Computer Vision
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
Tuesday 18 August
While adversarial prompt tuning can enhance robustness of vision-language models efficiently, we find that existing methods aggravate robust generalization overfitting on seen classes, leading to a rapid degradation in performance against adversarial examples of unseen classes as training progresses. We empirically identify that this degradation stems from the tendency of the model to learn pseudo-robust features (i.e., non-generalizable shortcuts). To mitigate this, we propose ADAPT (Adversarial Disentangled Prompt Tuning), a robust prompt tuning framework following the philosophy of “Learning What Not to Learn”. Specifically, ADAPT uses a dual-prompt mechanism with a target prompt and a pool of decoy prompts. During training, the decoy prompts are guided to entrap diverse pseudo-robust features, while the target prompt is constrained to be orthogonal to the decoys in the embedding space to learn robust features. By disentangling the robust features from the pseudo-robust features, ADAPT effectively prevents robust generalization overfitting. We further provide an analysis showing that the orthogonal loss bounds the effect of shifts in pseudo-robust features on unseen classes, yielding a testing error guarantee. Empirically, extensive experiments demonstrate that ADAPT substantially improves the robustness of the target prompt on unseen classes. The code is available at https://github.com/cheny02/ADAPT-ACMMM2026.
Information Theory
Chi-Squared Geometry for Robust Finite-Blocklength Information and Dispersion Analysis
We develop a column-wise chi-squared geometry for discrete memoryless channels (DMCs) yielding tight, logarithm-free bounds on mutual information, channel dispersion, and finite-blocklength coding rates without evaluating logarithms of the channel matrix. The key parameter is~\(\eta\)—the worst-case relative deviation of a transition probability from its output marginal, which is small precisely when the channel is close to the fully noisy channel $t_{ij}=s_j$. We prove three main results: (1) a third-order ratio expansion showing \(I(X;Y)/\chi^2(X;Y)\to 1/2\) as \(\eta\to 0\) with an \(O(\eta)\) skewness correction; (2) a two-sided dispersion equivalence bounding \(V(X;Y)\) above and below by \(\chi^2(X;Y)\) with explicit constants \(c_{\pm}(\eta)\to 1\); and (3) a certified robust design rate \(R_{\mathrm{cert}}(n,\varepsilon)\) with total certification gap \(O(\eta)+O(\eta/\sqrt{n})+O(\log n/n)\). The certified bounds on \(I\) and \(V\) require only addition, multiplication, division, and square roots; the final rate also uses \(Q^{-1}(\varepsilon)\).
Read the full abstract
Hardware Architecture
NeuroAbs: A Neuro-Symbolic RTL Abstraction Framework for Property Checking Acceleration
Formal verification is a crucial technique for ensuring the functional correctness of hardware designs. In the context of property checking, a key challenge is how to efficiently prove a user-specified property in the face of increasingly complex RTL designs. To address this challenge, abstraction techniques are often employed to reduce system complexity and accelerate the verification process. However, prior RTL abstraction methods either require significant manual effort or rely on rule-based techniques that lack flexibility. This paper introduces NeuroAbs, a neuro-symbolic framework for RTL abstraction. NeuroAbs first uses LLM-assisted RTL analysis to identify signals suitable for abstraction. It then combines LLM-based abstraction with an AST-based symbolic RTL representation to better align the generated abstraction with the intended transformation. The soundness of each abstraction is checked using satisfiability modulo theories (SMT) solving. If the abstraction is too coarse for a successful proof, NeuroAbs applies counterexample-guided abstraction refinement (CEGAR) to iteratively refine the model. Experimental results show that NeuroAbs significantly improves the efficiency of hardware property checking across a range of verification tasks.
Read the full abstract
Artificial Intelligence
SignalReasoner: Assessing the Upper Bound of 3B Models for Signal Mathematical Reasoning
Post-training with supervised chain-of-thought fine-tuning and reinforcement learning from verifiable rewards has substantially improved the mathematical reasoning capabilities of large language models (LLMs). However, their application to signal processing problems remains relatively under-explored. This report investigates reinforcement fine-tuning strategies for adapting Qwen2.5-3B-Base to graduate-level signal mathematical problems from WirelessMATHBench-XL, a comprehensive benchmark for mathematical reasoning in this domain. We examine two training paradigms: (i) direct reinforcement learning (RL) on WirelessMATHBench-XL with verifiable rewards; and (ii) supervised fine-tuning (SFT) on a distilled wireless-domain chain-of-thought corpus, followed by the same domain-specific RL stage. Across both paradigms, we benchmark Group Relative Policy Optimization (GRPO), Group Sequence Policy Optimization (GSPO), and Geometric-Mean Policy Optimization (GMPO). We aim to assess whether domain-aware CoT SFT serves as an effective initialization for subsequent RL, and whether GSPO or GMPO offer advantages in stability or accuracy over GRPO for signal reasoning tasks. Our best model achieves an overall accuracy of 39.12%, representing a more than threefold improvement over the untrained Base model (12.37%).
