Peer Revenue Returns subject to peer review

Papers

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

Computer Science

cs Watch Computer Science

Wednesday 7 October

Machine Learning

MUNITE: Unified Multimodal Latent Inference for Any-to-Any Multimodal Generation

Kyeongmin Yeo and Minhyuk Sung

We introduce MUNITE, a latent-variable framework for flexible any-to-any multimodal generation that treats encoding and latent generation as the same inference problem under different amounts of observed evidence. Given any subset of modalities, MUNITE models the conditional distribution over the latent representation associated with the complete observation. Full observation recovers deterministic encoding, no observation recovers the latent marginal, and intermediate subsets define conditional latent inference, all within a single conditional flow model. A shared latent sample captures variation that must remain consistent across generated targets, while modality-specific generative decoders model the remaining uncertainty independently. To learn these conditional distributions from incomplete training examples, we extend conditional flow matching through self-distillation: predictions conditioned on richer available observations supervise the same model conditioned on smaller subsets at the same intermediate latent state. When the richer-evidence trajectory follows the exact conditional flow, this provides the same expected learning signal as full-target denoising. Across PolyMNIST-D-Q, FFHQ64, and image-text-audio, MUNITE achieves competitive or better generation quality and source-target alignment, with higher joint-generation coherence. In particular, it attains the highest coherence in all one-to-many and unconditional image-text-audio comparisons, showing the effectiveness of unified latent inference across diverse multimodal settings.
Read the full abstract

Computer Vision

Global Average Precision for Representation Learning

Bill Psomas, Mohammad Mahdi, Michalis Thomas et al. (+3 more)

Standard information retrieval metrics, such as mean Average Precision (mAP), assess performance one query at a time, based on how the similarities between a query and its positives compare against those with its negatives. The same holds for common representation learning losses, such as InfoNCE and per-query AP surrogates. None of them considers whether similarities are comparable across queries, which any system with a single decision threshold relies on. Global Average Precision (gAP) does, by ranking all query-candidate pairs in one list and computing a single AP. We introduce gSAP, a differentiable surrogate of gAP. It needs only a similarity matrix and a binary matrix marking the positive pairs, the same input as existing losses, so it is a drop-in replacement for them and agnostic to the encoder, the modality, and the source of supervision. Since it considers all possible pairwise comparisons in the batch jointly, it also remains trainable at low temperatures, a regime where per-query surrogates run out of gradient. Swapping it into established recipes improves supervised metric learning, cross-modal alignment, and self-supervised pretraining, where, to our knowledge, it is the first ranking loss to replace the community standard InfoNCE in the latter two. Its similarities are more consistent across queries, which drives the gains under a universal threshold. gSAP retrieves up to four times as many positive pairs as the strongest AP surrogate at the same precision, and it degrades the least when queries with no positives in the database are added. Beyond thresholding, models trained with gSAP also learn better representations, with higher transfer, $k$NN and zero-shot classification accuracy.
Read the full abstract

All authors Bill Psomas, Mohammad Mahdi, Michalis Thomas, Danda Pani Paudel, Giorgos Tolias, Giorgos Kordopatis-Zilos

Sound · Cross-listed from eess.AS

Toward Part-Aware Choral Transcription with singing voice assignment

Hanyu Meng, Zhanhong He, Zixun Guo et al. (+1 more)

Single-instrument automatic music transcription (AMT) has advanced substantially, yet choral applications require soprano, alto, tenor, and bass (SATB) to be transcribed as separate parts. Recent note-level choral AMT instead produces a single merged note track, limiting rehearsal, education, and score reconstruction. To address this limitation, we introduce Part-aware Choral Transcription (PawCT), to our knowledge the first end-to-end neural framework that identifies active SATB parts from choral audio and transcribes each into a separate note-level track. PawCT combines part-specific onset, offset, and frame prediction with part-presence estimation, union-level supervision, and structured training targets using a range prior (RP) based on SATB pitch ranges and its ordered-continuity (OC) extension, which adds within-part melodic continuity and cross-part pitch ordering. On YouChorale, PawCT-RP-OC achieves a macro part-aware note F1 of 0.225 at a 50-ms onset tolerance, outperforming an adapted choral baseline (0.165) by 36.4% relative and a two-stage post-hoc assignment pipeline (0.175). Its part-agnostic variant, PagCT, achieves a 50-ms onset F1 of 0.382, compared with 0.237 for the previous state-of-the-art choral AMT model. Cross-dataset evaluations on CSD and Cantoria further assess performance under dataset shift. These results demonstrate the benefit of jointly modeling note transcription and vocal-part assignment. Code and demos are available at https://hanyu-meng.github.io/Paw_Choral_AMT_Demo/.
Read the full abstract

All authors Hanyu Meng, Zhanhong He, Zixun Guo, Yaolong Ju

Computer Vision

DeepTopoClustering: Unsupervised Derivation of Surface Process Taxonomy from 4D Point Clouds for Topographic Monitoring

Jiapan Wang, Daan Hulskemper, Mathilde Letard et al. (+2 more)

4D point clouds acquired by permanent laser scanning (PLS) enable accurate high-frequency monitoring of surface change in dynamic topographic environments. However, existing methods remain limited in organizing detected surface activities into meaningful process types. We propose DeepTopoClustering (DTC), an unsupervised framework for deriving a hierarchical process taxonomy from object-based surface activities, so-called 4D objects-by-change (4D-OBCs). We transform each 4D-OBC into a GeoMorphogram, a distributional sequence representing the temporal evolution of topographic change within a spatially bounded surface activity. A convolutional autoencoder learns latent embeddings from GeoMorphograms, which are jointly optimized using a hierarchical deep clustering objective to organize surface activities into a hierarchy. We evaluate the learned hierarchy using expert annotations on two 4D datasets of sandy beach sites and their combination. DTC with GeoMorphograms achieves the highest agreement with expert judgment at the taxonomy level comprising eight major process types ($F_1=0.78$, match accuracy $=0.92$), outperforming dimensionality reduction and conventional flat clustering. The learned taxonomy separates major erosion- and deposition-dominated activities and distinguishes finer subtypes based on change magnitude, duration, compactness, and temporal evolution. DTC thus provides a scalable and interpretable route from 4D change detection to a data-driven, expert-supported surface process taxonomy, advancing automated knowledge derivation for understanding surface dynamics in topographic monitoring.
Read the full abstract

