal modeling. However, existing thermal prediction studies often rely on different
datasets, simulators, data splits, preprocessing pipelines, and metrics, while most
datasets and implementations remain unavailable, making fair and reproducible comparison difficult. We introduce IC-ThermBench, an open and progressive
benchmark that combines established 3D-IC steady-state, transient, and industrial package tasks with a new 50,000-sample 2.5D chiplet extension designed to evaluate progressively broader represented physical variation and cross-package OOD transfer. Five Generalization Scopes cover 3D-IC fixed-design prediction, Within-Family Generalization under layout, material, and boundary-condition variation, and Cross-Package OOD transfer to unseen package systems. We evaluate eight representative baselines under common data, splits, labels, and metrics. Performance degrades gradually from S2 to S4 as represented physical support broadens, but Cross-Package OOD produces a much sharper degradation: the best RMSE and MAE increase from 1.216 and 0.938~K at S4 to 15.99 and 15.00~K at S5, respectively. With only 10 labeled samples per OOD case, target-domain adaptation reduces the best MAE to 2.60~K. IC-ThermBench further provides a unified generation, training, inference, and evaluation pipeline, enabling reproducible and fair comparison of existing and new thermal predictor.
Read the full abstract ▾
All authors David Hang, Wenkai Yang, Kuiye Ding, Haiyang Xin, Jacky Wei
Computer Vision
Boot-and-Feedback Framework for Generalist-Expert Model Collaboration in Breast Ultrasound Diagnosis
Ming Cheng, Hongyu Sun, Zhaolin Chen et al. (+3 more)
Breast ultrasound (BUS) is widely used for breast cancer diagnosis yet remains operator-dependent. While deep learning shows promise, ensuring diagnostic reliability and interpretability is challenging. Recent
Multimodal Large
Language Models (MLLMs) often generate spurious descriptions due to limited domain knowledge, which mislead downstream expert models and compromise clinical validity. To address these challenges, we propose the Boot-and-Feedback (BooF) model collaboration framework for synergistic MLLM-expert interaction. Specifically, in the Boot Stage, the MLLM is guided by the BI-RADS lexicon and preliminary benign-malignant
vision-expert predictions, enabling it to transfer general
reasoning to BUS analysis while avoiding
hallucinations. Subsequently, the Feedback Stage integrates these descriptions with visual features via a lightweight
Attention-Gated Cross-Modality Fusion Module. This allows the expert to leverage textual feedback while adaptively filtering noise. Extensive experiments on multiple BUS
datasets demonstrate that BooF substantially outperforms state-of-the-art methods in terms of diagnostic accuracy and interpretability.
Read the full abstract ▾
All authors Ming Cheng, Hongyu Sun, Zhaolin Chen, Jun Liu, Hossein Rahmani, Qiuhong Ke
Robotics
Safety-aware Model Predictive Path Integral Control with Signal Temporal Logic
Yiqi Zhao, Taekyung Kim, Hideki Okamoto et al. (+4 more)
Safety-aware motion planning remains a challenge in robotics, especially when missions are time-critical and are under complex specifications. In this paper, we propose safety-aware-stl-mppi, a computationally efficient sampling-based receding-horizon planning framework designed to promote satisfaction of constraints expressed in Signal Temporal Logic (STL). Our approach encodes discrete-time STL formulas into candidate time-varying control barrier functions (CBF), which are integrated into a model predictive path integral (MPPI) controller. Our method inherits the benefits of low computational cost from an efficiently parallelizable sampling based planner and utilizes CBF for constraints expressed in STL. We compare against several MPPI baselines using four artificial Mars Rover planning case studies with a diverse environment and cost setups, where we show our method consistently achieving high safety and efficiency. We show a quadcopter planning experiment with NVIDIA Isaac Lab.
Read the full abstract ▾
All authors Yiqi Zhao, Taekyung Kim, Hideki Okamoto, Bardh Hoxha, Jyotirmoy V. Deshmukh, Lars Lindemann, Georgios Fainekos
Artificial Intelligence
Giraffe: A Mapping Architecture from Hidden Text Representations to Visual Embeddings for Efficient Graphic Design
Nejla Ghaboosi
Multimodal large
language models (MLLMs) have made significant progress in understanding and interpreting mul- timedia content. However, their ability to generate me- dia remains limited. Recent approaches have attempted to bridge this gap by translating the hidden representations of
token sequences into the
embedding space of visual models or directly into raw image data. However, these methods often represent each image using multiple specialised to- kens which significantly increases the input length. This be- comes a major limitation for tasks such as graphic design generation where the output typically involves a seamless blend of thousands of
tokens across text, multiple images, and layout information. To address this challenge, a novel architecture is proposed that maps hidden
token represen- tations to the
embedding space of visual models, such as CLIP ViT-L/14, using a single [IMG]
token per image. The architecture employs two shallow MLP blocks, each with a separate compression module followed by a shared expan- sion module, trained with six distinct loss functions. One block aids the other during training and is omitted during inference, resulting in a lightweight solution. Strong perfor- mance is demonstrated in both image-to-design and text-to- design generation tasks.