Read the full abstract
Artificial Intelligence
LiveHouse-TS: An Open-world Living Benchmark for Time Series Foundation Models
Time Series Foundation Models (TSFMs) have recently emerged as a highly promising paradigm for cross-domain zero-shot forecasting. However, existing evaluation protocols predominantly rely on static benchmarks with fixed historical test windows. While these benchmarks provide a valuable baseline snapshot, they evaluate an average performance on a fixed history, failing to capture how models behave in continuously evolving real-world environments characterized by seasonal variations, distribution shifts, and unexpected events. To bridge this gap, we introduce LiveHouse-TS, the first open-world living benchmark infrastructure for TSFMs. By evaluating models prequentially on real future data in open-world environments, LiveHouse-TS shifts time series benchmarking from snapshot accuracy to continuous temporal validity. Rather than acting as a one-off leaderboard, our infrastructure serves as a continuous time series infrastructure designed to explore vital, long-term scientific questions: Can model rankings be maintained over the long term? Which models remain genuinely robust under distribution shifts? Extensive streaming evaluations across 11 domains with 17 datasets demonstrate that static rankings undergo a dramatic reshuffling under a live protocol.
Read the full abstract
Graphics
3D Gaussian Accelerated Ray Tracing: Fast training through particle-based backward propagation
3D Gaussian Splatting has made Gaussian primitives a highly efficient representation for real-time novel view synthesis, but its rasterisation-based formulation relies on screen-space approximations that limit accurate view-dependent ordering and the integration of secondary ray effects such as reflections, refractions, and shadows. Gaussian ray tracing addresses these limitations by evaluating explicit ray-primitive intersections, yet it remains costly to train. We observe that the main bottleneck is not ray traversal alone, but the pixel-centric backward propagation, where many threads concurrently accumulate gradients into the same primitive parameters, causing severe atomic contention and thread serialisation. We present 3DGART, a practical training framework for ray-traced Gaussian rendering. Our key idea is to reorganise backward propagation around primitives rather than pixels. Using conservative perspective-correct screen-space bounds, we build a compact intermediate buffer and a tile-primitive mapping that allows each thread to accumulate the contribution of one primitive over its covered pixels within a tile. This transforms gradient computation from a contention-heavy scatter operation into a structured gather-like process. On Mip-NeRF 360, 3DGART achieves an $\approx 3-3.5\times$ raw training speedup over per-pixel baseline and $\approx4 \times$ over 3DGRT on Mip-NeRF 360 while improving quality. More importantly, 3DGART makes fully ray-traced Gaussian training practical, reaching runtimes competitive with rasterisation-based pipelines while preserving benefits of ray tracing.
Read the full abstract
Human-Computer Interaction
SleuthTalk: Supporting Historical Photo Identification with Private Workspaces for Collective Sensemaking and Deliberation
Identifying individuals in historical photographs is a critical task across fields such as history, journalism, genealogy, and archival research. While AI-based facial recognition can efficiently generate candidate matches, it often produces ambiguous results that require deeper analysis and contextual interpretation. Existing platforms lack robust support for collaborative deliberation, especially in uncertain or high-stakes cases. We present SleuthTalk, a private collaborative workspace integrated into Civil War Photo Sleuth, designed to scaffold structured comparison, discussion, and group decision-making. SleuthTalk enables users to curate custom shortlists, annotate facial features, and build consensus through structured feedback. In a mixed-methods evaluation with experienced historical photo researchers, SleuthTalk enhanced self-reported confidence, surfaced diverse perspectives, and supported transparent, reflective identifications.
Read the full abstract
Game Theory
Fairness–Stability Trade-offs in Many-to-One Matching
We study the trade-off between firm-side fairness and coalition stability in many-to-one matching markets with transferable payments. For a fixed matching $X$, we characterize the largest supportable core factor by a bottleneck financing problem: $\alpha(X)=1/\Phi(X)$, where $\Phi(X)=\min_{z\ge0}\max_i R_i(X,z)$. This yields a polynomial-time linear program and local sensitivity formulas for one-worker reallocations. We then develop a maximum-edge round algorithm and a broader class of mutual-top safe choices. Every safe execution is EF1 and, with $t=\delta(A)$ denoting the minimum positive-edge quality, guarantees $\alpha(X)\ge\max\{t,1/[m-(m-1)t]\}$ and $SW(X)/OPT\geq t+(1-t)/m$. These bounds give finite-firm lower and upper bounds for the EF1–core minimax frontier, with exact results for two firms and for three firms when $\delta\le1/2$; as the number of firms grows, the tight scale-free stability rate is $\delta$. We also extend the financing formulation to stronger $EFX^+$ fairness and capacity-constrained markets.
