Peer Revenue Equity in science

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science

cs Watch Computer Science

Tuesday 6 October

Computer Vision

MacJEPA: Missingness-Robust Audio-Visual Recognition from Untrimmed Egocentric Videos

Souptik Sen and Zahra Ahmadi

Audio-visual models improve egocentric action recognition by exploiting complementary cues, yet typically assume that both streams remain available at inference. Existing missing-modality methods operate on trimmed, single-event clips in which a stream is entirely present or absent, whereas real sensors fail and recover within long, untrimmed observations. We redefine egocentric modality missingness as temporally localized sensor outages within untrimmed, multi-event observations, with whole-clip absence as the limiting case. We introduce MacJEPA, a missing-modality-robust Masked-context query JEPA that recognizes visual actions and acoustic events from supplied interval queries over audio-visual context. Window-local modality dropout simulates these sensor outages during training. MacJEPA further repurposes masking in JEPA from a self-supervised pretext into a supervised robustness objective, aligning masked and clean latent representations of both multimodal content tokens and the task-conditioned queries. All objectives are optimized jointly with recognition in a single stage, requiring no test-time adaptation. Across Epic-Kitchens-100 and Epic-Sounds, a single checkpoint remains competitive under complete input and consistently surpasses published missing-modality baselines when either the dominant or auxiliary stream is removed. MacJEPA thus unifies strong full-input recognition with temporal missing-modality robustness in a single model operating on untrimmed multi-event videos.
Read the full abstract

Hardware Architecture

On the Impact of Degradation-Balanced Scheduling in Manycore Systems

Maria Pantazi-Kypraiou, George Goudroumanis, George Floros et al. (+2 more)

Electromigration degradation in manycore systems depends strongly on how runtime workload is distributed across cores. Task placement affects power, temperature, and current density, which in turn shape the spatial distribution of degradation over time. However, a scheduler that ranks cores only by their current reliability state may not always distinguish among candidate cores, especially when reliability values are close at decision time. In this case, scheduling decisions can be driven by secondary factors rather than by long-term degradation balance. This paper studies degradation-balanced scheduling and proposes a Reliability-Balanced (RB) policy that predicts the full-task effect of each candidate assignment. RB penalizes added imbalance in electromigration exposure, stress, active time, and task count while guarding the predicted Rvalue floor. On randomized continuous mixed workloads, the proposed scheduler preserves completed tasks while improving average Rvalue by 0.37% and minimum Rvalue by 4.35%. It reduces Rvalue, exposure, stress, and active-time standard deviation by 10.84%, 9.39%, 8.35%, and 16.38%, respectively.
Read the full abstract

All authors Maria Pantazi-Kypraiou, George Goudroumanis, George Floros, Anuj Pathania, George Stamoulis

Databases

Strength Is Not Value: Rethinking Approximate Functional Dependency Measures for Error Detection

Xiaolong Wan and Xixian Han

Approximate functional dependency (AFD) strength measures are commonly used to prioritize candidate dependencies, including when selected dependencies support downstream tasks such as error detection. This practice implicitly assumes that dependencies ranked higher by strength are also better candidates for the task they are intended to support. We test this assumption directly for error detection. We evaluate detection value through precision and recall without imposing a single weighting on different error types, and introduce an explicit false-positive/false-negative cost only when a single candidate-selection objective is required. Across three real-world relations from an established AFD benchmark, $1{,}155$ corruption-and-detection runs, and $513{,}480$ candidate-level evaluations, we find a clear asymmetry. Strength measures tend to track precision positively, but their associations with recall are substantially less stable, with both magnitude and direction varying across measures and experimental settings. Strength-based prioritization is also unreliable at the ranking level: top-$k$ sets chosen by strength often have little overlap with those favored by actual detection performance, and strength-based selections can incur substantial regret. An important part of the recall mismatch is associated with structural detection opportunity under the candidate equivalence-class partition, which tracks recall more consistently overall than the strength measures we evaluate. Corruption mechanism and several static structural properties account for selected effects but do not provide a common explanation across settings, leaving part of the mismatch unresolved. Overall, AFD strength and detection value are distinct and should be evaluated separately.
Read the full abstract

Computer Vision

PIE-PS: Photometric Stereo from Physical Irradiance Event Streams

Xiangze Meng, Guangyu Li, Jing Li et al. (+4 more)

Event cameras record asynchronous log-image-irradiance changes with microsecond latency and high dynamic range. These properties are useful for photometric stereo under moving illumination, but raw events are sparse and depend on an unknown contrast threshold. We start from the event trigger model and derive a physical relation between adjacent events, light motion, and surface normals. This relation gives a direct physics-only solver, but the solver needs the threshold, enough events at each pixel, and independent per-pixel optimization. To address these limits, we introduce PIE-PS, a learning-based framework for dense surface normal reconstruction from raw event streams and known lighting. We form Physical Irradiance Events (PIEs) by pairing two adjacent events at the same pixel with their corresponding light directions. Each PIE provides a Physical Irradiance Event Feature (PIEF), defined as the signed event rate. PIEF does not require the unknown contrast threshold. To share spatial and temporal context across nearby PIEs, we introduce PIE-GNN, which treats each PIE as a graph node and encodes it with its light-pair geometry. Since the reliability of PIE observations can vary with local appearance, illumination geometry, and sensor noise, Reliability-Grading Attention (RGA) predicts reliability weights to down-weight unreliable PIEs. Pixel aggregation then produces dense normals. Experiments on synthetic and real data show that PIE-PS outperforms prior event-based photometric stereo methods and the direct solver baseline.
Read the full abstract