All authors Jiapan Wang, Daan Hulskemper, Mathilde Letard, Roderik Lindenbergh, Katharina Anders

Computation and Language

Training Advisors for LLM Agents from Task Outcomes

Sergei Polezhaev, Barys Liskavets, Ori Press et al. (+1 more)

Large language model agents tackle multi-step tasks by interleaving reasoning and tool calls with observations from the environment. Prior work has shown that natural-language feedback can help these agents revise their decisions during task execution. We introduce Caddie, a method for training critics to provide natural-language analysis and advice as agents work through a task. Unlike approaches that rely on step-level labels or reference critiques, Caddie learns from whether the agent ultimately succeeds after receiving the critic's feedback. We optimize the critic through reinforcement learning while keeping the base model frozen. Trained on multi-hop question answering with a single base model, our Qwen3-4B critic improves success rates across four base models of different scales and architectures, including three not used during critic training. On the MuSiQue benchmark, the trained critic improves Qwen3-4B's success rate by more than 25 percentage points, surpassing the performance of Kimi K3 without a critic. The same critic also yields gains on out-of-domain interactive benchmarks, including $\tau^3$ and DeepDive, with no additional training. Our results show that agents can decide when to seek help from a critic at inference time and that outcome-based critic training can produce guidance that transfers across base models and task domains.
Read the full abstract

All authors Sergei Polezhaev, Barys Liskavets, Ori Press, Alexander Golubev

Robotics

Towards Accurate End-Effector Localization for UMI-Style Robotic Manipulation Teaching

Junjie Zhang, Deteng Zhang, Zhisong Xu et al. (+4 more)

Robot demonstration learning requires accurate and temporally complete end-effector localization during close-range manipulation and camera occlusion. Existing SLAM benchmarks emphasize navigation motions, whereas manipulation datasets prioritize policy learning over localization evaluation. We introduce MILD, a Manipulation-Interface Localization Dataset with real-world and simulation sequences. The real-world subset provides 86 sensor sequences from Insta360 X5 and Insight9 across 15 repeated tabletop tasks, calibration assets, and a per-execution robot end-effector reference trajectory. The simulation subset, MILD-Sim, extends task coverage in Isaac Sim for controlled manipulation-replay studies. Benchmarking visual-inertial and fiducial-aided systems on instrumented real-world recordings reveals large differences in both TCP-relative trajectory error and temporal coverage, even under the same nominal task. To support marker-augmented teaching workspaces without a pre-surveyed fiducial map, we present AprilVINS, which combines fisheye visual-inertial estimation with sequence-local AprilTag geometry and separates prior admission from guarded export of the jointly optimized state. On Insta360 AprilTag4 recordings, AprilVINS(full) under a unified protocol with sequence-specific profiles reaches millimeter-level SE(3)-aligned TCP-relative APE RMSE with high time completion and lower reported error than the tested routes under their respective protocols, whereas fisheye VIO without tag factors remains at centimeter scale. Ablations separate accuracy from exportability, and a MILD-Sim replay study provides task-specific tolerance references for interpreting those error magnitudes. Together, MILD and AprilVINS provide a diagnostic benchmarking framework for UMI-style demonstration collection. Code, datasets, and evaluation manifests will be released upon acceptance.
Read the full abstract

All authors Junjie Zhang, Deteng Zhang, Zhisong Xu, Bo Sun, Liuyang Li, Yihong Tian, Jie Yin

Artificial Intelligence

Self-Evolve With a Reference:Anchored Training of Tool-Integrated Agents

Wenjie Liao, Liangjie Zhao and Zehong Cao

Self-evolving tool-integrated agents learn from tasks and feedback generated within their own training loop. A Curriculum Agent generates tasks, while an Executor Agent learns from self-consistency signals through reinforcement learning. However, relying solely on the current Executor for feedback has two limitations: group-relative advantages vanish under full consensus, while uncertainty-based curriculum rewards favor disagreement without showing whether the generated tasks support further learning. These limitations motivate an additional reference beyond the current Executor. We propose AnchorLoop, which introduces a frozen copy of the previous iteration's Executor as a historical reference and reuses it on both sides of the training loop. For the Executor, the anchor provides a cross-reference advantage that evaluates current outputs against both current and historical majority answers. For the Curriculum, it provides an agreement-based reference based on differences in sampled majority agreement. Since the Executor and anchor have identical parameters during Curriculum training, this comparison serves as a proxy for task selection rather than evidence of inter-version improvement or correctness. Across 13 reasoning benchmarks, AnchorLoop improves over Agent0 by 2.5% on mathematical reasoning and 2.8% on general reasoning tasks. It also maintains higher effective-advantage variance and continues improving in later iterations as the unanchored baseline shows diminishing gains. These results demonstrate the benefit of introducing a lightweight historical reference into self-evolving tool-integrated agents without external task or answer supervision.
Read the full abstract

Computer Vision

DeltaSplat: Iterative Gaussian Refinement for Pose-Free Feed-Forward 3D Gaussian Splatting

Chanung Park, Seunghyeon Song, Joo Chan Lee et al. (+2 more)