Read the full abstract
Discrete Mathematics · Cross-listed from math.CO
Enumerating forcing and strongly forcing (0,1)-matrices
Let $Q$ be a nonzero $s\times t$ $(0,1)$-pattern, and let $m\ge s$ and $n\ge t$. An $m\times n$ matrix is strongly $Q$-forcing if every $1$-entry belongs to an $s\times t$ submatrix equal to $Q$. Let $F^{*}(m,n,Q)$ count these matrices. Put $H=m-s+1$ and $W=n-t+1$. We prove \[ F^{*}(m,n,Q)\ge 2^{HW}. \] Writing $r$ and $c$ for the numbers of nonzero rows and columns of $Q$, equality holds if and only if \[ (H=1\text{ or }r=1)\qquad\text{and}\qquad(W=1\text{ or }c=1). \] Thus the minimum over all nonzero $s\times t$ patterns is $2^{HW}$, attained exactly by singleton patterns when $H,W>1$, and every fixed nonzero pattern has square growth rate $1$. We also refine the count by weight. If $o(Q)$ is the number of $1$-entries of $Q$, then the number of strongly $Q$-forcing matrices at the minimum positive weight $o(Q)$ is $\binom{H+r-1}{r}\binom{W+c-1}{c}$; at every fixed density in $(0,1)$, the logarithmic growth rate is the binary entropy when $m$ and $n$ are comparable. For ordinary forcing, where every $s\times t$ submatrix contains the $1$-entries of $Q$ in their prescribed positions, let $F(m,n,Q)$ be the number of forcing matrices and let $\mathfrak m(m,n,Q)$ be their minimum weight. We prove \[ F(m,n,Q)=2^{mn-\mathfrak m(m,n,Q)} \quad\text{and}\quad 2^{\,mn-\mathfrak m(m,n,Q)+HW} \le F(m,n,Q)F^{*}(m,n,Q) \le 2^{mn}. \] The lower product bound has the same equality cases as the strong-forcing lower bound above, while the upper product bound is attained exactly by singleton patterns. In particular, the product is at least $2$, with equality exactly when $s=m$, $t=n$, and $Q$ is the all-ones pattern.
Read the full abstract
Machine Learning
Beyond MSE: Rethinking the Evaluation Metric and Benchmarking for Irregular Time Series Forecasting
Existing research on irregular time-series forecasting has primarily focused on model design, while evaluation metrics remain insufficiently studied. Existing benchmarks typically use mean squared error (MSE) as the evaluation metric. We show that, in irregular forecasting, MSE is determined not only by the model prediction but also by the sample-specific timestamp sampling distributions, leading to a biased assessment of the models' continuous-time predictive performance. To address this issue, we propose the Continuous-time Squared Error (CSE), which employs importance weighting to eliminate the influence of the timestamp sampling distributions. We further theoretically prove that CSE's asymptotic estimation error with respect to continuous-time risk is no greater than that of MSE. Finally, we construct a systematic benchmark covering synthetic, semi-synthetic, and eight real-world datasets to validate the effectiveness of CSE and systematically evaluate models' continuous-time predictive performance. Experiments show that CSE can recover continuous-time risk more accurately than MSE, while relying solely on MSE may not fully reflect models' continuous-time predictive performance in real-world scenarios. Our code can be obtained at https://github.com/hnu-vis/ITS-Bench.
Read the full abstract
Computer Vision
B-Spline Embedded Structure Learning for 3D Tooth Segmentation
Accurate 3D tooth segmentation forms the cornerstone of digital dentistry, yet it remains a formidable challenge due to the inherent intricacy of real-world dentitions, such as crowding, misaligned teeth and high morphological similarity between adjacent teeth. To resolve this, we present B-Spline Embedded Structure Learning, a novel framework that distills the inherent sequential arrangement of teeth into a continuous structural constraint to regularize representation space. Our approach parameterizes the global dental topology by fitting a parametric B-spline trajectory to tooth centers, assigning each point a continuous structural embedding that forces the shared backbone to capture global arch organization. To fully exploit these embedded priors, we introduce a Structure-Aware Dynamic Classifier (SADC) to substitute rigid static templates with adaptive, case-calibrated decision boundaries. SADC regularizes dynamic prototype pooling via a localized Gaussian proximity gate and contextually co-evolves them through an attention block modeling spatial relations and bilateral symmetries across teeth. Extensive evaluations on the 3DTeethSeg22 benchmark demonstrate that our method establishes a new state-of-the-art accuracy with exceptional structural robustness and efficiency in computational overhead, markedly enhancing the model's capacity to handle complex dental configurations.