All authors Xiangze Meng, Guangyu Li, Jing Li, Di Mei, Songchen Ma, Mingkun Xu, Rui Ma

Computational Engineering

Hours-of-service-aware siting of charging and battery-swapping stations for long-haul electric trucks under adoption uncertainty

Elnaz Irannezhad and Jia Guo

Planning en-route charging and battery-swapping infrastructure for long-haul battery-electric trucks (BETs) requires models that reflect how trucks actually operate. This paper develops a mixed-integer programming framework that jointly sites charging or swapping stations and schedules each truck's charging, swapping and mandatory driver rest, so that charging time overlaps with regulated rest instead of being added to it. Energy use is derived segment by segment from road terrain with a tractive-force model, and the truck battery is modelled as a set of independently swappable packs. Staged investment under uncertain BET adoption is formulated as a multistage stochastic program with Markovian demand and solved by stochastic dual dynamic integer programming (SDDiP) with Lagrangian cuts; we show that the Lagrangian multipliers can be bounded by each station's annualised cost without weakening the cuts. Applied to twelve freight corridors on Australia's East Coast, the algorithm jointly optimises charging and battery-swapping events as well as mandatory break events. A +/- 30% change in adoption alters the final charge-only network by only -12% to +13% of stations, with over 95% of stations built by the second stage. Hedging against uncertainty mainly changes which sites are chosen (71% overlap with a deterministic rolling-horizon model), not when they are built.
Read the full abstract

Hardware Architecture

A gem5-based Simulation Framework for Computing-in-DRAM

Alexander Kusnezoff, João Paulo C. de Lima, Jeronimo Castrillon et al. (+1 more)

Computing-in-Memory using DRAM (CIMD) has demonstrated substantial energy and throughput gains for memory-bound workloads consisting of bulk-bitwise operations, by performing computation directly within DRAM subarrays. Realizing CIMD, however, requires a redesign of the memory controller and careful mapping of operands onto the memory arrays. Presently, accurate FPGA-based testbeds exist, but they are costly and labor-intensive, while open-source simulators are largely trace-based and cannot execute full application runs. The only full-application CIMD simulator available, included with MIMDRAM, is built on an outdated version of gem5 and its toolchains. We present gem5-CIMD, a full-application CIMD simulation framework built on the latest version of gem5. We extend the simulated instruction set from four bitwise operations to sixteen CIM instructions spanning arithmetic, relational, conditional, and utility operations across a range of data types and bitwidths (4 to 64 bits). A complete memory-management stack comprising multi-level huge-page support, a CIM-aware allocator, and a subarray-aware address-mapping scheme ensures that CIM operands are co-located in the same DRAM subarray. In addition, gem5-CIMD is aware of the DRAM refresh interval and faithfully models and schedules the mandatory refresh operations when needed. We provide a CIM standard library intended as a compiler target and evaluate gem5-CIMD on multiple case studies including an end-to-end KNN workload. The simulator sources, including the evaluated workloads, are publicly available.
Read the full abstract

All authors Alexander Kusnezoff, João Paulo C. de Lima, Jeronimo Castrillon, Asif Ali Khan

Robotics

Compact Robot Policies Need Fine-Grained Visual Representations

Nanhe Chen, Runqiu Yang, Jiawei Tang et al. (+2 more)

Multi-task manipulation policies differ in architecture, scale, and pretrained priors all at once, so published comparisons cannot attribute performance to any single component. We argue that most of it comes from the visual representation, and that parameter scale and generative priors are largely incidental. To test this, we build CoRP (Compressed Representation Policy), a deliberately compact policy (48.9M parameters, no vision-language model and no video-generative prior) that factorizes into a representation extractor and a flow-matching action generator. It reaches 97.0% on LIBERO and 75.78%/73.36% on RoboTwin 2.0 Clean/Randomized, matching systems 40.9-163.6x larger. Holding the action generator fixed, we then vary one extractor property at a time. Pretrained initialization is decisive: a random ViT-S/14 drops to 78.1% and an ImageNet ResNet-34 to 74.5% on LIBERO. Pretraining alone is not enough, as freezing the encoder costs 19.8 points. Compression matters as much: resampling each view to 48 tokens beats passing all patch tokens (97.0% vs 83.2%), and a variational information bottleneck over those tokens is worse than a hard token budget, cutting LIBERO-Goal from 95.8% to 33.0% by suppressing the instruction-dependent token selection the policy relies on. Language conditioning contributes only where the observation leaves the goal ambiguous (LIBERO-Goal: 9.2% to 95.8%), while on RoboTwin 2.0, where observations are unambiguous, removing it slightly improves success. Therefore, we argue that a compact policy works when its representation is pretrained, task-adapted, and compressed. Project page: https://corp-policy.github.io/
Read the full abstract

All authors Nanhe Chen, Runqiu Yang, Jiawei Tang, Sichao Liu, Yuquan Wang

Sound · Cross-listed from eess.AS

CTAG-FX: Reinterpreting Synthesizer Parameter Spaces for Expressive Tone-Shaping Audio FX Design