Pose-free feed-forward 3D Gaussian Splatting (3DGS) reconstructs a scene from sparse, unposed images in a single network pass, removing the need for camera calibration and per-scene optimization. However, camera estimation errors propagate into the predicted Gaussians and compound the geometric and photometric inaccuracies of single-pass prediction. To correct these errors, we introduce DeltaSplat, a lightweight Gaussian refinement module for pose-free feed-forward 3DGS. It iteratively renders the current Gaussians at the input context views and predicts per-Gaussian updates from the resulting residuals. A 2D residual alone, however, underdetermines the 3D correction. DeltaSplat therefore conditions each update on per-pixel Plücker rays and rendered depth as a soft geometric prior. A dual-branch convolutional mixer efficiently encodes these inputs, and per-attribute heads decode the fused features into position, opacity, and color updates. The module adds only ~2.2% parameters to the backbone and remains fully feed-forward at inference. On DL3DV, DeltaSplat reaches 26.64 dB PSNR in the pose-free setting, improving its state-of-the-art backbone by 1.75 dB and surpassing even baselines supplied with ground-truth cameras; consistent gains hold across 6-24 views and all camera regimes.
Read the full abstract

All authors Chanung Park, Seunghyeon Song, Joo Chan Lee, Eunbyung Park, Jong Hwan Ko

Computer Vision

Hard, Yet Reducible: Controlled Forward Transfer for Synthetic Degradation Curation

Chunming He, Kailai Zhou, Jiaming Zuo et al. (+6 more)

Selecting synthetic degradations for dense prediction requires an estimate of their training utility, the generalization gain they bring under a finite training budget. Clean and degraded twins share content and labels, suggesting a score based on how much short training reduces the excess error caused by degradation. However, this gap can also shrink when clean performance deteriorates. Measuring the improvement on degraded images alone avoids that confound, but it still credits progress that the same amount of clean training would have produced. We propose the controlled Reducible Degradation Gap (cRDG) for regions defined by degradation type and severity. From a common checkpoint, cRDG runs two budget-matched probes that differ only in one augmentation slot, which holds either a synthetic degradation or a clean augmentation. The score is the gain on held-out degraded images relative to the clean-control probe. Clean harm is a separate feasibility constraint. cRDG reveals a correctable severity band in which training on the degradation yields high controlled gain under the available budget, and the band moves with the predictor, the starting checkpoint, and the training budget. Curation of Reducible Bands (\method) uses cRDG to select synthetic data without changing the predictor. On semantic segmentation and salient object detection, \method{} improves representative predictors under matched synthetic-data budgets and training schedules, extends to existing data-generation pipelines, and preserves clean performance. Code and supporting materials will be publicly released.
Read the full abstract

All authors Chunming He, Kailai Zhou, Jiaming Zuo, Hanqi Liu, Fengyang Xiao, Youwei Pang, Xiaofeng Liu, Weisi Lin, Xiaoqi Zhao

Machine Learning

Stream-Based Active Learning with Cooperative Neural Networks for Data-Efficient Partial Inverse Design: An Automotive Glass Run Channel Case Study

Agung Nugraha, Hyerin Kwon, Heungjun Im et al. (+2 more)

Inverse design in engineering often runs into a simple problem. Each labeled training sample must be produced through expensive simulation, so building a large dataset is slow and costly. This study addresses that problem for partial inverse design, where only some design variables are specified and the rest must be inferred to reach a target performance value. We propose CoNN-AL, a framework for data-efficient partial inverse design that adds stream-based active learning to the Cooperative Neural Network with Denoising Autoencoder (CoNN-DAE). The model estimates predictive uncertainty through Monte Carlo dropout and uses it to decide, in real time, which incoming candidate samples are worth labeling, so the limited labeling budget is spent on the most informative designs. We validate the framework on a real-world automotive glass run channel dataset of more than 900,000 unique simulated designs. With only 20,000 actively selected labels, about 2.3% of the training pool, CoNN-AL reaches R-squared values of 0.967 to 0.982 across all missing-variable levels, approaching the upper-bound models trained on far more data. It reaches R-squared of at least 0.95 with 30 to 40% fewer labels than random sampling at the more difficult missing-variable levels and, at the most challenging level, is the only strategy in this study to reach R-squared of 0.98. Together with this work, we publicly release the dataset to support future research on data-driven design.
Read the full abstract

All authors Agung Nugraha, Hyerin Kwon, Heungjun Im, Gian Antariksa, Jihwan Lee

Graphics

DynaConTalk: Wavelet-Constrained Diffusion for Long-Form and Controllable Holistic Co-Speech 3D Motion

Yifei Zhu, Yangyang Cai, Mingyi Shi et al. (+4 more)

Holistic co-speech animation is prone to averaging in both motion representation and speech conditioning. In coordinate-space diffusion, slow body posture, mid-frequency gesture strokes, and fast hand or facial details are entangled in one prediction target, often producing low-variance, over-smoothed motion. Meanwhile, dense rhythmic and acoustic cues can dominate sparse content-specific information under fixed multimodal fusion. We present DynaConTalk, a wavelet-constrained diffusion framework for long-form and controllable holistic co-speech motion generation. Diffusion operates in stationary wavelet transform (SWT) coefficient space, whose temporally aligned bands separate coarse posture evolution, gesture strokes, and fine expressive details. Our dynamic gating network preserves HuBERT and speaker identity as a base and selectively adds rhythm, mel, and transcript features through motion-state- and noise-aware residual gates. Attention pooling and learned depth routing deliver complementary conditions to each denoising stage, while a frame-resolution rhythm path preserves precise timing. A signed proposal-consensus update then reconciles these conditions with the evolving motion state. Matched-noise constraint injection uses the same sampling interface for history continuation and localized keypose repair, and extends to reference-guided control. Separate body-hand and facial denoisers, followed by inverse SWT and a pose-driven root regressor, produce holistic motion. Experiments evaluate generation quality, facial accuracy, temporal continuity, and controllable editing. Code, models, and the interactive editing interface are available at https://github.com/zhuyifeiabcd1/DynaConTalk.
Read the full abstract