Read the full abstract ▾
Computation and Language
Investigating Knowledge Transfer Across Interactive Dialogue Games
Filippo Momentè, Mir Nafis Sharear Shopnil, Andrea de Varda et al. (+5 more)
Dialogue games represent a challenging setting where complex cognitive skills are required to accomplish tasks while coordinating with other players. Considering that
language represents an interface for both understanding the game rules and executing actions, it is reasonable to assume that training on a specific
language game will enhance specific capabilities that might be relevant for other tasks as well. Motivated by this rationale, in this paper, we investigate how knowledge transfers across different dialogue games. We study transferability by finetuning LLM models on games from the clembench suite (Chalamalasetti et al., 2023) and performing two analyses: i) we derive a task-transferability graph using a binary integer optimization program from Zamir et al. (2018), using task performance as the main metric; and ii) we compute task vectors (Ilharco et al., 2022) for each game to study similarities across finetuned models and their task transferability. In our first analysis, we find that some games benefit more from transfer than finetuning, and that the visuospatial family (e.g., exploration games) transfers best. With our task vector analysis instead, we find that similarity-based approaches capture game-role relationships but almost no transferability patterns, suggesting that more complex metrics are required.
Read the full abstract ▾
All authors Filippo Momentè, Mir Nafis Sharear Shopnil, Andrea de Varda, Pavel Merinov, Raffaella Bernardi, Oswald Lanz, Alessandro Suglia, Alessandro Torcinovich
Human-Computer Interaction
When LLMs Slow Down: How Environmental Impacts Mediate University Students' LLM Usage
Hyeonwook Kim, Xuesi Chen, Alex Cabral et al. (+3 more)
Large
Language Models (LLMs) are increasingly being embedded into all facets of society, from search to education, industrial, and financial applications. These systems' carbon and water footprints raise important sustainability concerns, particularly with adoption rates exceeding 80% among university students, despite limited insight into the environmental impacts of individual usage. Eco-feedback interfaces offer a promising approach to encourage more sustainable behaviors, yet their role in shaping LLM users' sustainability awareness and decision-making remains underexplored. We design and deploy the interface that visualizes latency-carbon trade-offs during live LLM interactions. We study its use with undergraduate computer science students (N=89, ages 18-24), enrolled in a computing ethics course, providing an empirical look at how a technically sophisticated and values-oriented user population responds to sustainability-aware AI interfaces. We found that the likelihood of choosing the eco-feedback system significantly decreased as perceived response latency increased (p < .001), while users' willingness increased when they recognized the carbon-saving impacts (p < .01). Also, students with stronger eco-mindedness demonstrated higher baseline willingness to adopt lower-carbon modes and reported increased awareness of the environmental impacts of LLM use, though this effect diminished as latency increased. These results position eco-feedback interfaces as a promising sustainability intervention and highlight their potential as an educational opportunity to promote more sustainable LLM use among university students and beyond.
Read the full abstract ▾
All authors Hyeonwook Kim, Xuesi Chen, Alex Cabral, Cindy Kaiying Lin, Udit Gupta, Josiah Hester
Human-Computer Interaction
Who Chooses How Preferences Are Aggregated? Auditing Aggregation-Rule Authority in LLM-Based Group Recommendation
Yuxuan Du
AI systems increasingly make joint recommendations for users with conflicting preferences. However, when reasonable aggregation rules support different actions, a further question arises: who may choose how those preferences are combined? We study this interaction-level problem as aggregation-rule authority. Using synthetic preference profiles and profiles constructed from empirical ratings, we conduct a controlled behavioral audit of three LLMs under three authority conditions: unspecified, explicitly retained by users, and delegated to the model. In cases where two witness rules supported different actions, models almost never committed when users retained authority, but committed in every delegated case. All three models executed both witness rules perfectly when directly instructed. Yet when authority was unspecified or delegated, their aggregation-consistent outcome distributions differed across models and preference settings. Together, these results separate rule-execution capability from aggregation-rule authority: delegation assigns the model discretion to resolve the aggregation choice, but does not determine which collective outcome follows.
Read the full abstract ▾
Cryptography and Security
RAGSentinel: Certifiable Geometric Consensus for Robust Retrieval-Augmented Generation
Yueyang Quan, Anjun Gao, Yufei Xia et al. (+2 more)
Retrieval-augmented generation (RAG) improves the factuality of large
language models by grounding responses in external documents, but it also exposes a critical security vulnerability:
adversarial documents injected into the knowledge database can enter the context window and steer the model toward targeted incorrect answers. Existing post-
retrieval defenses rely on instruction following, parametric knowledge, or text-level consistency, all of which can be imitated or optimized against by adaptive attackers. We propose RAGSentinel, a training-free, label-free defense for black-box RAG systems. RAGSentinel uses a surrogate encoder to measure query-conditioned hidden-state shifts induced by retrieved documents, removes shared topic directions, and filters poisoned documents as
geometric outliers from a robust majority consensus. We prove that, under an honest-majority assumption and a representation-level separation condition, RAGSentinel exactly recovers a poison-free majority-sized context. Experiments across three question-answering
datasets, three LLM families, and multiple poisoning attacks show that RAGSentinel consistently achieves low attack success rates while preserving competitive accuracy and remaining effective against adaptive attacks with full pipeline knowledge.