Geonung Jo and Jongeun Choi

Recent text-guided audio FX research has focused on translating natural-language descriptions into parameters or chain configurations within existing FX systems. However, comparatively little attention has been paid to how the internal control space of an individual FX processor can itself be constructed. To address this gap, we propose CTAG-FX, which functionally reinterprets the roles of 78 parameters in a text-conditioned synthesizer configuration as controls of a tone-shaping FX processor. Each synthesizer configuration defines a fixed tone-shaping processor that can be repeatedly applied to new audio inputs rather than producing only a one-off rendered output. The text-conditioned synthesizer configurations are generated using A&R-CTAG, a retrieval-enhanced extension of CTAG. We evaluate CTAG-FX through signal-level analysis, together with a scrambled-mapping ablation in which parameter-to-control assignments are randomly permuted to assess the contribution of the proposed role assignment. Role-based mapping produces prompt-distinct spectral and nonlinear behavior, whereas scrambling reduces this prompt-specific differentiation and yields a more prompt-insensitive nonlinear profile. Overall, these results suggest that synthesizer parameter spaces can serve not only as sound-generation spaces but also as design resources for constructing new audio-FX control spaces. Audio samples are available at https://taylor3527eg-hub.github.io/ctag-fx-platform-demo/.
Read the full abstract

Computer Vision

View Matters: Keyframe-Guided Text-Driven 3D Gaussian Editing

Kaizhe Zhang, Yijie Zhou, Weizhan Zhang et al. (+3 more)

Text-driven 3D Gaussian editing commonly does not distinguish the editing reliability of rendered views, although different viewpoints provide supervision of substantially different quality. Views that clearly show the scene and match the edit instruction provide reliable guidance, while less informative views may weaken the edit when all views are treated equally. We present View Matters, a view-importance-aware framework that conducts editing around reliable keyframes. Keyframe Importance Estimation (KIE) identifies reliable views using geometric visibility, semantic distinctiveness, and edit relevance. Keyframe-Guided Editing (KGE) then propagates their editing signals asymmetrically to non-keyframes without noisy reverse influence, while Importance-Aware Optimization (IAO) preserves this reliability preference during 3DGS optimization. Across 23 scene-prompt pairs, View Matters achieves the highest average CLIP text-image similarity of 0.2822 and directional similarity of 0.2564 among the evaluated methods, with a four-minute editing time. Additional adjacent-view analysis indicates that the fidelity-oriented editing process maintains cross-view coherence.
Read the full abstract

All authors Kaizhe Zhang, Yijie Zhou, Weizhan Zhang, Xuanyu Wang, Feng Lei, Sha Gong

Machine Learning

On the Intrinsic Limited Robustness of Latent-Based Watermarking

Cheng-Han Yeh, Kuan-chun Yu, Cheng-Chang Tsai et al. (+1 more)

Existing latent-based watermarking methods for diffusion models have overestimated their robustness to image distortions, including geometric transformations such as rotation, scaling, and translation (RST). Moreover, this paradigm of watermarking approaches may suffer from inherent limitations arising from the domain in which the watermark is embedded. In this paper, we provide the first theoretical analysis explaining why these methods lack invariance to perturbations. By relaxing the invariant relation, we derive a maximum perturbation bound that characterizes the relationship between pixel-space perturbations and their corresponding effects in latent space. In addition, we present the first analytical formulation that captures all components of practical detection mechanisms. Finally, we conduct experiments to validate the theoretical findings and the limitations of latent-based watermarking methods. Our theoretical and empirical results indicate that, under the current design paradigm, latent-based watermarking methods intrinsically exhibit limited robustness. We conclude by providing the analytical tool and design guidelines that future research could follow.
Read the full abstract

All authors Cheng-Han Yeh, Kuan-chun Yu, Cheng-Chang Tsai, Chun-Shien Lu

Robotics

Nested Power Models for Multirotor Propulsion: From Aerodynamic Drag to Electrical Losses

Antonio Franchi, Aaron Saini, Ahmed Ali et al. (+1 more)

Speed-only aerodynamic power models for multirotor propulsion cannot represent acceleration-dependent effects. This work develops a nested sequence of propulsion-power models that starts from aerodynamic power dissipation and progressively introduces a reversible kinetic-energy rate, torque-dependent electromechanical dissipation, and lumped speed-proportional dissipation. The models are identified using one subset of experiments and validated using the other on a motor-drive-propeller unit. Independent estimates of rotational inertia and aerodynamic drag complement predictive validation by assessing whether the models correctly attribute the measured power to reversible kinetic-energy exchange and irreversible dissipation and, within the latter, to aerodynamic and electromechanical losses. The results show that the reversible kinetic-energy rate is necessary but insufficient for accurate dynamic power prediction. Dissipation proportional to the squared motor torque provides the main additional improvement, while speed-proportional dissipation further prevents irreversible losses from being attributed to reversible kinetic-energy exchange. The resulting methodology provides a reusable and experimentally verifiable basis for developing and selecting dynamic propulsion-power models for multirotor systems.
Read the full abstract

All authors Antonio Franchi, Aaron Saini, Ahmed Ali, Chiara Gabellieri

Artificial Intelligence

LFHE: Local-First Heuristic Evolution for Bounded Local Topology Search in Decentralized Learning with Non-IID Data

Yin-Kuan Liang, Yan Gao and Yang Long

Decentralized learning is highly sensitive to communication topology under non-IID data. Adaptive peer-selection methods can exploit local model information, but broader peer discovery may require increasingly large control state, whereas direct spectral optimization typically relies on graph-wide information. We study the intermediate setting of bounded local topology search and propose Local-First Heuristic Evolution (LFHE), a representation-driven rewiring framework whose candidate discovery and scoring use only ego-neighborhood and friend-of-a-friend (FoF) information. The structural score admits an exact interpretation through graph Dirichlet energy: its sum across clients equals twice the representation Dirichlet energy, which under standard linear consensus dynamics governs the instantaneous dissipation of representation disagreement. LFHE combines this state-dependent structural signal with early exploration and degree control, while algebraic connectivity remains an offline graph diagnostic. Under bounded sparse degree, its FoF candidate state remains local rather than expanding toward population-wide peer tracking. Across four image, speech, and text benchmarks, LFHE achieves competitive decentralized learning performance. Matched-protocol controls identify the structural term as the principal empirical topology-selection signal, while comparison with broader peer discovery exposes a trade-off between predictive performance and discovery-state locality. Together, these results motivate state-aware bounded local topology search between pairwise peer selection and globally informed topology optimization.
Read the full abstract