All authors Yifei Zhu, Yangyang Cai, Mingyi Shi, Miao Cheng, Lin Gu, Taku Komura, Yoshifumi Kitamura

Machine Learning

For Those Who Believe in Faithfulness: Optimizing the Area Under Insertion and Deletion Curves for Ranking Relative Feature Importance

Bjørn Leth Møller, Bulat Ibragimov and Christian Igel

The adoption of machine learning for socially relevant tasks requires effective explainable artificial intelligence (XAI) methods to better understand the behavior of machine learning models. Attribution methods are a popular XAI approach in which input-output relationships are characterized by heat maps that reflect the relative importance of input features for a particular prediction. The quality of such maps is often assessed by measuring faithfulness based on the area under insertion and deletion curves, which measures changes in the model output as features are added and removed. In this study, we derive an objective function from this notion of faithfulness and a way to approximate its gradient. We establish the connection between insertion curves and top-$k$ feature selection, which leads to a loss function measuring the quality of attributions. Randomization of the loss allows us to efficiently approximate its gradient. To show the effectiveness of the general approach, we combine the loss function with the neural explanation mask framework. The resulting method, termed Ra-NEM, can be used with any differentiable model without affecting the model's performance. Experiments demonstrate that Ra-NEM provides accurate attributions robustly and efficiently. Compared to other algorithms, the attributions have not only higher faithfulness but also perform well in terms of other XAI metrics. The high inference speed of Ra-NEM makes the method suitable for online applications. The code is available online: https://github.com/baerminator/Ra_Nem
Read the full abstract

Game Theory

Minimizing Cumulative Envy in Allocating a Sequence of Items

Paul W. Goldberg, Isaac Robinson and Nicholas Teh

We study temporal fair division with indivisible goods that arrive sequentially and must be allocated irrevocably. In contrast to the usual online model, we assume that valuations and future arrivals are known in advance, and ask how unfairness evolves during the process. We introduce cumulative maximum envy: the sum, over all rounds, of the maximum pairwise envy at that round. Equivalently, this is the area under the worst-envy curve, and it captures both the magnitude and the duration of envy. For a fixed arrival order, we show that the corresponding decision problem is strongly NP-complete and that minimizing this objective admits no constant-factor approximation unless P = NP, even under identical valuations and even under binary valuations. We complement these hardness results with a dynamic program that gives pseudopolynomial-time solvability for a constant number of agents, polynomial-time algorithms in further restricted settings, and an FPTAS for fixed $n$ under identical integer valuations. We then study a sequencing variant where the algorithm may choose the arrival order. This variant remains NP-complete even for two agents with identical valuations; however, a simple greedy algorithm achieves a $3/2$-approximation for $n=2$ agents, an $n/(n-1)$-approximation for any number of agents, and an additive guarantee depending on the maximum value of any good.
Read the full abstract

Human-Computer Interaction

Empowering Users in Graph Rule Mining via Large Language Models

Francesco Cambria, Francesco Invernici, Andrea Colombo et al. (+1 more)

In the era of interconnected data, graphs have emerged as an effective abstraction for modeling complex systems in an intuitive format, especially with the rise of Property Graphs, which offer an intuitive and scalable way of navigating non-intuitive structures. In this context, graph mining techniques have been developed for testing complex graph-based rules, as the MINE GRAPH RULE operator, which, however, require users to have prior expertise both in graph theory and formal query language. In this work, we propose to bridge the gap between users and the graph-association rule-mining process by showing how Large Language Models (LLMs) can be easily prompted to formulate, refine, and interpret complex relational rules, directly producing MINE GRAPH RULE queries.
Read the full abstract

All authors Francesco Cambria, Francesco Invernici, Andrea Colombo, Anna Bernasconi

Computer Vision

ORCA: Hunting Compositional Failures in Text-to-Image Diffusion

Arshia Hemmat, Amirhossein Vahidi, Amitis Shidani et al. (+4 more)

Text-to-image diffusion models fail predictably on compositional prompts: attributes bind to the wrong objects, spatial relations invert, and multi-object scenes lose count. Recent architectures already augment CLIP with a T5 encoder precisely because CLIP's contrastive embedding loses compositional structure, yet these failures persist. We argue the binding problem is therefore not one of missing information but of misaligned information: a text encoder preserves compositional structure, but in a representation space shaped by language modelling rather than vision, and the denoising objective does not directly reward aligning the two. We show this correspondence can be supplied as an explicit training signal, that the relevant cross-modal information is concentrated in a low-rank subspace of self-supervised visual features, and that supplying it can be folded into diffusion training as a single auxiliary loss. Our method, ORCA (Orthogonal Residual Compositional Alignment), aligns the latent of a diffusion transformer with a low-rank target derived from a frozen visual encoder, through a predictor whose orthogonal basis is parameterised by a learned residual between T5 and CLIP embeddings, which provides a prompt-dependent signal for selecting the visual readout subspace. We prove that the cross-modal information recoverable at a given rank is bounded by the spectral mass of the visual encoder's covariance in the top components. Across three diffusion-transformer backbones (DiT-B/2, DiT-L/2, U-ViT-L), ORCA improves FID and GenEval over both vanilla and REPA baselines at zero inference-time cost; on DiT-L/2 it reaches FID 16.65 and GenEval 0.291 at 200K steps, exceeding the strongest 400K baseline at half the training cost, with the largest gains concentrated on attribute binding, spatial relations, and multi-object prompts.
Read the full abstract