Read the full abstract ▾
All authors Yueyang Quan, Anjun Gao, Yufei Xia, Minghong Fang, Zhuqing Liu
Artificial Intelligence
More GPUs or a Smaller Cache? Tensor Parallelism versus KV Compression for Memory-Bound LLM Serving
Srikanta Datta Tumkur, Mehar Simhadri, Anshu Bansal et al. (+5 more)
When an LLM serving deployment runs out of KVcache room, there are two well-established ways out. Tensor parallelism shards the weights and the KV cache across two, four, or eight devices, buying
memory headroom at the price of an all-reduce on every layer and a hardware bill that grows with the device count. The
algorithms community shrinks the cache in place, with KV quantisation and eviction keeping a single GPU and spending a little quality instead. Compression papers report
memory ratios, parallel-scaling papers report throughput curves, and almost nobody puts the two on the same cost axis. We place tensor-parallel configurations (degree 1 to 8) and KV-compressed configurations (16/8/4-bit, keep-ratios down to 0.25) on one costnormalised axis, cost per million
tokens against latency, using a profiled simulator calibrated on A100, A40, and H100 hardware, and we go looking for the cost-equivalence crossover. We do not find one. Across two models (Llama-2 at 7B and 70B), three GPU types, and every level of
memory relief we could construct, compression is cheaper by 1.20x to 2.00x. A 7B model on an 80 GB device cannot exhaust its KV budget within its own context window, and the boundary that decides between the strategies is model size relative to device
memory, at roughly 36B parameters for an 80 GB card. Below that wall, compression dominates and extra GPUs are largely wasted spend; above it, tensor parallelism stops being a choice and becomes an entry ticket: Llama-2-70B is infeasible on one A100 at any KV setting, because the binding resource is weights, which KV compression does not touch. Tensor parallelism is the only lever that improves latency (compression makes per-
token latency worse, by 8 to 93%, through batching contention), while compression is the only lever that multiplies capacity per dollar (16.5x, against 1.21x for an eightfold spend on GPUs).
Read the full abstract ▾
All authors Srikanta Datta Tumkur, Mehar Simhadri, Anshu Bansal, Jay Iyer, Sai Pavan Kumar, Sai Kapil Kumar, Ramesh Nampelly, Raj Dandekar
Software Engineering
Evaluating Language Models on Cross-Language Code Functional Equivalence
Hui Sun, Anderson Uchôa, Rohit Gheyi et al. (+1 more)
Background: Large
Language Models (LLMs) have demonstrated strong performance across a variety of code-understanding tasks, leading many to believe that they can reason about program semantics. However, existing evaluations primarily focus on single-
language settings or rely on synthetically generated code, raising concerns about whether current results reflect true semantic understanding. Aims: We investigate whether LLMs can accurately judge functional equivalence across different programming
languages in human-written code, a setting that requires deeper
reasoning beyond superficial similarity. Method: We introduce PolyHuman, a
dataset of human-written programs in CPP, Java, and Python. Using this
dataset, we evaluate intra- and inter-
language equivalence detection across open-weight and proprietary LLMs, selecting GPT-o4-mini as a representative model to assess stability. We then manually analyze 81 cases of systematic disagreement in which models incorrectly judge functional equivalence, examining the code logic and the generated Chain-of-Thought
reasoning. Finally, we categorize these failures and compare them across GPT-o4-mini, Claude-Opus-4.7, and Gemini-3-Flash to determine whether they reflect model-specific issues or broader limitations of state-of-the-art LLMs. Results: We identify a difficulty-dependent breakdown in equivalence judgment (harder problems make the model increasingly prone to misclassifying non-equivalent code as equivalent), a model-specific sensitivity to programming
language for the best-performing model (particularly a more conservative behavior on Python), and a partial reliance on similarity-based cues. GPT-o4-mini also shows substantial run-to-run instability under identical settings, indicating inconsistent rather than absent capability. Conclusions: Current LLMs do not reliably capture functional equivalence within or across
languages.
Read the full abstract ▾
All authors Hui Sun, Anderson Uchôa, Rohit Gheyi, Wesley K. G. Assunção
Cryptography and Security
NeuronGuard: Robust LLM Safety Alignment via Ablation-Aware Safety Signal Redistribution
Anjun Gao, Yueyang Quan, Yufei Xia et al. (+2 more)
Safety
alignment in large
language models (LLMs) remains brittle against a growing spectrum of attacks. Jailbreak attacks bypass safety mechanisms through crafted
prompts, while
neuron-level attacks directly prune safety-critical
neurons post-deployment. Both exploit a common weakness: safety-relevant information concentrates in a sparse
neuron subset. We present NeuronGuard, a fine-tuning-stage defense that simultaneously hardens LLMs against both attack classes by redistributing safety signals across a broader set of
neurons. NeuronGuard dynamically identifies safety-critical
neurons via periodically refreshed per-layer linear classifiers, forces refusal behavior under deliberate
neuron ablation, and applies KL-divergence regularization for distributional consistency. A randomized gradient projection strategy preserves downstream task utility by resolving conflicts between the defense and task objectives. We provide a formal guarantee that NeuronGuard strictly reduces the attack success rate (ASR) upper bound, and experiments across three LLMs, six state-of-the-art attack strategies, and
multimodal settings confirm near-zero ASR while maintaining task accuracy, including against white-box adaptive adversaries.