Cryptography and Security

FBAN: A Fully Homomorphic Encryption Compatible Bottleneck Attention Network for Privacy-Preserving Behavioral Authentication

Jichao Xiong, Atsuko Miyaji and Jiageng Chen

Continuous authentication (CA) strengthens session security by repeatedly verifying the user during device interaction, yet it inherently relies on highly sensitive behavioral traces (e.g., fine-grained touch dynamics) that are often outsourced to cloud/edge services for scalable inference. This raises a fundamental privacy-in-use challenge: protecting behavioral features during computation, not only in transit or at rest. Fully homomorphic encryption (FHE) offers a principled solution, but deploying modern CA models under FHE remains difficult due to non-linearities and attention-style operations that incur high ciphertext cost. We propose FBAN, a TFHE-compatible Bottleneck Attention Network and an end-to-end encrypted CA framework. FBAN is designed for integer-only execution via a two-stage pipeline (floating-point pretraining followed by quantization-aware training) and is compiled into TFHE circuits for homomorphic inference. We further specify a client-server protocol with session-bound blinding and decrypt-and-return verification, enabling the server to authenticate users without observing raw behavioral features. We provide a cryptographic security analysis against an honest-but-curious server under TFHE IND-CPA security, and formalize resistance to replay and impersonation without the TFHE secret key. Experiments on two public touchscreen datasets demonstrate that FBAN achieves strong authentication utility under encrypted inference while maintaining a lightweight model footprint, with TFHE parameters instantiated at $\geq 128$-bit security.
Read the full abstract

Software Engineering

Beyond the Leaderboard: Multi-Dimensional Evaluation of Dense and Mixture-of-Experts Models for Automated Program Repair

Anvi Kalpesh Shah and Umamaheswara Sharma B

Automated Program Repair (APR) with language models is usually evaluated by whether a generated patch passes the test suite, which can hide differences in maintainability, security, and computational cost. We propose a Weighted Quality Index (QI), inspired by the ISO/IEC 25010 software quality model, that combines functional correctness, maintainability, security, and generation efficiency under configurable weighting schemes. We evaluate three dense Qwen2.5-Coder models (3B, 7B, 14B) and the 16B-parameter DeepSeek-Coder-V2-Lite Mixture-of-Experts (MoE) model (2.4B active parameters) on 40 QuixBugs and 90 Defects4J bugs, all run locally on identical hardware to control for infrastructure effects. Model rankings change with the weighting scheme, showing that single-metric evaluation can hide trade-offs. The MoE model shows almost no statistically significant difference in correctness from the 7B and 14B dense models (McNemar's exact test) while using 3-6 times fewer active parameters, whereas correctness increases significantly across the three dense scales. These results suggest that active parameter count can be a more informative lens than total parameter count for sparse code models.
Read the full abstract

Multiagent Systems

Visual Orchestration Tax in Agentic VLM Pipelines: Auditing and Certifying Visual Evidence Reuse

Lingteng Zeng

Agentic VLM pipelines increasingly pass the same static visual evidence through multiple specialist agents and tools. This design creates an orchestration-level redundancy mode: semantically unchanged images are repeatedly reconstructed as image-conditioned requests at the VLM API boundary. We call this phenomenon visual orchestration tax and develop a measurement-to-certification framework for visual evidence reuse in agentic VLM pipelines. The audit side defines $\mathrm{M1}_{\mathrm{trace}}$ to count raw visual-evidence touches and M2 to measure structural touch redundancy, with query-level distributions, bootstrap confidence intervals, and paired quality tests. Across SeeingEye and MAMMQA on chart, document, general-VQA, and multi-modal-QA tasks, audits reveal 66.8-75.6% visual-evidence touch redundancy, and every audited query exceeds the predefined gate. The certification side introduces SharedVisCache, a contract-aware evidence reuse hook keyed by image content, preprocessing fingerprint, and encoder assumptions. On SeeingEye, contract validation certifies 75.0-75.5% repeated touches as reusable while preserving 350/350 output strings and $\Delta\mathrm{M5}{=}0$. At the physical layer, certified hits reduce $F_{\mathrm{vision}}$ from 800 to 200 in ChartQA-200 trace replay and from 200 to 50 inside live SeeingEye translator-stage physical integration, preserving 800/800 replay strings and 200/200 integrated call outputs. The results position visual reuse as a measurable, behavior-preserving property of agent orchestration and define an agent-layer contract that makes backend prefix or token reuse semantically interpretable.
Read the full abstract

Numerical Analysis

AdHImEx: Adaptively High-Order Implicit-Explicit Transport for Large Time Steps

Amber J. te Winkel, Hilary Weller, Christian Kühnlein et al. (+1 more)