All authors Arshia Hemmat, Amirhossein Vahidi, Amitis Shidani, Mohammad Vali Sanian, Hesam Asadollahzadeh, Aryan Yazdan Parast, Mohammad Lotfollahi

Machine Learning

Fully Interpretable Minimal Transformers: From Geometry to Algorithm

Raneem Mahajne and Toviah Moldwin

We present a framework for building and interpreting minimal transformer models. By constraining a transformer's embedding dimension and head size to 2, we enable full two-dimensional visualization of its internal representations. Embeddings, query/key/value transforms, attention outputs, residual streams, and decision boundaries can all be seen directly. Our central claim is that the learned geometry implies an algorithm; the arrangement of points and boundaries in R^2 can be read as a step-by-step procedure. We train a transformer on a simple task where it must produce the most recently observed even number whenever the '+' operator appears in a sequence of digits. Once trained, we visually walk through every step of the transformer's computation. We show how the model embeds the tokens and their respective positions in the sequence, transforms them via the Q, K, and V matrices, uses the dot product between the Q and K representations to form the attention matrix, and uses the attention matrix to select values that move the representation of each input token to the region of the domain of the output layer that will correctly predict the next token. We introduce a suite of interpretability visualizations that make the algorithmic interpretation of this procedure explicit. Our framework offers a pedagogical and experimental testbed to explore how transformers use informational geometry to implement next-token prediction.
Read the full abstract

Machine Learning · Cross-listed from cond-mat.mtrl-sci

Origins of Universal Machine Learning Force-Field Errors in Multicomponent Materials

Hongwei Du, Dingyang Lv, Baole Wei et al. (+11 more)

Universal machine learning force-field generalization to multicomponent environments generated by compositional design remains insufficiently assessed. We construct a benchmark of 7,599 multicomponent configurations inspired by high-entropy design, elemental substitution and anion mixing. Eleven pretrained models are evaluated against density functional theory for energies, forces and stresses, with assessment extended to elastic, vibrational and adsorption-related properties. Force errors are analysed through training-reference coverage, local geometric heterogeneity, distance directionality and elemental response. Distances to training-reference environments reveal a qualitative association between coverage differences and increasing errors, while substantial variation remains at similar distances. Higher-error groups show greater local geometric heterogeneity, although OMat24 provides broad coverage of these environments. Relative to training-reference pair medians, errors remain low near the median, rise steeply on the compression side and increase more weakly on the extension side. After matching element pairs and absolute distance deviations, compression-side force errors are 1.81-1.95 times extension-side errors. Model-predicted pairwise interaction curves show greater curvature under compression. Fitting difficulty in independent elemental systems correlates with electronic band-energy responses to atomic displacements and Fermi-level shifts, and a similar pattern is observed in multicomponent systems. In parameter-matched comparisons, spherical-harmonic representations with maximum degrees of 2 and 4 lower test force errors for 38 and 40 of 43 elements, respectively, while differences in elemental difficulty remain. These findings inform force-field selection for experimental compositional design and identify targets for training-data sampling and model representations.
Read the full abstract

All authors Hongwei Du, Dingyang Lv, Baole Wei, Yu Ren, Feng Yu, Xin He, Bonan Zhu, Jiahui Liu, Yongda Huang, Yongheng Li, Jianjun Liu, Siqi Shi, Hong Wang, Ziheng Lu

Computation and Language

A Deafening Silence: Catastrophic Forgetting Lives in the Output Embeddings of Tokens the Data Never Speaks

Jonghyun Han, Younghoon Song and Jongyoul Park

Continual pre-training and fine-tuning in Large Language Models (LLMs) inevitably induce catastrophic forgetting, typically mitigated by replay using often-inaccessible original data. In this data-free regime, we analyze where forgetting occurs and why. Systematic parameter freezing across five settings up to 1.4B reveals that forgetting concentrates selectively in the output embeddings of tokens rarely seen in the new corpus, whereas the same sqrt(v-hat) band of the body is inert and new learning resides elsewhere. This localization is governed by the vocabulary deficiency of the corpus rather than the training mode, allowing pre-retraining risk ranking from token counts alone within a fixed base model. Mechanistically, absent tokens receive persistent one-sided softmax gradients that Adam's second-moment (sqrt(v-hat)) normalization amplifies into full-sized updates. We therefore propose an intervention: raising Adam's epsilon exclusively for the output projection during training. Across eight settings spanning 160M to 12B parameters and four model families, this removes 39.4% to 67.9% of forgetting across all seven stable configurations without degrading target learning or requiring per-model tuning. The defense combines additively or better with replay (79.8% on Qwen/Korean) and rescues released-head LoRA from a 23-fold forgetting surge. Because post-hoc editing of the drifted rows recovers under 5% of forgetting, the intervention must operate during training. Our findings indicate that a single-line optimizer adjustment may serve as the primary defense against catastrophic forgetting where the corpus starves the vocabulary.
Read the full abstract

Artificial Intelligence

SkillForge: Co-Evolving Skills and Agents via Dynamic Skill Lifecycles

Yuyao Ge, Yiwei Wang, Yuchen He et al. (+5 more)