Read the full abstract
Systems and Control
Universal Approximation of Maximal Lyapunov Functions with Anchored Neural Networks
Maximal Lyapunov functions encode the entire domain of attraction of an asymptotically stable equilibrium, but preserving strict decrease under neural approximation is difficult because its margin vanishes at the equilibrium. For systems locally dominated by an asymptotically stable homogeneous vector field, we construct a continuously differentiable maximal target and an anchored, positivity-preserving neural family. We prove semiglobal universal approximation: strict neural Lyapunov functions and their first derivatives can approximate the target on nested invariant sublevel sets that exhaust the domain of attraction. We also provide directly verifiable conditions under which a candidate neural Lyapunov function can be formally certified, and illustrate the effectiveness of the proposed neural architecture through numerical examples.
Read the full abstract
Artificial Intelligence
PlanPO: Group Planning-Aware Policy Optimization for Multi-Turn Agentic LLMs
Group-relative policy optimization has emerged as a key paradigm for training agentic large language models (LLMs) on multi-turn interactive tasks. However, most existing variants fail to distinguish advantages among successful trajectories even when these trajectories differ substantially in their interaction efficiency. For instance, circuitous successes are often assigned the identical outcome reward, causing advantage collapse and severe performance bottlenecks. To this end, we propose Group Planning-aware Policy Optimization (PlanPO), a simple yet effective RL method for learning generalizable planning abilities beyond task-specific high-quality behavior patterns. Specifically, PlanPO introduces coarse-to-fine advantage signals, which capture the relative differences in trajectory-level lengths and turn-level response lengths conditioned on successful trajectories sampled for the same task. Within the group-relative optimization structure, this enables agents to actively learn generalizable and deliberate behaviors spanning interaction planning and textual generation from high-quality rollouts, without degenerating into vanilla length minimization. Experimentally, PlanPO improves over GRPO by 27.2% on average across the challenging multi-turn benchmarks ALFWorld, WebShop, and SciWorld, outperforming recent powerful baselines while incurring negligible additional training cost.
Read the full abstract
Computation and Language
Q-Interference: Memory-Efficient Phase-Aware Quantum-Inspired Attention
GPT attention measures token compatibility through dot-product similarity. This mechanism is simple, effective, and memory-efficient. But it does not explicitly model whether strong token features should reinforce or suppress one another. We introduce Q-Interference, a fully classical quantum-inspired attention mechanism for autoregressive language modeling that augments each query and key feature with an amplitude and a learned phase. The resulting attention score is phase-aware which aligned phases contribute constructively while conflicting phases contribute destructively. Although Q-Interference yields a richer interaction rule than similarity alone, a naive implementation of Q-Interference requires a large token-pair-feature interaction tensor, making it memory-intensive and often impractical. To address this limitation, we propose an exact trigonometric factorization that computes the same score using two standard matrix multiplications avoiding materialization of the large intermediate tensor. Q-Interference fits directly into a Transformer block in GPT and leaves the remainder of the model architecture and next-token prediction objective unchanged. Experiments on public benchmark datasets and baseline models show that the proposed reformulation trains stably in a controlled GPT-style setting and provides a consistent memory advantage over naive phase-aware interference attention. These results support the specific contribution of this work: an exact memory-efficient reformulation that makes phase-aware interference attention practical within a standard GPT pipeline.
Read the full abstract
Systems and Control
Integrated Heat and Power System Scheduling with Continuous-Time Thermal Dynamics via Bernstein-Galerkin Optimization
Coordinated scheduling of district heating networks (DHNs) and electric power systems can improve operational flexibility and reduce costs by exploiting thermal inertia. Most existing formulations rely on simplified discrete-time DHN models, which may inadequately represent continuous spatiotemporal thermal dynamics and can lead to biased flexibility estimation and suboptimal schedules. In this paper, an integrated heat and power system scheduling framework that explicitly incorporates the continuous-time thermal dynamics of DHNs is proposed. A Bernstein-Galerkin transform method is developed to convert the underlying partial-differential thermal-dynamics constraints into a finite set of algebraic constraints, enabling tractable optimization while retaining dynamic fidelity. The resulting model transforms the original infinite-dimensional variational problem into a finite-dimensional coefficient optimization that can be solved using optimization solvers. Compared with conventional discretization approaches, the proposed method provides a more accurate representation of thermal dynamics and yields schedules with improved economic performance and reliability.