Adaptively Implicit-Explicit (AdImEx) time stepping provides stability for large time steps for mass-conservative transport, making it attractive for the numerical representation of advection in weather and climate prediction. Existing AdImEx transport schemes become first-order accurate in the large-time-step, implicit regime and may require multiple sparse matrix solutions per time step. This work introduces AdHImEx, a new AdImEx scheme that addresses both limitations simultaneously. AdHImEx provides second-order accuracy for large Courant numbers while requiring only a single matrix solution per time step. It is a Runge-Kutta scheme that blends third-order accurate explicit time stepping and Crank-Nicolson implicit time stepping to provide numerically verified stability for Courant numbers up to 100. AdHImEx is designed for flows that are predominantly explicit, activating implicit time stepping only where large Courant numbers occur. It retains the full efficiency and third-order accuracy of the explicit time stepping in regions with small Courant numbers. In this work, it is combined with a fifth-order accurate finite-volume discretisation in space for mass conservation. A second contribution of this work is a novel stage-dependent treatment of the divergence operator that preserves constancy despite different implicit and explicit Runge-Kutta stage time steps and spatially varying implicitness, removing spurious divergence. This innovation is applicable to other AdImEx time stepping schemes. Convergence, transport, and efficiency tests demonstrate substantial improvements over previous first-order AdImEx schemes, little phase and amplitude error, accurate transport on locally refined meshes with large Courant numbers, and promising efficiency characteristics for atmospheric transport.
Read the full abstract

All authors Amber J. te Winkel, Hilary Weller, Christian Kühnlein, James Kent

Artificial Intelligence

Which alloy composition,what process parameters? Inferring the recipe from optimized metallic microstructure and texture

Mahish K. Guru, Jan Bohlen, Louam Lemjid et al. (+3 more)

The mechanical properties of a metallic alloy are set by its microstructure and texture: the size and shape of its grains and the orientation of their crystals. That structure is in turn set by a recipe, the alloy composition together with the processing parameters. Alloy development runs this chain forwards, tuning the structure until a target property is met. Running it backwards, from an optimized structure to the recipe that would produce it, still relies on expert knowledge. We ask whether this backwards step can be learned. On an in-house dataset of 107 magnesium alloy extrusion conditions across 14 alloys, each with optical micrographs and an X-ray texture measurement, we compare three descriptors of microstructure and texture: conventional grain and texture statistics, a vision embedding from a pretrained image encoder, and a graph neural network on the grain network. Each is paired with prediction heads for two tasks: the alloy composition given the process (Task A), and the process parameters given the composition (Task B). Under 5-fold cross-validation, the conventional descriptors identify the correct alloy for 65% of held-out conditions, against 17% for always guessing the most common alloy, while the learned embeddings stay below 30%. The process parameters are recoverable but noisier: compared with using the composition alone, the microstructure roughly halves the temperature error. Because only a few alloys were cast and only a few press settings were used, both answers are discrete, and heads that pick from these known options, while respecting their order, worked better than heads that predict a free value.
Read the full abstract

All authors Mahish K. Guru, Jan Bohlen, Louam Lemjid, Marius Tacke, Roland Aydin, Noomane Ben Khalifa

Machine Learning

Align, Then Correct: Training-Free Two-Stage Low-Rank Compensation for Extremely Quantized Large Language Models

Seobin Song, Geonho Lee, Janghwan Lee et al. (+1 more)

Low-rank quantization error compensation (LQEC) recovers the accuracy lost under aggressive weight quantization by attaching a closed-form rank-$r$ adapter beside each frozen quantized weight, without any training. We show that existing compensators are limited by two shared simplifications. They calibrate symmetrically, evaluating the full-precision and compensated weights on the same activation, which yields a compensation target that is inherently high-rank – so a fixed rank budget captures only a small fraction of it. And they minimize only the second-order term of the loss, although the compensated model is not stationary: a first-order descent direction larger than the applied compensation itself remains in every layer, and no reconstruction objective can absorb it. We propose a two-stage closed-form framework that removes both simplifications. Stage 1 aligns each layer's output with the full-precision model under a Fisher-weighted asymmetric objective, concentrating the rank budget on a rank-compressible target. Stage 2 re-measures statistics on the compensated model and applies a rank-constrained natural-gradient step that absorbs the remaining first-order signal. Every adapter is the result of a single truncated SVD; backward passes serve only to collect statistics. At 2 bits under QuIP#, our method reduces WikiText-2 perplexity from 12.43 to 10.26 on Qwen3-8B and from 21.11 to 13.22 on Qwen3-4B. On the held-out C4 corpus, it recovers 51% and 84% of the gap to FP16, versus 31% and 63% for the strongest baseline, with consistent gains in the seven-task zero-shot average, at higher bit-widths, and under a distinct quantizer.
Read the full abstract

All authors Seobin Song, Geonho Lee, Janghwan Lee, Jungwook Choi

Computer Vision

The Failure Is in the Readout: Fine-Grained Emotion Recognition Benchmarks Measure Elicitation, Not Perception

Tobias Hallmen, Fabian Deuser, Robin-Nico Kampa et al. (+2 more)