Read the full abstract ▾
All authors Anjun Gao, Yueyang Quan, Yufei Xia, Zhuqing Liu, Minghong Fang
Artificial Intelligence
Shengxin Zhang, Xiaomin Wu, Xiyang Wu et al. (+1 more)
Test-time
reasoning methods such as iterative refinement, decomposition, and repeated sampling are often evaluated in isolation, making their gains difficult to compare across models,
benchmarks, and evaluation pipelines. We introduce a unified view of these methods as recursion operators over an
agent's
reasoning trace: GROW, which deepens a single
reasoning path; PRUNE, which decomposes and recomposes the problem; and BRANCH, which samples alternative
reasoning paths and selects among them. We evaluate all three operators against a single-pass chain-of-thought baseline under a shared harness with identical
prompts,
token budgets, and grading code. Across five
benchmarks and three frontier models, comprising 14 model-
benchmark settings, 49,327 graded items, and 151,876 model calls, BRANCH improves accuracy in all 14 settings by an average of 5.98 percentage points and is the best-performing operator in 12. In contrast, GROW yields a mean gain of 2.18 points and degrades performance in two settings, while PRUNE improves accuracy by 0.94 points on average. Analysis shows that BRANCH's advantage arises not only from exploring multiple
reasoning paths, but also from recovering from truncation: its gains strongly correlate with the baseline rate of empty, budget-exhausted outputs (r = 0.72). These results weaken the hypothesis that different problems require routing among test-time
reasoning operators; at this level of abstraction, repeated branching is consistently dominant. Finally, we show that unpaired evaluation and treating scoring-pipeline failures as model errors can materially change, and even reverse, comparative conclusions, motivating paired scoring as a standard protocol for test-time-compute evaluation.
Read the full abstract ▾
All authors Shengxin Zhang, Xiaomin Wu, Xiyang Wu, Jing Xie
Neural and Evolutionary Computing
Trust, but Verify: Rigorously Profiling Best-Effort High-Performance Computing for Digital Evolution
Matthew Andres Moreno, Santiago Rodriguez Papa, Charles Ofria et al. (+2 more)
Developments in high-performance computing (HPC) technology continue to drastically increase quantities of available processing power. In the context of digital
evolution, this explosive growth offers opportunities to advance both hypothesis-driven explorations of multi-scale biological phenomena and application-driven evolutionary optimization targeting hard problem domains. A particular opportunity arises from emerging next-generation AI/ML hardware accelerator platforms, such as the 880,000-processor Cerebras Wafer-Scale Engine (WSE). Such hardware, however, constrains on-device data storage and movement — a challenge compounded by vulnerability to failures arising over numerous device components. Best-effort relaxations that depart from a traditional deterministic computing paradigm can help accommodate such constraints, but complicate reproducibility and risk introducing artifactual biases. We explore these concerns, developing a framework to measure runtime behavior of best-effort code and examining case studies of best-effort computing in digital
evolution projects. The first case study applies best-effort CPU-cluster multiprocessing to a multicellularity
evolution model, which provides 92% scaling efficiency at 64 processes ($2.1\times$ speedup) and exhibits robust median quality of service, even under hardware anomalies. The second case study examines WSE-based simulations, demonstrating best-effort strategies to track spatiotemporal population history — through sparse, asynchronous device-to-host sampling that tolerates hardware faults. In sum, across potential forms and scopes of best-effort relaxation, we argue that digital
evolution is uniquely positioned to contribute in developing post-deterministic HPC paradigms.
Read the full abstract ▾
All authors Matthew Andres Moreno, Santiago Rodriguez Papa, Charles Ofria, Luis Zaman, Emily Dolson
Software Engineering
The Empire, Long Divided, Must Unite: Architectural Convergence in Three LLM Agent Harnesses
Dai Jiahong
An
agent harness is what turns a
language model into an autonomous
agent: the surrounding code that builds the model's context, mediates its tools, runs the loop, and persists state across a long-horizon run. This layer, not the model it wraps, is increasingly the binding constraint on
agent behaviour. We present a source-level, multi-case study of three open coding-
agent harnesses built from deliberately opposing philosophies: LangChain's deepagents (batteries-included), Earendil's pi (radical minimalism), and DeepSeek's dsh (everything-is-a-plugin). Reading each at a pinned commit and following its commit history, we find that the two mature harnesses have travelled in opposite directions (deepagents subtracting authored scaffolding, pi accreting durable infrastructure), yet converged toward one architectural middle form of five recurring elements: a commoditised loop, an append-only replayable session record, model quirks kept as data, progressive disclosure of context, and explicit extension seams. A third harness, read afterward as a held-out check, exhibits all five, and in one seam reuses another's implementation outright. We therefore do not claim independent invention, and decompose the convergence into parallel discovery,
diffusion, and literal reuse. Finally, one load-bearing dimension shows no convergence, and indeed no presence: external verifiability, a tamper-evident record an outside party can check without trusting the runtime. We read this absence not as an oversight but as a predictive gap, the next axis on which harnesses for provenance-sensitive domains will differ.
Read the full abstract ▾
Cryptography and Security
STAIN-FL: Stealthy Targeted Attack Injection with Contextual Triggers in Federated Learning
Ashlinder Kaur, Purnima Murali Mohan, Zengxiang Li et al. (+1 more)
Federated video anomaly detection trains model collaboratively without sharing raw surveillance footage, but limited server-side visibility lets compromised clients to inject
backdoor via malicious updates. This paper introduces STAIN-FL, a stealthy targeted
backdoor attack injection framework that uses naturally occurring surveillance conditions, including low-light scenes, indoor settings, and crowd density, as contextual triggers. STAIN-FL combines anomaly-to-benign label manipulation with gradient masking over least-updated coordinates to preserve clean accuracy while inducing trigger-conditioned misclassification. We evaluate STAIN-FL on UCF-Crime using 1024-dimensional I3D features in a non-IID four-client multi-agency setting, comparing FedAvg and FedProx under sparse and continuous attacks. Results show that sparse attacks have low-detectability, operationally significant attacks rather than high-intensity attacks: they keep the mean clean-accuracy drop below $2\%$, yet still misclassify more than half of triggered anomalies at peak
backdoor accuracy under FedAvg ($56.7\%$) and FedProx ($54.2\%$). Under FedAvg, the sparse
backdoor remains above the $25\%$
backdoor-accuracy threshold for an average of $336$ post-attack rounds, highlighting the persistence risk of contextually triggered attacks in surveillance systems.