Read the full abstract
Machine Learning
Abra: Scaling Diffusion Image Training
Compute-optimal scaling laws guide the training of frontier language models yet remain largely unexplored for visual generation. We present a systematic scaling law study for text-to-image diffusion models using Abra, a controlled family of flow-matching transformers trained across three orders of magnitude worth of compute ($10^{19}$ to $10^{22}$ FLOPs), reaching significantly larger compute budgets than previous works. We demonstrate that diffusion models scale just as predictably as language models but require far more data to train optimally: compute optimality occurs at approximately $200$ image tokens per parameter, ten times the Chinchilla compute-optimal prescription for LLMs. We show that unlike language models, diffusion models are robust to overtraining and that practitioners should err on the side of more data rather than a larger model. Finally, we show that this predictability extends beyond training loss to generative quality metrics, optimal CFG settings, representation quality, and even the shape of the training curves, which collapse onto a universal form.
Read the full abstract
Machine Learning
Rethinking Irregular Time Series Forecasting from the Perspective of Basis Functions
Irregular time series forecasting is crucial in many domains, such as healthcare and meteorological observation. However, due to the inherent characteristics of irregular time series, including sparse observations and non-uniform sampling, accurately predicting future dynamics remains challenging. In light of these two characteristics, many existing methods aggregate irregular observations into fixed-dimensional estimated response coefficients through predefined basis functions and use these coefficients as sequence representations. Nevertheless, this modeling paradigm still suffers from two key limitations: (i) a potential non-vanishing asymptotic bias caused by ignoring the sampling density of timestamps; and (ii) the limited adaptability of predefined basis functions to diverse temporal patterns. In this study, we propose a Debiased Neural Basis-Function Network (DNBNet) to address these challenges. Its core is a debiased neural basis-function response mechanism, which corrects asymptotic bias through importance sampling while parameterizing basis functions with neural networks to adapt to diverse temporal patterns. In addition, considering the sparsity of irregular data, we design a novel multi-scale decomposition module based on average pooling, together with a mass-aware fusion mechanism, to obtain richer representations. Finally, a dual-branch decoder is employed for forecasting. Extensive experiments on multiple real-world datasets demonstrate the effectiveness of DNBNet and its strong generalizability across diverse irregular time series scenarios. Our code can be obtained at https://github.com/hnu-vis/DNBNet.
Read the full abstract
Computer Vision
UniQuery4R: Unified 4D Scene Reconstruction from a Single Query
Reconstructing dynamic 4D scenes requires jointly estimating correspondence, geometry, object motion, and camera motion. Existing feed-forward methods typically predict dense task-specific maps or independently process source-target pairs, leading to unnecessary computation for sparse queries and limited feature reuse across different frame pairs. We present UniQuery4R, a query-conditioned framework that encodes a multi-frame clip once and selects the source view, target view, and continuous source-image coordinate only at decoding time via source-to-target cross-attention. Each query jointly predicts target correspondence, target-time 3D position, and scene flow, along with source depth, while camera parameters are estimated per view. This design allows the encoded clip to be reused across arbitrary source-target selections and supports both sparse inference and dense reconstruction through batched queries, without learned temporal embeddings tied to a fixed clip length. We further introduce a direction-magnitude parameterization of scene flow with separate supervision for moving and static points. Among the evaluated methods, UniQuery4R achieves the best macro-average results on WorldTrack for both scene-flow estimation and dynamic-point reconstruction.
Read the full abstract
Artificial Intelligence
DeAR: Decentralized Agentic Reasoning via Capability Grounding and Collaborative Thought Navigation
Existing agentic reasoning systems typically rely on centralized protocols. This design introduces routing bottlenecks and static role allocations that often fail when handling complex multimodal queries. We propose DeAR (Decentralized Agentic Reasoning), a framework that shifts from central control to autonomous peer-to-peer collaboration. DeAR is built on three mechanisms: (1) decentralized capability grounding for query-dependent agent specialization, (2) thought map navigation for targeted peer interactions, and (3) topology update for adaptive error correction. Evaluations across 9 diverse multimodal reasoning and text-based QA benchmarks indicate that DeAR consistently outperforms recent baseline methods, validating that decentralized and adaptive collaboration among agents enhances accuracy in knowledge-intensive reasoning tasks. The source code will be available at https://open_upon_acceptance.