Memory-augmented reinforcement learning strengthens LLM agents' ability to solve complex long-horizon tasks. Skills are one such form of memory, pairing instructions with an applicability condition over task types. However, retaining every skill indiscriminately as the policy improves lets obsolete or harmful entries accumulate and mislead the agent. We propose SkillForge, an agentic RL method that compiles and evolves the skill library through a fitness-driven skill lifecycle of trial, active, stable, and retired states, so that the skills and the model co-evolve throughout training. A pre-RL evaluation phase first uses the base model's own rollouts to pre-retire low-fitness skills, yielding a filtered library that then seeds supervised fine-tuning. Reinforcement learning takes over from this checkpoint, and at each iteration selective retirement, stabilization, and LLM-guided mutation continue to forge the skill library alongside policy optimization. Across multiple interactive agent benchmarks, SkillForge achieves the highest aggregate success rate, delivering up to 7.8% relative improvement over the strongest baseline while keeping the skill library compact throughout training. We introduce SkillFurnace, a dataset of 5k+ annotated records bundling retirement-filtered SFT trajectories, evolved skill libraries with fitness annotations, and retirement events with human-annotated failure categories to support research on skill quality and lifecycle management.
Read the full abstract

All authors Yuyao Ge, Yiwei Wang, Yuchen He, Baolong Bi, Lingrui Mei, Jiayu Yao, Lizhe Chen, Shenghua Liu

Sound · Cross-listed from eess.AS

Training-Free Instruction TTS Gender Bias Calibration Using Model-Adaptive Steering

Kuan-Yu Chen, Yi-Cheng Lin, Jeng-Lin Li et al. (+1 more)

Instruction text-to-speech (ITTS) systems systematically encode acoustic gender skews from descriptive style prompts, such as occupations or personas, even when demographic attributes are left unspecified. Calibrating the implicit gender distribution of the synthesized voices, across heterogeneous architectures and without model retraining or overriding explicit user prompts, is an open problem. In this work, we propose model-adaptive steering, a training-free bias calibration method that steers post-encoder conditioning representations using a group deviation vector paired with a coarse-to-fine strength search on a development set. A deterministic lexical gate bypasses intervention whenever explicit gender keywords are detected, preserving intended prompt semantics. Evaluated on a 12,800-prompt held-out benchmark across four ITTS models (three architectures), the proposed method reduces aggregate calibration error from 11.5–28.1 to 0.8–5.9 percentage points (e.g., shifting Parler-TTS Mini from 78.1% and PromptTTS++ from 27.3% female to 50.8–52.4%). Output rates stay within 0.9 pp across anchor set sizes at fixed operating points, while UTMOS decreases by at most 0.08 and WER degrades by at most 3.3 pp.
Read the full abstract

All authors Kuan-Yu Chen, Yi-Cheng Lin, Jeng-Lin Li, Jian-Jiun Ding

Computer Vision

MOTIF: Person-of-Interest Deepfake Detection Beyond 3DMM Coefficients

Giovanni Affatato, Sara Mandelli, Paolo Bestagini et al. (+1 more)

Video deepfakes targeting a specific individual, the Person-of-Interest (POI), are the most harmful ones, and, since a public figure is abundantly recorded, a detector can be built from genuine footage of that individual. Such detectors commonly describe a subject through a 3D Morphable Model (3DMM) and adopt its coefficients as a whole, so which part of that description carries the signal has never been measured. We dissect it, holding the encoder, the training corpus and the enrollment protocol fixed and varying only what the encoder observes. The groups of coefficients prove largely redundant, since the shape block alone recovers almost all the accuracy of the full vector, and their temporal evolution contributes a real but bounded amount. We further show that the dense surface the same fit returns, which these detectors discard, carries identity information that the coefficients do not, and that it helps precisely where they are weakest. We assemble the best configuration into MOTIF, a visual-only detector trained on real videos only, with no manipulated video and no POI-specific data. It improves on both state-of-the-art POI detectors in every dataset and manipulation of our benchmark and at two quality levels. Our experimental code will be released at https://github.com/polimi-ispl/MOTIF.
Read the full abstract

All authors Giovanni Affatato, Sara Mandelli, Paolo Bestagini, Stefano Tubaro

Data Structures and Algorithms

Packing Diverse Shortest Cycles

Akanksha Agrawal, Fedor V. Fomin, Petr A. Golovach et al. (+3 more)

Bentert, Fomin, Golovach, Korhonen, Lochet, Panolan, Ramanujan, Saurabh, and Simonov (SODA 2025) initiated the parameterized study of Edge-Disjoint Shortest Cycle Packing: given a weighted graph $G$ and an integer $k$, decide whether $G$ contains $k$ edge-disjoint cycles of minimum weight. They showed that the problem admits an algorithm running in time $n^{O(k^6)}$ and asked whether it is fixed-parameter tractable or $W[1]$-hard parameterized by $k$. We resolve this question by proving that Edge-Disjoint Shortest Cycle Packing is $W[1]$-hard parameterized by $k$, even on unweighted subcubic graphs. The same lower bound also applies to the vertex-disjoint variant. For planar graphs, they provides a construction of a kernel with $O(k^2)$ vertices and an algorithm running in time $k^{O(k)} \cdot n^{O(1)}$, and explicitly asked whether the problem admits a single-exponential algorithm of running time $2^{O(k)} \cdot n^{O(1)}$. Rather than addressing this question in isolation, we introduce a more general framework, Diverse Shortest Cycle Coverage, which asks for $k$ shortest cycles that may overlap in a controlled way while maximizing the total weight of covered edges. This framework simultaneously captures edge-disjoint shortest cycle packing, the problem of finding diverse shortest cycles, and the problem of maximizing edge coverage by shortest cycles. Our main algorithmic result shows that Diverse Shortest Cycle Coverage can be solved on planar graphs in time $2^{O(k)} \cdot n^{O(1)}$, thereby giving a single-exponential algorithm for Edge-Disjoint Shortest Cycle Packing as a special case. The key idea of our algorithm is a structural analysis of the Laminar Shortest Cycles Tree, a tree-like decomposition that reveals a laminar interaction pattern among shortest cycles in planar graphs and enables an efficient dynamic programming algorithm.
Read the full abstract