Read the full abstract ▾
All authors Ashlinder Kaur, Purnima Murali Mohan, Zengxiang Li, Tram Truong-Huu
Computational Geometry
On Angle-optimization and Simplification of Degree-1 Homology Representatives
Emerson G. Escolar and Yuta Shimada
In topological data analysis, in particular persistent homology analysis, extracting "optimal" representatives for homology classes is crucial for identifying
geometric regions of interest. In prior work, optimality is defined in terms of minimizing length or volume. In this work, we restrict our
attention to a single homology class in degree $1$ and introduce the total absolute
curvature of cycles as the cost function. We show that this cost function, based on angles between edges of cycles, penalizes departures from planarity, convexity, and simple-ness of the cycle representative. We formulate the "angle-optimal homologous cycle problem", recast it as a binary quadratic optimization problem, and show the results of experiments on artificial toy data.
Read the full abstract ▾
Computer Vision
Luce: Relightable Gaussians for 3D Asset Generation
Mayank Singh, Michele Stoppa, Alvise Memo et al. (+7 more)
High-fidelity image-to-3D generation requires a 3D representation that captures both geometry and appearance. However, preserving fine detail across the physically based rendering (PBR) modalities needed for relighting remains challenging. To address this, we propose Luce, a 3D representation that unifies geometry and PBR materials within a voxelized
multimodal Gaussian cloud, using dedicated Gaussian primitives for albedo, metallic-roughness, and surface normals. A variational autoencoder compresses this representation into a unified material-aware latent space. A rectified-flow
transformer generates this latent from a single image using multi-layer features from a pretrained image encoder that preserve both semantic context and fine spatial detail. The latent is then decoded into relightable PBR Gaussians and an optional textured mesh with a tangent-space normal map. On Toys4K, Luce achieves state-of-the-art single-image-to-3D generation, improving FID by 28% over the strongest baseline. We further evaluate Luce on a
benchmark of AI-generated images depicting diverse subjects and materials, where it improves the CLIP image-
alignment score over the best baseline (0.8519 vs. 0.8299). Luce generates relightable, geometrically accurate, and materially faithful assets that preserve fine details such as text, logos, and inscriptions.
Read the full abstract ▾
All authors Mayank Singh, Michele Stoppa, Alvise Memo, Rui Yu, Sree Harsha Kalli, Srimanth Gunturi, Muhammad Ahmed Riaz, Behrooz Shahsavari, Waleed Abdulla, David E. Jacobs
Artificial Intelligence
More Rejective, Not More Discriminative: The Unit of Verification in Pre-Execution LLM Oversight
Yuchen Han, Cheng Yan and Wuyang Zhang
Pre-execution oversight is core to trusted monitoring in AI control: a fallible LLM monitor vets planned actions before irreversible execution. Over-blocking forfeits usefulness and pressures deployers to disable it. Every protocol must fix a unit of verification: how many actions one call reviews. Existing designs take the unit as given; its effect on fallible monitors is unmeasured. Natural traces cannot isolate it: review length co-varies with error type and position. Catch alone misleads: rejecting everything catches everything. Measuring this needs boundary variation alone and a matched clean control. We introduce the twin-prefix framework, which supplies both. Each gold plan yields a prefix with one injected, environment-accepted error and a clean twin differing in one write. Judging each pair at five nested lengths ties verdict changes to the unit alone. Discrimination is scored by pre-registered informedness, catch minus false rejection. Longer review raises catch; false rejection climbs in lockstep. Informedness peaks at one or two actions for all six judges in both domains: longer windows make zero-shot monitors more rejective, not more discriminative. Replaying withheld observations traces the failure largely to observation deprivation. Safety cases should state the unit and co-report the clean series. Our framework is the first controlled, pre-registered instrument for this choice and never reads catch alone. Our calibrated short unit recovers up to 0.95 informedness over eight-action review, and no tested label-blind policy consistently beats it.
Read the full abstract ▾
Machine Learning
CoDrift: Compositional Drifting for Offline Reinforcement Learning
Xiewei Ni, Ruofeng Mei and Xiangyu Xu
Offline reinforcement learning is intrinsically multi-objective: a policy must remain compatible with the behavioral support of a fixed
dataset while preferentially selecting high-value actions. We recast these objectives in a common form by viewing each as an action-space motion field that specifies how generated actions should move. This perspective enables heterogeneous learning objectives to be combined directly through field composition. Inspired by drifting models, we propose CoDrift, a compositional framework for one-step generative policy learning. CoDrift combines three objective-level fields into a unified policy field. The conditional field preserves state-dependent behavioral structure, while the marginal field pools actions across states to provide a more stable generative signal in the single-positive-sample regime of continuous-control offline RL. The value field moves generated actions toward higher-value regions. The composed field is absorbed into a
stochastic generator that produces an action with a single forward pass at deployment. We evaluate CoDrift on 73 tasks from OGBench and D4RL in both offline and offline-to-online settings. CoDrift compares favorably with state-of-the-art methods and achieves the best average rank in both settings.