Fine-grained emotion recognition supports therapy tools and social robots, but it needs facial data, which raises privacy and data-protection concerns. EmoNet-Face-HQ answers that with generated portraits, expert-rated over a $40$-category taxonomy far finer than the usual six to eight basic emotions. Under the protocol it ships with, vision-language models (VLMs) score poorly on that taxonomy, and the benchmark concludes that a dedicated fine-tuned model is necessary: Empathic-Insight-Face (EIF; Small/Large). We show that off-the-shelf VLMs match or beat that fine-tuned model when the answer is not generated but read from the logits, as one binary query per category. We keep the benchmark's images, taxonomy and ratings, and change only how the answer is read. Experts agree at $\kappa_w = 0.468$ on the five categories they measure most reliably. Generatively, no interval among eleven open-weight VLMs lies entirely above that anchor ($\kappa_w=0.268$-$0.486$). Under verification all eleven clear it, each of them significantly better at $\kappa_w=0.507$-$0.586$. Three also significantly beat EIF sitting at $\kappa_w = 0.551$ (Small; $0.534$ Large). The gain comes from the graded probability and not from asking a yes/no question: as a control, thresholding those same probabilities to yes/no costs 142% of the average gains and drops binarization below generative elicitation to $\kappa_w=0.254$-$0.423$. A replication on real photographs (FACES) is weaker and mixed: of the ten models that pass a validity gate, six gain, three are neutral to positive and one is negative, so the effect is not confined to synthetic data.
Read the full abstract

All authors Tobias Hallmen, Fabian Deuser, Robin-Nico Kampa, Norbert Oswald, Elisabeth André

Machine Learning

Symphony for Text Generation: Benchmarking Clinical Note Generation

Daniel Varab, Victor Petrén Bach Hansen, Asbjørn W. Helge et al. (+8 more)

Ambient documentation systems are rapidly gaining adoption, yet their impact on clinical note quality remains poorly characterized. We introduce MedConv, a multilingual dataset of 300 clinical encounters in English, Danish, and German, and use it alongside the Ambient Clinical Intelligence benchmark (ACI-BENCH) to compare Corti, a clinical AI platform, with two leading, accessible ambient scribe software applications built on general-purpose AI. We present a controlled clinical evaluation framework that combines entailment metrics with LLM-judged pairwise comparisons across eight dimensions adopted from PDSQI-9. Results show that Corti's API-based text-generation infrastructure is on par with or outperforms leading commercial scribes. We further show that Corti's configurable API provides the flexibility necessary to fine-tune quality dimensions for specific documentation use cases. We present the evaluation methodology and release a dataset to support future reproducible comparison of ambient documentation systems.
Read the full abstract

All authors Daniel Varab, Victor Petrén Bach Hansen, Asbjørn W. Helge, Kevin Pelgrims, Mathias Baltzersen, Adrian Young-San Roessler, Vanessa Klungtvedt, Maximilian Brand, Lasse Krogsbøll, Henrik Cullen, Lars Maaløe

Computation and Language

Making COMET Comparable Across Scripts: Diagnosis and Correction of Tokeniser-Induced Script Bias in Indic MT Evaluation

G. L. John Salvin and Swapnil Hingmire

COMET reports translation quality as a single number, and that number is routinely compared across target languages written in different scripts. Such a comparison assumes Script Invariance: the score should not depend on the writing system that carries the target. We test it on IndicMT Eval by re-encoding the target into Latin script, which changes orthographic form while holding content and human ratings fixed. Script identity then accounts for 22.9% of native-script COMET variance, and agreement with annotators falls in all five languages studied. We trace the effect to the tokeniser and measure it with three label-free diagnostics. The bias is two faults, not one. Scores from different scripts occupy incompatible ranges, and within a single script the metric orders translations less accurately. No order-preserving transform of the score can repair the second fault. The first is removed exactly by COMET-QN, which maps the score distribution of each (language, script) pair onto a shared reference. Pooled agreement with annotators rises from 0.300 to 0.399, which is what makes scores from different scripts safe to place on one axis, and every within-language ordering is provably preserved. A regressor over parity features recovers a further 17.1% of the lost sensitivity. The remainder belongs to the encoder, and no post-processing can reach it. We therefore recommend publishing the normalised score, the three diagnostics, and the identity of the tokeniser they were computed against, so that a reader can tell how much of a score reflects translation quality and how much reflects the writing system.
Read the full abstract

Computational Engineering

Vibe Building

Yongqing Jiang, Haoran Luo, Jianze Wang et al. (+3 more)

Automated building design must comply with seismic and wind codes and satisfy structural mechanics constraints, yet most existing agents produce visually plausible models without verification grounded in mechanical analysis and code compliance. We introduce the Vibe Building task and propose PE-Loop (Physics-Engine-in-the-Loop), an agent in which a deterministic physics engine is the sole source of evaluation signals, mapping code constraints to a physics process reward, while the language model is confined to proposing discrete revisions (section menu, topology, and lateral system). Designs are verified by held-out seismic and wind time-history checks and a constructability gate. On VB-Bench, 3,577 physics-adjudicated building instances across six code families, PE-Loop achieves the highest verified success rate under three of four backbone LLMs, the highest held-out seismic pass rate under all four, and the highest held-out wind pass rate under three. Replacing the physics verdict with a language-model judge, all else fixed, leaves 58.43% of accepted designs noncompliant. These results suggest that reliable structural design rests less on a stronger LLM proposer than on an adjudicator the proposer cannot influence, a division of labor for agents whose outputs must hold up in the physical world.
Read the full abstract

All authors Yongqing Jiang, Haoran Luo, Jianze Wang, Xin Zhou, Kaoshan Dai, Zhiqi Shen

Multiagent Systems

Token-Efficient Multi-Agent Collaboration via System One-Guided Computational Division of Labor

Zihan Zhou, Xinzhe Hu, Hanxu Yang et al. (+2 more)