All authors Akanksha Agrawal, Fedor V. Fomin, Petr A. Golovach, Vinod Gupta, Yash Hiren More, Vidya Sagar Sharma

Robotics

Enhancing Robotic Perception and Adaptability through Sensor Fusion and Origami-Inspired Designs

Namai Chandra, Jaison Jose, Kavi Arya et al. (+1 more)

Compact mobile robots must recover scene geometry under changing lighting and surface texture while working within tight payload and cost limits. We present a compact mobile robot that uses origami-inspired wheels for locomotion and active control of its sensing geometry. As the wheels move between terrain-adaptive configurations, the changing chassis pitch sweeps a 2D LiDAR through intermediate elevations; held wheel positions provide a chosen viewing angle. An IMU accounts for chassis attitude, and a fusion node projects LiDAR returns into the RGB-D depth stream supplied to RTAB-Map. The arrangement uses the wheel actuation already present on a sub-300 USD, sub-2 kg prototype to extend the scanner's viewing geometry. We assess depth fusion in a textureless indoor corridor and an outdoor sunlit area, with three runs per sensor configuration in each setting. Mean full-frame invalid-depth fractions fell from 21% to 11% indoors and from 48% to 18% outdoors. The prototype combines improved depth coverage with a continuously adjustable LiDAR viewpoint using the same actuation that reconfigures its wheels.
Read the full abstract

All authors Namai Chandra, Jaison Jose, Kavi Arya, Shivaram Kalyanakrishnan

Machine Learning

Dual-QK: Sharp Queries and Flat Keys for Prunable 2-bit KV Caches

Sunjoo Whang, Jungjun Oh, Minsung Kim et al. (+5 more)

Long inputs and extended generation increase the storage and access costs of the key-value (KV) cache. Low-bit quantization reduces storage and memory traffic, while query-channel pruning can further reduce key-cache reads. Rotation-based quantization redistributes the energy of key outliers across channels. To maintain computational invariance, the same orthogonal transform must be applied to queries, preserving query-key dot products. However, this rotation can disperse query energy, weakening the separation between a few large components to retain and many small ones to prune. We introduce Dual-QK, which uses paired non-orthogonal query and key transforms to address this conflict. Using calibrated query and key statistics, Dual-QK combines partial key whitening with a query-aligned basis to balance key scales for INT2 quantization and concentrate query energy for dynamic channel pruning. Channel-0 protection and bucket-relative RoPE support low-bit accuracy over long contexts. Experiments on four models across five generative benchmarks and long-context retrieval tasks show improved accuracy over OSCAR on most tasks at 40% query-channel sparsity. At a 128K context, Dual-QK provides $6.8\times$ KV-cache compression and an estimated $8.3\times$ reduction in KV read volume relative to unpruned BF16. Under the evaluated configurations, our SGLang implementation achieves up to $3.75\times$ the decoding throughput of unpruned BF16.
Read the full abstract

All authors Sunjoo Whang, Jungjun Oh, Minsung Kim, Dongho Seo, Jisu Shin, Gregory Kielian, Hoi-Jun Yoo, Sangjin Kim

Cryptography and Security

Hybrid Hierarchical Runtime Verification for Edge-IoT Security: Combining MonPoly and RTLola

Nikolaos Kekatos, Marinelio Chintri, Panagiotis Katsaros et al. (+5 more)

Security monitoring of edge-IoT fleets faces three structural challenges. (i) A per-node monitor is cheap but cannot see attacks that coordinate across devices. (ii) A cloud monitor sees the full fleet but pays for that view in bandwidth. (iii) Even at the cloud, a monitor built on a single RV engine can be fooled by an attacker who compromises a device, raises one malicious request, and then goes silent: once the events stop, an event-triggered monitor has nothing left to evaluate. We propose a three-layer hierarchical runtime-verification framework that addresses all three. The edge layer classifies events as they happen, the gateway layer aggregates short windows of per-device behaviour, and the cloud layer runs two complementary RV engines. MonPoly handles first-order temporal correlation over the merged alert stream: coordinated overflow (which genuinely quantifies across devices) plus per-device multi-vector APT, escalation, and persistent-campaign patterns. RTLola handles a time-triggered silent-node property that an event-triggered engine cannot detect within a bounded delay under fleet silence. We evaluate the framework on a 15-actor Docker testbed covering eight attack profiles plus a silent-bypass scenario. In the controlled labelled testbed, every device-attributable incident the framework raises names an attacker-labelled device, and the RTLola tier catches silent-bypass attempts the event-triggered tier misses. Per-event monitoring stays in the microsecond range at the edge and gateway, with low end-to-end alert-to-incident latency at the cloud.
Read the full abstract

All authors Nikolaos Kekatos, Marinelio Chintri, Panagiotis Katsaros, Alexios Lekidis, Tom Nianios, Ioannis Seitoglou, Anastasios Temperekidis, Stylianos Basagiannis

Machine Learning

Homogenization in Multi-Agent Systems

Prakhar Ganesh, Kyra Wilson, Luca Zappella et al. (+4 more)

Multi-agent systems (MAS) leverage interactions between agents to perform complex tasks. Despite their success, we show that these interactions can also lead to homogenization, i.e., agents converging to similar behaviors. Homogenization in MAS can reduce agent diversity and reinforce shared failures. In this paper, we operationalize homogenization using three metrics: conformity to the majority, polarization towards extremes, and growing inertia against changes over subsequent interactions. We evaluate homogenization in MAS for code generation, hiring, and scientific peer review. Across these tasks, we show that homogenization translates to concrete downstream risks: in code generation, it hides and amplifies correlated errors which can create systemic vulnerabilities; in hiring, it allows the influence of biased agents to persist long after their removal; and in peer review, it creates uneven evaluation standards across research areas. Our results establish homogenization as a failure mode of MAS, demonstrating that MAS evaluations must move beyond aggregate performance to carefully analyze interaction dynamics. Finally, we show that simple approaches to increase diversity—leveraging sampling stochasticity and mixed-models MAS—fail to reduce homogenization risks, highlighting the need for strategies to effectively leverage agent diversity.
Read the full abstract