Read the full abstract ▾
Machine Learning · Cross-listed from stat.ML
Generalization, memorization, and overfitting for diffusion models trained in the lazy high-dimensional regime
Hugo Latourelle-Vigeant, Sinho Chewi, Aram-Alexandre Pooladian et al. (+2 more)
Modern score-based generative models have achieved remarkable empirical success in high-dimensional tasks such as image, audio, and
video synthesis. These models reduce distribution learning to a sequence of regression problems that, if solved exactly on finite data, would ultimately reproduce the training samples. Their ability to generalize must therefore arise from the implicit or explicit regularization during training. In this work, we develop a generative counterpart to the theory of benign overfitting and algorithmic regularization for overparameterized neural networks in the supervised lazy-training regime. We study denoising score matching in a vector-valued reproducing kernel
Hilbert space with an inner-product kernel. In the proportional high-dimensional regime $n\asymp d$, we derive exact risk trajectories under gradient flow training. These trajectories exhibit three phases governed by qualitatively distinct
estimators: a spectral
estimator that generalizes, a pure-noise score with localized peaks that interpolate the training objective, and an empirical Bayes
estimator that memorizes the data. We then analyze how these
estimators combine along the reverse-time SDE and characterize the distribution of the resulting samples. The analysis reveals familiar mechanisms from supervised learning, including kernel linearization and self-induced regularization from the nonlinear part of the kernel, but also reveals a distinct phenomenology specific to generative modeling.
Read the full abstract ▾
All authors Hugo Latourelle-Vigeant, Sinho Chewi, Aram-Alexandre Pooladian, John Sous, Theodor Misiakiewicz
Computers and Society
An Echo Chamber of One: Should AI Psychosis Be a Distinct Clinical Entity?
Joshua Au Yeung, Hamilton Morrin, Vincent Ng et al. (+2 more)
"AI psychosis" has entered public and clinical discourse as a label for the onset or exacerbation of psychotic symptoms, most commonly delusions, following intensive interaction with large
language model (LLM)-based chatbots. Current evidence is limited to media reports, case reports, and early observational data, yet the scale of potential exposure is considerable, and public concern has prompted responses from industry and regulators. We examine whether AI-associated psychosis warrants recognition as a distinct clinical entity, drawing on clinical and technical viewpoints. We outline the proposed mechanism: LLM sycophancy, a tendency to agree with and flatter users that is reinforced through preference-based fine-tuning, combines with increasingly anthropomorphic design to create a bidirectional "echo chamber of one" capable of amplifying and co-constructing unusual beliefs. We then weigh arguments for and against nosological recognition. Potential benefits include improved case identification, tailored interventions, standardised research criteria, post-
market surveillance, and pressure on developers and regulators to act. Reasons for caution include the risk of prematurely reifying a syndrome from anecdotal evidence, the possibility that existing diagnostic constructs already accommodate AI use as a contributing factor, the unproven causal claim in the term itself, stigma, and the risk that a psychosis-centric label obscures a broader spectrum of AI-associated mental health harms. We conclude with recommendations for clinicians, developers, researchers, and regulators, including a "technological history" in psychiatric assessment, pre-deployment benchmarking for sycophancy and delusion reinforcement, and post-deployment surveillance. Regardless of whether AI-associated psychosis earns a place in psychiatric nosology, the phenomenon it describes demands coordinated
attention now.
Read the full abstract ▾
All authors Joshua Au Yeung, Hamilton Morrin, Vincent Ng, Zeljko Kraljevic, Richard Dobson
Machine Learning
MnemoDyn: Learning Resting State Dynamics from 40K FMRI sequences
Sourav Pal, Viet Luong, Hoseok Lee et al. (+5 more)
We present a dynamical-systems based model for resting-state functional
magnetic resonance imaging (rs-fMRI), trained on a
dataset of roughly 40K rs-fMRI sequences covering a wide variety of public and available-by-permission
datasets. While most existing proposals use
transformer backbones, we utilize multi-resolution temporal modeling of the dynamics across parcellated
brain regions. We show that MnemoDyn is compute efficient and generalizes very well across diverse populations and scanning protocols. When benchmarked against current state-of-the-art
transformer-based approaches, MnemoDyn consistently delivers superior reconstruction quality. Overall, we find that with such large-scale pre-training on (non-proprietary) rs-fMRI
datasets, we get a highly performant model for various downstream tasks. Our results also provide evidence of the efficacy of the model on small sample size studies which has implications for neuroimaging studies at large where resting state fMRI is a commonly acquired imaging modality.
Read the full abstract ▾
All authors Sourav Pal, Viet Luong, Hoseok Lee, Tingting Dan, Guorong Wu, Richard Davidson, Won Hwa Kim, Vikas Singh
Artificial Intelligence · Cross-listed from quant-ph
QML for Quantum Sensing under Measurement-Induced Information Loss
Sounak Bhowmik and Himanshu Thapliyal
Nitrogen-vacancy (NV) centers in diamond can serve as highly sensitive solid-state
quantum sensors for high-sensitivity magnetometry. However, in the noisy intermediate-scale
quantum (NISQ) era, extracting reliable information from noisy, finite-shot, and measurement-limited sensing data remains a considerable challenge. Whereas,
quantum machine learning (QML) offers a potential path to improve parameter estimation by learning nonlinear relationships between
quantum-sensing data and the underlying physical signal. In this work, we investigate the role of QML in
magnetic-field estimation within an NV center-inspired magnetometry setting. We formulated
magnetic field sensing as a supervised regression task. We compared the performance of several classical machine learning models trained on measurement-based classical data with that of
quantum kernel-based models trained on pre-measurement coherent
quantum states. Our objective is to isolate the impact of measurement-induced information loss and therefore provide a theoretical upper bound on the sensing performance. The upper bound is achievable only when coherent
quantum information is directly available to the learning model. Our results show that QML-based sensing performance improves significantly with coherent
quantum-state information, and not much with changes in model complexity or learning paradigm. This observation underscores the importance of learning pipelines that tightly integrate
quantum sensors and QML models to enhance
magnetic field sensing under realistic constraints.