Large language model (LLM)-based multi-agent systems (MAS) have become a promising paradigm for complex information-seeking and reasoning tasks by enabling collaborative problem solving among specialized agents. However, existing MAS frameworks tightly couple task reasoning with coordination operations, including task selection, role assignment, message routing, and context management. As interactions grow, using powerful LLMs for these bounded control decisions introduces substantial token overhead and latency, limiting the scalability of agentic Web services. In this paper, we investigate whether coordination can be decoupled from expensive reasoning without compromising collaborative performance. We propose S1-MAS, a token-efficient multi-agent framework based on System One-guided computational division of labor. S1-MAS assigns bounded coordination decisions to lightweight System One models while reserving open-ended reasoning for capable LLM workers. Specifically, a lightweight controller selects inspection conditions, chooses subsequent tasks, and determines termination, while a compact reader retrieves condition-relevant evidence from authorized sources to support these decisions. Through a decision-evidence loop, selected tasks dynamically determine worker roles and source access, enabling adaptive collaboration without task-specific training. Extensive experiments on seven diverse benchmarks demonstrate that S1-MAS achieves superior accuracy while substantially reducing the inference cost. Across individual comparisons with AgentVerse, DyLAN, and SelfOrg on seven benchmarks, S1-MAS reduces GPT-4o token consumption by 44.9%-97.2% and measured end-to-end latency by 37.8%-93.0%. These results highlight its potential for scalable and cost-effective agentic Web applications.
Read the full abstract

All authors Zihan Zhou, Xinzhe Hu, Hanxu Yang, Liangjian Wen, Zhao Kang

Computation and Language

Penalty-Framed No-Valid-Option MCQA: Analyzing LLM Abstention under Invalid Choices

Jinhyeok Kim and Hye-Young Jung

Multiple-choice question answering (MCQA) is commonly used to evaluate large language models under the assumption that one of the provided options is correct, typically using answer-selection accuracy. However, in real deployments, users or retrieval systems may provide invalid option sets in which none of the listed choices is correct, and selecting one of them may incur downstream cost. We study this setting as penalty-framed no-valid-option MCQA. Using the mathematics subset of MMLU-Pro, we remove the labeled correct option, allow models to either choose a remaining option or output ABSTAIN, and penalize invalid forced-choice responses. We further introduce correct-conditioned analysis, evaluating abstention only on instances that the model originally answered correctly. Experiments show that high MCQA accuracy does not fully guarantee abstention reliability: even under explicit no-valid-option-aware instructions and penalty-based scoring, models still produce invalid forced-choice responses for a subset of originally correct instances. These results show that penalty-framed no-valid-option MCQA reveals an aspect of model reliability not captured by standard answer-selection accuracy.
Read the full abstract

Robotics

ViDAL: A Visual Dynamics-Grounded Action Latent Space for Vision-Language-Action Models

Yuan Xu, Yixiang Chen, Qisen Ma et al. (+10 more)

Vision-Language-Action (VLA) models have become a central paradigm for robot policy learning, which predict actions in three forms: raw action chunks, discrete action tokens, or continuous action latents. However, existing action representations primarily model action trajectories, with limited consideration of the visual dynamics induced by these actions. We introduce ViDAL, a Visual Dynamics-grounded Action Latent Space that anchors continuous action latents in the future visual dynamics of the scene. Specifically, ViDAL learns action latent space by training an Action Variational Autoencoder (Action VAE) to reconstruct action chunks while aligning its latent with future scene dynamics. When integrated into downstream robot policies, the proposed Action VAE serves as a plug-in action interface compatible with multiple VLA architectures and enables optional future-video prediction as an additional capability. Empirically, ViDAL outperforms competitive baselines on LIBERO with 98.1% average success, improves a multi-task $\pi_{0.5}$ policy on RoboTwin 2.0 from 54.3% to 65.5% (Clean) and from 33.2% to 43.1% (Random) success rates over 50 dual-arm tasks, and yields 20.0% and 23.4% absolute success-rate gains on real-world single-arm Franka and dual-arm ARX robot platforms.
Read the full abstract

All authors Yuan Xu, Yixiang Chen, Qisen Ma, Jiabing Yang, Peiyan Li, Kai Wang, Jianhua Yang, Jianlou Si, Jun Huang, Jing Liu, Nianfeng Liu, Yan Huang, Liang Wang

Artificial Intelligence · Cross-listed from math.AP

Navier-Stokes lost in translation: Why Lean verification of AI autoformalisation does not guarantee correct natural language proofs

Alexander Bastounis, Fabian Circelli and Anders C. Hansen

Autoformalisation is increasingly used to verify mathematical texts, including those generated by AI, as in OpenAI's announced proof of blow-up of solutions to the Navier-Stokes equations. In this process, an AI system translates the text from a natural language (NL) into a formal language such as Lean. Once this translation is done, the argument expressed in the formal language can easily be mechanically verified. The purpose of this article is to demonstrate why this process may offer no confidence in the original NL argument, owing to the various difficulties in performing the translation semantically faithfully. In particular, we highlight that the problem of resolving ambiguities in mathematical NL text, which is necessary in order to provide semantically faithful translation, is arbitrarily high up in the Solvability Complexity Index (SCI) hierarchy/arithmetical hierarchy (the SCI $= \infty$). Hence, informally, providing semantically faithful AI autoformalisation is harder than any computational problem including the Halting problem (which has SCI $= 1$). To demonstrate the effect of this result we provide several examples of AI mistranslations of NL statements and proofs into Lean in practice, resulting in mismatches between NL proofs and their Lean `verifications'. These include OpenAI's announced Navier-Stokes proof. In particular, we show that the formalised Lean proof does not correspond to the NL proof of blow-up of solutions to the Navier-Stokes equations.
Read the full abstract