All authors Prakhar Ganesh, Kyra Wilson, Luca Zappella, Barry-John Theobald, Nicholas Apostoloff, Lucas Monteiro Paes, Nivedha Sivakumar

Computer Vision

UltraText Bench: A Comprehensive Bilingual Benchmark for Evaluating Visual Text Rendering in Image Generation

Deyuan Liu, Yihao Hu, Jingxuan Zhang et al. (+13 more)

Dense visual text requires image generators to reproduce long strings across multiple regions with correct placement and legibility. As short-string rendering improves, evaluation must test sustained performance across more demanding scenes. We introduce UltraText Bench, a bilingual benchmark for prompt-only generation of dense visual text. It contains 432 prompts spanning 24 real-world scene categories and three difficulty levels, split equally between English and Chinese. Each human-reviewed prompt supplies exact strings for four to twelve text regions, paired with structured references for their content, placement, and visual attributes. We use the Q-Judger vision-language model to assess each image against the complete reference, reporting text fidelity, text clarity, spatial quality, and scene quality. Across 24 model configurations, these dimensions reveal different strengths: Z-Image-Turbo gains 3.81 clarity points over Z-Image-Base while losing 14.76 fidelity points under the reported settings. Performance also varies with workload; Qwen-Image-2512's English composite falls from 86.50 at L1 to 42.86 at L3. Ten participants took part in human evaluation of the automatic scores. Repository: https://github.com/LINs-lab/UltraText_Bench.
Read the full abstract

All authors Deyuan Liu, Yihao Hu, Jingxuan Zhang, Xingying Li, Jun Xie, Jiacheng Liu, Jungang Li, Yu Huang, Xuanyi Liu, Yue Ding, Zecheng Wang, Lei Zhao, Mingda Wang, Zhenglin Cheng, Peng Sun, Tao Lin

Graphics

Simulation Methods for Multiphysics Phenomena in Visual Computing

Fabian Löschner, Stefan Rhys Jeske, José Antonio Fernández-Fernández et al. (+1 more)

Physics simulation is a cornerstone of many computer graphics applications, ranging from video games and virtual reality to visual effects and computational design. The number of techniques for physically-based modeling and animation has thus skyrocketed over the past few decades, facilitating the simulation of a wide variety of materials and physical phenomena. These course notes provide an in-depth introduction to multiphysics simulation methods for computer graphics, covering the mathematical foundations, key algorithms, and practical considerations behind the most widely used approaches. We focus on methods developed by the computer graphics community for simulating various physical phenomena and materials – including rigid and deformable bodies, fluids, and granular materials – as well as the interactions between them. For each method, we present the underlying mathematical framework with detailed derivations and discuss how different materials and coupling strategies fit into the formulation. A selection of software frameworks that offer out-of-the-box multiphysics modeling capabilities is also presented. Finally, we touch on emerging trends in physics-based animation, including machine learning-based methods which have become increasingly popular in recent years.
Read the full abstract

All authors Fabian Löschner, Stefan Rhys Jeske, José Antonio Fernández-Fernández, Jan Bender

Computer Vision

Efficient 3D Gaussian Head Avatars for Edge Devices

Umar Farooq, Jean-Yves Guillemaut, Adrian Hilton et al. (+1 more)

Generative 3D Gaussian head avatars provide high-quality, efficient rendering, but synthesising the Gaussian representation remains computationally expensive, limiting deployment on resource-constrained and edge devices. We introduce an efficient generator architecture for unconditional 3D Gaussian head synthesis, based on a parameter-efficient synthesis block and depth-wise separable convolutions while retaining style-based conditioning. Our architecture reduces generator complexity without requiring model compression or quantisation. Compared with the baseline model, our approach reduces FLOPs by 94%, parameter count by 70%, and model size by 81%, while maintaining competitive generation quality. We further demonstrate practical CPU inference and browser-based execution on mobile devices using ONNX Runtime, enabling 3D Gaussian avatar synthesis without dedicated GPU hardware or application-specific software. In addition to conventional image-quality metrics, we evaluate multi-view consistency, training cost, and deployment performance. Code, trained models, and evaluation tools will be released publicly.
Read the full abstract

All authors Umar Farooq, Jean-Yves Guillemaut, Adrian Hilton, Marco Volino

Information Retrieval

The Impact of Backbone Evolution on LLM-Based Relevance Assessments

Chuting Yu, Guido Zuccon and Teerapong Leelanupab

LLMs are evolving rapidly, with newer models offering stronger capabilities. This suggests that in LLM-based relevance judging, more capable models will achieve higher agreement with human judgements under the same prompt. We challenge this understanding by investigating the behavior of LLM-based relevance judges under backbone evolution. Keeping the prompts fixed, we evaluate a representative single-prompt (UMBRELA) and a rubric-based prompt (EXAM) across sequential model versions of commercial (Gemini, GPT) and open-weight (Qwen, Llama) models. Overall, we find no consistent evidence that newer versions lead to better relevance judges. Crucially, similar or improved aggregate performance does not imply judgment stability: correct judgements made by an earlier version of an LLM backbone are not necessarily preserved by later versions. We investigate the potential drivers of these regressions. Our findings caution against the assumption that judging prompts designed and validated for one backbone version will perform equivalently or better when the model is updated, even within the same family.
Read the full abstract