Read the full abstract
Computer Vision
Key-Frame Reasoning with SAM3: Third Place Solution for the MeViS-Text Track of the 8th LSVOS Challenge
This report presents a two-stage, training-free solution for the MeViS-Text track of the 8th LSVOS Challenge. The task requires a model to localize and segment the object specified by a natural-language expression throughout a video. Such expressions often depend on temporal cues, including actions, interactions, directions, and relative positions. Our first stage uses Gemini-3.1 Pro via API to decompose a video-level event into instance-level targets, select a key frame for each target, and generate a discriminative description aligned with that frame. In the second stage, SAM3-agent produces a pixel-level seed mask on the selected frame, and the SAM3 video tracker propagates the mask bidirectionally through the video. Valid instances are grounded and propagated independently before their frame-wise masks are merged. All local SAM3 processing runs on a single NVIDIA GeForce RTX 4090 without task-specific training or model ensembling. Our method ranked third on the challenge test set, obtaining J&F, J, F, N-acc., T-acc., and Final scores of 0.761, 0.7367, 0.7852, 0.8333, 0.9755, and 0.856593, respectively.
Read the full abstract
Cryptography and Security
When Agents Act on Web3: An Attack-Surface Survey of MCP, Skills, and Tool Calling
AI agents increasingly act rather than merely read: across the Model Context Protocol (MCP) ecosystem, the share of deployed tools that modify external state has risen from 27% to 65% of tool use. When agents exercise this authority on public blockchains through MCP, skills, and tool calling, the consequences of an attack are governed by the blockchain execution layer rather than by conventional software assumptions. This survey argues that four properties of that layer (irreversibility, signing authority, continuous autonomy, and sequence-level composition) qualitatively change the threat model, turning the recoverable failures of generic agent security into a standing, irreversible loss. We organize the fragmented MCP-security literature into an attack-surface taxonomy, then contribute a Web3 risk-mapping matrix that ties each attack class to its amplified impact, the responsible amplifiers, a representative mitigation, and the residual gap. We synthesize defenses, including emerging blockchain-based mechanisms, and find them improving but insufficient: measured protections stop fewer than 30% of attacks, and model-level safety refuses fewer than 3%. We close by positioning the work against adjacent surveys and deriving a research agenda from the matrix's open cells.
Read the full abstract
Artificial Intelligence
ASI-Bench: At the Dawn of Artificial Superintelligence
Artificial superintelligence (ASI) requires AI to move beyond mastering existing knowledge toward exploring the unknown, creating new knowledge, and turning new ideas into verifiable results. However, the capabilities of today's AI systems are still largely built on learning, compressing, and applying existing human knowledge. Accordingly, existing benchmarks primarily test whether AI can produce correct answers based on learned knowledge, or whether it can complete tasks under extensive human guidance. We therefore introduce ASI-Bench, the first benchmark to jointly evaluate AI systems' capabilities of innovative exploration and autonomous scientific execution across general research domains, and the first to progressively withdraw human methodological guidance within the same research project to test how far AI can proceed on its own. Built by over 40 experts with the cost of 31,000+ human hours, ASI-Bench contains 60 project-level research tasks across 11 scientific domains and progressively reduces methodological guidance to test whether AI can independently select methods, conduct research, and produce verifiable results. All tasks undergo expert review, AI-assisted auditing, sandbox execution, and scorer validation. Across 18 state-of-the-art agent–model configurations, the average score drops from 50.91 with full methodological guidance to 29.10 with only the method specified and 26.62 when agents must determine the method themselves. This sharp decline shows that current systems remain heavily dependent on human guidance and are still far from autonomously conducting end-to-end, project-level scientific research. ASI-Bench is open to the world. We invite researchers and builders everywhere to contribute new tasks, challenge the limits of today's AI, and help accelerate humanity's collective path toward artificial superintelligence at https://asibench.apexin.ai/submit.
Read the full abstract
Artificial Intelligence
Do LLMs Know a Good Hypothesis When They See One? Logit-Based Energy Scoring Outperforms Prompted LLM-as-Judge for Scientific Hypothesis Ranking
Large language models (LLMs) are increasingly used for scientific hypothesis generation. However, evaluating generated hypotheses remains a challenge for trustworthy AI-enabled scientific workflows. Existing approaches often use LLMs as judges or rely on semantic similarity, which can favor familiar ideas over novel ones. We propose a logit-based energy scoring method that evaluates hypotheses using a language model's intrinsic confidence rather than comparative judgment. We benchmarked seven language models on 1,323 papers across 12 disciplines. Each paper was paired with its hypothesis and fifteen incorrect alternatives. Intrinsic scoring reached 33.0% Hit@1 pooled across both scorers, compared with 16.6% for prompted listwise ranking. The strongest configuration, a 1-billion-parameter model using logit-based energy scoring, reached 53.1%, though this was the maximum across 14 model-by-scorer combinations selected post hoc. Overall, intrinsic model confidence shows potential for scientific hypothesis evaluation. This study also motivates future research on confidence-based methods for trustworthy AI-enabled scientific discovery.