Read the full abstract ▾
Cryptography and Security
SoK: ARCUS: On the Efficiency and Efficacy of Hardware Fuzzing
Alenkruth Krishnan Murali, Raghul Saravanan, Sai Manoj P D et al. (+1 more)
This work presents a comprehensive analysis of contemporary hardware fuzzing techniques applied across three major abstraction layers: Instruction Set Architecture (ISA), microarchitecture, and Register-Transfer Level (RTL). Our study examines key factors including input stimulus quality, mutation strategies, feedback mechanisms, target platforms, reference models, and achieved coverage. We find challenges, goals, and design trade-offs vary significantly across abstraction layers. We further identify several unmet needs in current hardware fuzzing practices, such as intelligent input generation, reliable and scalable golden reference models, expressive feedback channels, and cross-layer integration. Building on these insights, we outline future research directions, including hybrid fuzzing frameworks, AI-assisted test generation, scalable reference models, standardized evaluation metrics and
benchmarks, and human-in-the-loop automation for guided exploration and analysis. Together, they aim to unlock efficient, reliable, and comprehensive hardware verification solutions.
Read the full abstract ▾
All authors Alenkruth Krishnan Murali, Raghul Saravanan, Sai Manoj P D, Ashish Venkat
Artificial Intelligence
Evolutionary Recurrent Decision Model in Developing Adaptive and Maladaptive Behaviors
Andrew Hu
This study introduces the evolutionarily recurrent decision model (ERDM), a computational reinforcement learning framework designed to examine how evolutionary mismatch, bounded rationality, and satisficing contribute to adaptive and maladaptive behavior. ERDM simulates
agents across evolutionary recurrent environments, including threat, prey/goal-pursuits, and alliances.
Agents learn through competing rewards abstracted from survival metrics. A validity study under varying adverse childhood experiences demonstrates that distinct adaptive and maladaptive strategies, such as learned helplessness, avoidance, healthy relationships, and aggression, emerge naturally without being hardwired. These results align with empirical literature, showcasing ecological validity. The results suggest that many psychopathology-relevant aspects may be interpreted as bounded cognitive systems operating under modern-ancestral environmental mismatch, positioning ERDM as a key computational cognitive tool that can be extended to other studies.
Read the full abstract ▾
Computer Vision
SceneReGen: Generative Reconstruction of 3D Scenes from a Single Image
Zefan Tian, Yuteng Ye, Yiheng Zhang et al. (+5 more)
Single-image 3D scene reconstruction must complete partially observed objects and place them coherently in a shared observation-aligned scene frame. Object-level generative priors offer strong completion ability, but their centered, scale-normalized outputs are typically expressed in an object frame, creating a fundamental representation gap between object generation and scene reconstruction. We introduce SceneReGen, a generative reconstruction framework that reinterprets scene reconstruction as the generation and assembly of complete object assets in a shared observation-aligned scene frame. SceneReGen addresses the generation-reconstruction gap through selective pose factorization: each object's observed orientation is encoded directly in the generated mesh, while translation and scale are estimated from instance-level and global scene evidence. Given a scene image and instance masks, a geometry encoder extracts dense cues; learnable shape queries condition a pretrained DiT-based 3D generator to produce complete meshes in their observed orientations, while position queries fuse object and scene features to assemble them in the shared frame. On the 3D-FUTURE evaluation subset, SceneReGen achieves the best scene-level CD, scene-level F-Score, and 3D bounding-box IoU among the evaluated methods, ties the best object-level CD, and ranks second in object-level F-Score. Qualitative outputs in autonomous-driving and
embodied-AI scenes further illustrate the potential of asset-centric reconstruction beyond indoor furniture.
Read the full abstract ▾
All authors Zefan Tian, Yuteng Ye, Yiheng Zhang, Yuhang Yang, Xueqiang Lv, Shizhou Zhang, Le Liu, Di Xu
Computational Complexity · Cross-listed from quant-ph
Quantum Algorithms and Hardness for Point-Count Approximation over Finite Fields
Yota Maeda and Hiroshi Yano
We study the approximation of the number of solutions of Laurent
polynomials over finite fields. For a Laurent
polynomial \[f(x)=\sum_{j=1}^{s}a_jx^{u_j}\in \mathbb{F}_q[x_1^{\pm1},\ldots,x_n^{\pm1}], \] let $U$ be its augmented support
matrix whose columns are $(1,u_j)$ with rank $\rho$ and $N(f) := \# \{x\in (\mathbb{F}_q^\times)^n \mid f(x)=0\}$ be its
torus point count. Our first main result is a
quantum algorithm that outputs $\widehat{N}(f)$ satisfying \[ |\widehat{N}(f) - N(f)| \le \varepsilon q^{n+s/2-\rho} \] with success probability $1-\delta$. Provided that $\rho$ and $\|U\|_\infty$ are bounded, the
algorithm runs in both classical bit and
quantum gate complexity $\mathrm{poly}(n, s, \log q, 1/\varepsilon, \log(1/\delta))$. It provides finer resolution than relative-error approximations in general settings. To the best of our knowledge, in the explicit finite-field input model considered here, no previous
algorithm achieves this additive accuracy with running time
polynomial in $\log q$. Van Dam (arXiv:quant-ph/0405081) conjectured the existence of such an
algorithm under the assumption of an oracle reflecting the algebraic properties of the
polynomial. In contrast, by exploiting a point-counting formula derived from character sums over finite fields, we develop an alternative approach that efficiently approximates the number of points without assuming the existence of such an oracle. As a second main result, we prove that the same approximation problem becomes $\#$P-hard under randomized
polynomial-time Turing reductions when the support
matrix $U$ varies freely as part of the input. Thus, taken together, our results clarify how the effectiveness of the
quantum approach depends on the tradeoff between the accuracy scale and the support parameters of the input
polynomial.