Artificial Intelligence

Partially Observable Zero-shot coordination by Predicting Intention of Partner

Jinnyeong Yang, Yuhwan Jeong, Hoyong Kwon et al. (+3 more)

Zero-shot coordination in embodied settings requires acting while the partner is intermittently out of view, leaving existing methods with ambiguous partner representations and uncertainty over hidden partner states. We propose Predicting Intention of Partner (PIP) to jointly address these challenges. PIP uses a Joint-view VAE to distill richer training-time evidence from the union of both agents' local observations into a partner representation available from local observations alone. Partner-state Belief networks further infer the partner's hidden location and behavioral tendencies from the ego agent's interaction history. We evaluate PIP in Burrito-PO, Overcooked-PO, and a Melting Pot substrate, together with a human evaluation in Burrito-PO. PIP attains the highest mean performance among the compared methods across all three benchmarks. Human evaluation and diagnostic analyses further support coordination with unseen partners and the contributions of both components under partner occlusion.
Read the full abstract

All authors Jinnyeong Yang, Yuhwan Jeong, Hoyong Kwon, Minseok Kim, Jihun Kim, Kuk-Jin Yoon

Artificial Intelligence

Test-Time Agent Evolution for Long-Horizon Legal Reasoning

Haotian Chen, Shuaicheng Niu, Haocong Rao et al. (+5 more)

Legal intelligence aims to support reliable decision-making across long-horizon legal processes involving evolving case states and multiple roles. However, real-world legal deployment exhibits substantial case heterogeneity in facts, evidence, and procedural contexts, exposing the limitations of static agent strategies. Moreover, legal reasoning is inherently interdependent across roles and procedural stages, making global reliability fundamentally different from isolated role competence. To address these challenges, we study training-free test-time agent adaptation, where agents continuously exploit deployment-time signals from preceding cases and ongoing interactions without updating model parameters. We propose \method, which introduces Test-Time Memory Evolution to retrieve reusable experience from previous cases, adapt it to the current factual and procedural context, and consolidate accumulated experience for subsequent decision-making. Further, Rubric-Aligned Collaboration verifies and revises role-specific actions according to behavioral and procedural requirements, enabling coordinated decision-making across roles and stages. Extensive experiments on J1-EVAL and LegalWorld across five backbone models demonstrate consistent improvements over representative reasoning and agent baselines with reasonable interaction and computational costs. Ablation and case studies further show that the two components provide complementary benefits in experience adaptation and cross-role coordination, improving the reliability and efficiency of long-horizon legal reasoning.
Read the full abstract

All authors Haotian Chen, Shuaicheng Niu, Haocong Rao, Kaisong Song, Jun Lin, Lizhen Cui, Zhiqi Shen, Yonghui Xu

Cryptography and Security

Rethinking Visual Provenance: Detection and Watermarking Across Direct Visual Generation and LLM-Driven Code Rendering

Zheng Gao, Xiaoyu Li, Zhicheng Bao et al. (+1 more)

AI systems create images and videos with image/video generation models or by writing code and graphics descriptions that are then rendered. These routes can produce similar visible artifacts but expose different representations, intervention points, and provenance evidence. We develop a production-centered framework that compares detection and watermarking across both routes. An explicit verification specification distinguishes passive inference, message recovery, and authenticated provenance. We organize image, video, source-code, and rendering-aware watermarks by production stage. We examine the different requirements of generated images and video, plots and SVG, programmable video, and agent-composed workflows. Documented Claude, OpenAI, and rendering-tool interfaces connect the framework to concrete systems. We pose ten scoped research questions on identifiability, observability, fair comparison across stages, recoverable payload, reconstruction, synchronization, composition, hybrid local contribution, and private production-event authentication. The result is a conceptual research agenda grounded in published methods, inspected interfaces, and elementary boundary examples. It reports no experiments and claims no new theorems; its appendix results are elementary calculations, and documentation and source inspection establish interfaces, not empirical robustness.
Read the full abstract

All authors Zheng Gao, Xiaoyu Li, Zhicheng Bao, Jiaojiao Jiang

Information Retrieval

Adapting Generative Recommenders for Multi-Turn Interaction

Yu-Chen Den, Zhi Rui Tam, Yung-Yu Shih et al. (+4 more)

Generative recommenders decode items from a user's interaction history, but offer no way for users to correct a recommendation that misses their current intent. Adding conversation is natural since items and words share same output space, yet training the model to converse may overwrite the history-to-item mapping it relies on. We introduce INTEGER (**INTE**ractive **GE**nerative **R**ecommendation), which extends generative recommendation to multi-turn interaction with a learned routing token that lets the model decide when to recommend, history re-anchoring that conditions each item on both past behavior and the dialogue, and behavioral replay with instruction-data rehearsal that prevents forgetting during adaptation. Users can thus give feedback on recommendations within the dialogue, while recommendations stay grounded in behavioral history and accuracy is not traded for fluency. On Amazon Beauty and Toys, INTEGER matches or exceeds the strongest baselines in accuracy with competitive conversation quality, improving Hit@10 by 13.3% on Amazon Beauty, and significantly outperforms the generative recommender it starts from. Our analyses show that INTEGER learns behaviors that naive adaptation fails to acquire, recommending once the user's intent is clear and staying attentive to behavioral history at the moment of recommendation. INTEGER also learns an intent-agnostic replacement over the item space, which suppresses rejected items but points to attribute-aware feedback as the next step.
Read the full abstract

All authors Yu-Chen Den, Zhi Rui Tam, Yung-Yu Shih, Shih-Hsin Wang, Yun-Nung Chen, Pu-Jen Cheng, Eugene Yang