Read the full abstract
Machine Learning
Understanding Curriculum Learning in Large Language Models via Cross-Difficulty Optimization Dynamics
Curriculum learning has been widely adopted in the post-training of large language models by organizing training data from easy to hard. However, its effectiveness varies substantially across reasoning tasks, suggesting that no single curriculum is universally optimal and raising a fundamental question: what determines when curriculum learning works? In this paper, we answer this question by analyzing the optimization dynamics induced by different curriculum schedules. We show that the transfer relationship between different difficulty levels characterizes the optimization dynamics induced by curriculum learning, which in turn explains the effectiveness of different curriculum schedules, and formalize this relationship as Relative Transfer, a principled measure of cross-difficulty knowledge transfer. Based on this measurement, we derive Transfer-aware Dynamic Curriculum Sampling (TDCS), which dynamically adjusts the sampling distribution according to the estimated transfer relationship throughout training. Extensive experiments on multiple reasoning benchmarks demonstrate that TDCS consistently outperforms representative scheduling strategies across different tasks, model scales, and training paradigms. More importantly, our work provides a unified optimization-based explanation of curriculum learning through cross-difficulty transfer.
Read the full abstract
Hardware Architecture
The Road Less Traveled: Congestion-Aware NoC Placement and Packet Routing for FPGAs
To help scale to ever-larger and more complex designs, recent FPGA architectures now integrate network-on-chips (NoCs). NoCs help transfer high-bandwidth data over long distances within the chip without using scarce low-delay long routing wire segments. While NoC-enhanced FPGAs aid system integration and design reuse, they also complicate FPGA computer-aided design (CAD) flows by introducing new constraints and metrics. Placement and routing need to optimize NoC metrics like latency and bandwidth utilization and avoid link oversubscription (congestion), while simultaneously optimizing the programmable routing resource usage of the design modules attached to NoC routers. In this work, we develop several new approaches to reduce NoC congestion while minimizing the impact on other design metrics. First, we incorporate a NoC link congestion cost into the placement engine of the open-source CAD flow, versatile place & route (VPR). Second, we integrate turn model NoC routing algorithms into the placement engine to leverage path diversity to further reduce congestion. On average over a suite of 29 benchmarks, combining placement congestion modeling with turn model packet routing reduces NoC congestion by 90.7% at the cost of increasing aggregate bandwidth demand by 4%. In cases where the enhanced placement engine and NoC routing fail to fully resolve congestion, we formulate NoC routing as a Boolean satisfiability (SAT) problem. This approach yields significant additional improvements; the combined algorithm reduces congestion by 95.1% compared to the baseline placement. Finally, we enhance the reinforcement learning (RL) agent in VPR's placement engine by introducing a NoC-aware move type, resulting in an 8.8% reduction in wirelength on designs that make extensive use of the NoC.
Read the full abstract
Systems and Control
Nonadaptive Learning in Robust Nonlinear Output Regulation
This paper considers robust nonadaptive regulation for general nonlinear systems in an output-feedback setting with arbitrarily high relative degree. We develop a nonadaptive design that combines an input-driven filter and a generic internal model with a recursive backstepping law, thereby recasting the regulation problem as the robust input-to-state stabilization of an augmented error system. Unlike adaptive schemes, the proposed method does not rely on linearly parameterized regressors and does not require the construction of Lyapunov functions having merely nonpositive derivatives. Under standard assumptions on the exosystem, including purely imaginary and simple eigenvalues, together with a minimum-phase input-to-state stability condition on the internal dynamics, we establish global asymptotic regulation and derive explicit, verifiable inequalities for selecting the design gains. The resulting nonadaptive framework guarantees convergence of the estimation and tracking errors even when the controlled-system dynamics are complex or only partially known. The effectiveness of the theoretical results is demonstrated using a benchmark controlled Duffing system.