Read the full abstract ▾
Computer Vision
RefineRank: Joint Box Refinement and Ranking for Surgical Spatio-Temporal Grounding
Linzhe Jiang, Jiayuan Huang, Changhao Zhang et al. (+3 more)
Surgical spatio-temporal grounding (STG) requires locating, at each
video time specified by a procedural question, the object that the question asks about. Existing approaches face a trade-off:
vision language models understand the question context but produce imprecise coordinates, whereas open-set detectors provide localized candidate boxes whose confidence does not reflect which box answers the question. We introduce RefineRank, which closes this gap at the candidate-box level. A compact trainable module, RefineNet, combines the
language and regional features of a frozen medical
vision language model with the proposals of a frozen open-set detector: it predicts a bounded coordinate correction and a quality score for every candidate box, and a fixed decoding rule returns the original or refined box with the highest score. On the MedVidBench Official Rankings (Verified), RefineRank records 0.421 STG mIoU, the highest displayed STG score, while its global multi-metric rank is 11. In a controlled evaluation on separate training and evaluation
videos, coordinate correction raises the candidate oracle upper bound from 0.6772 to 0.7302, and ranking the joint pool of original and refined candidates by their RefineNet scores improves STG mIoU from 0.2719 to 0.4534, whereas separately trained selectors over the same pool reach at most 0.4186. These results show that a small box-level module can reconcile question understanding with precise localization without retraining either backbone. Code is available at [https://github.com/linzhe001/RefineRank](https://github.com/linzhe001/RefineRank).
Read the full abstract ▾
All authors Linzhe Jiang, Jiayuan Huang, Changhao Zhang, Chunyang Jiang, Zhehua Mao, Mobarak I. Hoque
Computer Vision
GlanceWAM: Sparse Test-Time Imagination for World-Action Models
Linhan Wang, Zijian An, Mingyuan Zhang et al. (+8 more)
Video generative models provide rich physical priors for
robot learning, yet existing world-action models (WAMs) face a fundamental trade-off: synchronous
video generation at control rate is latency-prohibitive, while abandoning test-time visual imagination sacrifices task success. We show that visual imagination achieves both real-time inference and superior success rates when generated asynchronously off the critical path and consumed directly in latent space. We introduce GlanceWAM, which decouples imagination from control on a single shared
video DiT backbone: an asynchronous proposer glances ahead on a slow clock to imagine a single lookahead frame seconds into the future in the background, while an action head decodes action chunks at control rate (48 ms) purely in latent space without blocking. Enabled by a non-interfering
attention mask that isolates
video representations and staleness-robust horizon training that accommodates asynchronous lookahead aging, GlanceWAM breaks the speed-success dilemma. Trained purely on demonstrations, it attains 72.2% on the 24-task RoboCasa kitchen
benchmark (vs. 67.1% for synchronous Cosmos Policy) and 99.0% on LIBERO while cutting per-chunk control latency 24x relative to synchronous world-action models (48 ms on one A100). In single-arm and bimanual real-
robot manipulation, it achieves higher average success than $\pi_{0.5}$ without any
robot-data pretraining. Code is available at https://github.com/linhanwang/GlanceWAM and real-
robot videos at https://linhanwang.github.io/glancewam/.
Read the full abstract ▾
All authors Linhan Wang, Zijian An, Mingyuan Zhang, Chen Dai, Yi Xu, Can Cui, Jiayan Wang, Zichong Yang, Yinlin Chen, Lifeng Zhou, Chang-Tien Lu
Systems and Control
Dynamical System-Based Imitation Learning and Neuroadaptive Control for Trajectory Recovery in Autonomous Ships
Yeyson A. Becerra-Mora and José Ángel Acosta
Repetitive maritime operations can be effectively learned using the Imitation Learning (IL) paradigm, which transfers human expertise directly to Unmanned Surface Vehicle (USV) control systems. Dynamical Systems (DS) are widely used to model non-linear human demonstrations while offering inherent stability guarantees. However, real-world execution under persistent marine perturbations reveals a critical trade-off: standard DS-based IL approaches prioritize global target convergence at the expense of localized trajectory reproduction fidelity. To address this limitation, we present a hybrid learning-control architecture that integrates a DS-based IL reference generator with a neuroadaptive controller. Our approach introduces a control action that drives the USV back to the demonstrated path following exogenous disturbances, enabling dynamic human-like reactive
alignment-termed behavioral tracking. The proposed methodology is validated using the Marine Systems Simulator (MSS) toolbox. Simulation results confirm that the framework generalizes complex maneuvering tasks while substantially improving trajectory tracking fidelity under disturbances compared to alternative control strategies.
Read the full abstract ▾