Read the full abstract
Information Theory · Cross-listed from math.ST
Minimax Optimal Estimator and Improved Error Rate for the MLE in Logistic Regression with Gaussian Design
We study finite-sample parameter estimation in logistic regression with Gaussian design, where the goal is to estimate $\mathbf{\theta}^*\in \mathbb{R}^d$ with $R=\|\mathbf{\theta}^*\|_2\ge 1$ from i.i.d. samples $\{(\mathbf{x}_i,y_i)\}_{i=1}^n,$ $\mathbf{x}_i \sim N(0,\mathbf{I}_d)$, $y_i\mid \mathbf{x}_i \sim \mathrm{Bernoulli}((1+\exp(-\mathbf{x}_i^\top \mathbf{\theta}^*))^{-1})$. In this paper, we provide the first minimax optimal estimator, and improve on the best known finite-sample error rate for the maximum likelihood estimator (MLE). These two accomplishments are due to a minimax optimal estimator for the parameter norm $R$. First, we establish the minimax lower bound $\Omega(\sqrt{R^3/n})$ for norm estimation. We then improve the best known norm estimation error rate of the MLE, i.e., $O(\sqrt{R^3d/n})$ from Chardon, Lerasle and Mourtada (2024), to $\tilde{O}(\sqrt{R^3/n}+R^2d/n)$. The additional term, $R^2d/n$, appears to be the intrinsic bias of the MLE, as evidenced by the high-dimensional asymptotic theory of Zhao, Sur and Candes (2022) and numerical examples. We show that, however, this additional term is not information-theoretically necessary. To this end, we construct an efficient debiased norm estimator that achieves the error rate $O(\sqrt{R^3/n})$ and is therefore minimax optimal. Combining this with the optimal direction estimator given by the MLE, we establish the minimax optimal rate $\Theta(\sqrt{Rd/n}+\sqrt{R^3/n})$ for estimating $\mathbf{\theta}^*$, as well as the improved finite-sample error rate $\tilde{O}(\sqrt{Rd/n}+\sqrt{R^3/n}+R^2d/n)$ for the MLE. Numerical experiments demonstrate that the proposed minimax optimal estimators outperform the MLE.
Read the full abstract
Systems and Control
Safe whole-body backstepping control for quadcopter path-following
This paper presents a novel whole-body Backstepping control strategy for safe quadcopter path-following. The proposed approach introduces an integrated control scheme that combines a translational guidance controller with a rigid-body attitude controller. To guarantee asymptotic path convergence, the method utilizes a nominal Integrated Guidance and Control (IGC) based on Artificial Vector Fields (AVF). To ensure reactive safety and collision avoidance, the control law is modified using a smooth distance function within the High-Order Control Barrier Function (HOCBF) framework. The quadcopter dynamics are modeled using quaternion algebra to represent position, velocity, and attitude. By combining the Backstepping approach with HOCBF, the controller guarantees that the vehicle avoids obstacle sets while successfully converging to the target path when unobstructed. The proposed methodology is validated through software-in-the-loop simulations and real-world experimental results using the Crazyflie platform.
Read the full abstract
Systems and Control
A Hybrid End-to-End and Modular Control Architecture Toward Safe Vehicle Lateral Control: Combining Soft Actor-Critic with Model Predictive Control
Connected and automated vehicles demand lateral controllers that are simultaneously accurate, low-effort, and safe under model error and sensor noise. Modular controllers such as model predictive control (MPC) are interpretable and constraint-aware but rely on accurate models and hand-tuned weights. End-to-end learned policies, in particular continuous-action deep reinforcement learning, are adaptable and require no hand-designed control law, but offer no intrinsic safety guarantees and limited interpretability. This paper presents a hybrid architecture that combines an end-to-end Soft Actor-Critic (SAC) policy with a constrained linear MPC into a single steering command, using the MPC's first-step optimum as the model-based anchor and a single monotone blending coefficient that interpolates between the two paradigms. The architecture is evaluated on a linearized lateral bicycle model against a PID baseline, a tuned linear MPC, and a stand-alone SAC policy, across nominal, single-axis robustness, and multi-initial-condition ensemble experiments. The hybrid retains the tracking quality of stand-alone SAC while remaining inside the MPC's actuator envelope and preserving a deterministic, model-based contribution to every steering command. The architecture provides an actuator-envelope guarantee by construction but does not establish recursive feasibility or terminal invariance, and the closed-form blend does not prevent all corner-case divergences at the boundary of the training distribution. A corner-case analysis shows that the blend attenuates but cannot prevent failure under distribution shift, motivating a connectivity-aware extension in which the blending coefficient is scheduled by vehicle-to-everything (V2X) signals to restore model-based authority. Limitations and a path toward a constrained-QP predictive safety filter are discussed.
Read the full abstract
Systems and Control
Balancing a Flying Inverted Pendulum with an Unknown Length Using Model Predictive Control and a Genetic Algorithm Estimator
This paper proposes an online Genetic Algorithm (GA) estimator and a Model Predictive Control (MPC) approach to solve the flying inverted pendulum problem in a practical experiment where the pendulum length is unknown. The performance of the MPC approach was demonstrated on a practical system through disturbance and trajectory tracking tests to assess controller robustness and tracking accuracy. The convergence speed and accuracy of the online GA estimator were validated on a practical system using different initial conditions.