https://arxiv.org/api/sSKnxKUVUq6VnPxpZe6Hadc5mOU 2026-09-06T22:13:28Z 300 15 0 http://arxiv.org/abs/2609.03584v1 2026-09-03T09:32:15Z

Synthetic chord corruption provides a controlled stress test for singing accompaniment generation (SAG), whereas complete automatic chord recognition (ACR) replay measures the condition delivered to a deployed system. We compare them by replaying CNN-CRF and DeepChroma+CRF predictions through one fixed MIDI-SAG generator, holding track, seed, context, and scoring window constant. Across 30 paired tracks and three seeds, a central four-second tritone produced larger changed-target and inside-output effects than CNN-CRF replay in STFT, CQT, and CENS; the CENS target gap was positive on 29/30 tracks (mean 0.462). Matching replay support and relation composition reduced this mismatch, with joint matching giving the lowest replay distance in the full-30 analysis. Relative-root substitutions produced 2.88-fold larger CENS full-window output change than same-root quality flips at near-equal dose. The matched surrogate was closer to replay for both recognizer paths. We conclude that localized corruption tests mechanisms, whereas complete replay evaluates deployed chord-condition propagation.

2026-09-03T09:32:15Z Weiwen Huang http://arxiv.org/abs/2609.03498v1 2026-09-03T08:01:34Z

Faster playback has become a common feature in modern online audiovisual services, allowing users to consume content in less time while still maintaining a coherent viewing experience. However, different modalities of media content, such as video, audio (including speech and music), and audiovisual, exhibit varying requirements for understandability and information integrity under faster playback. These differences lead to noticeable variations in perceived quality depending on the content type. Nevertheless, users' perceived quality at different playback speeds remains insufficiently investigated. To address this gap, this paper conducts a series of subjective experiments to analyze the relationship between playback speed and perceived quality for video, audio, and audiovisual content. Content-specific intrinsic features are extracted to capture temporal dynamics, including temporal information (TI) for video, words per minute (WPM) for speech, and beats per minute (BPM) for music. Predictive models of perceived quality under faster playback are then developed separately for video, speech, and music. By integrating these models, a unified content-adaptive perceived quality assessment model (CAPQ-FAST) is proposed for faster audiovisual playback. Experimental results demonstrate that the proposed model effectively predicts perceived quality under different playback speeds. This model can help service providers better understand users' viewing intentions and perceptual experiences under faster playback, thereby enabling more adaptive personalized recommendations and playback control to enhance user experience adaptability.

2026-09-03T08:01:34Z IEEE Transactions on Circuits and Systems for Video Technology, doi: 10.1109/TCSVT.2026.3716481 Jiarun Song Yuxin Song Fuzheng Yang Weisi Lin 10.1109/TCSVT.2026.3716481 http://arxiv.org/abs/2609.03415v1 2026-09-03T06:23:50Z

Automatic generation of hand gestures is essential for the transmission of Indian classical dance and critical for its preservation. Indian classical dance gesture datasets are inherently low-resource, and the canonical Sanskrit definitions of many mudras lack precise textual descriptions, limiting the effectiveness of conventional text-conditioned image generation models. We present \textbf{MudraGen}, a conditional diffusion framework that synthesizes realistic RGB images of \textit{Samyukta Hasta Mudras} -- interactive two-hand gestures from Bharatanatyam (an Indian classical dance form). Unlike prior work on simple hand signs or single-hand gestures, MudraGen introduces geometry-aware supervision to capture the precise coordination, anatomical validity, and cultural nuance of interacting hands. We formulate three geometry-aware objectives: Keypoint Loss for 3D joint alignment, Joint Offset Loss for inter-hand spatial coherence, and Shape Consistency, which serves as an anatomical regularizer by encouraging consistent hand morphology while allowing independent hand poses. Together, these objectives guide the diffusion model toward anatomically plausible and well-coordinated hand configurations, enabling the synthesis of photorealistic and pose-accurate gesture images. Experimental results show that MudraGen surpasses existing state-of-the-art generative approaches in visual realism, anatomical correctness, and preservation of fine hand-pose structure, enabling faithful reproduction of complex Samyukta Hasta mudras. Beyond quantitative gains, its ability to generate culturally grounded and structurally consistent gestures highlights practical applications in cultural preservation and dance education.

2026-09-03T06:23:50Z Accepted for publication in ACM Journal on Computing and Cultural Heritage (JOCCH) Special Issue on Visual Heritage Jagadish Kashinath Kamble Jayanta Mukhopadhyay Debaditya Roy Partha Pratim Das http://arxiv.org/abs/2609.02751v1 2026-09-02T15:51:26Z

As generative AI tools become increasingly powerful and easy to use, people can easily edit portrait images with a prompt, necessitating the task of image editing attribution, which predicts the involved editing tools from the given image. Existing attribution methods hold the single-tool assumption and can only attribute a specific editing tool, but struggle to handle the more complex and increasingly common multi-tool editing scenarios, where artifacts left by different editing tools are composite and overlapped. To address this gap, we explore Multi-Tool Image Editing Attribution (MIEA), which aims to identify multiple editing tools involved in a multi-tool edited facial image. To simulate the real-life editing operations on facial images, we then construct a new dataset, MultiEdit, which contains 500k+ edited facial images and covers six types of editing tools that support face swapping (Deepfake) and various facial enhancements. Inspired by the findings from data analysis, we design DPEC, a multi-tool attribution method that can capture distinguishable, locality-aware editing tool traces from both spatial and frequency domains with the support of an error-based curriculum learning strategy. Experiments show \Method\ outperforms nine methods for facial images edited in at most five steps.

2026-09-02T15:51:26Z Accepted to ACM Multimedia 2026 (MM 2026) Sheng Liu Qiang Sheng Danding Wang Yu Li Chenming Zhou Juan Cao 10.1145/3767308.3835066 http://arxiv.org/abs/2609.02367v1 2026-09-02T09:39:50Z

Joint audio-video generation models have made substantial progress in visual quality and audio-visual synchronization. However, they still provide limited control over when shot transitions occur and dialogue is spoken. This limitation constrains their application in script-driven content creation, where timing errors can undermine narrative coherence and the viewing experience. Current joint generators align video and audio representations on a shared temporal axis, yet the precise timing of shots and dialogue specified in a structured prompt is encoded only in the prompt's text representation and remains unaligned with the temporal coordinates of either modality. Consequently, video and audio may remain synchronized with each other while both fail to follow the script timeline. This mismatch motivates us to extend temporal alignment beyond video and audio to include the structured script. We therefore introduce Temporal Context Routing (TCR), which maps the script timing onto the shared temporal axis of video and audio generation and routes each prompt's guidance to the corresponding positions in both modalities. Compared with the baseline on 200 test scripts, TCR reduces Shot Boundary MAE by 96%, from 1.11 s to 0.042 s, and raises Dialogue Acc@0.5 s from 28.3% to 84.1%. TCR achieves these improvements while maintaining visual quality and audio-visual synchronization comparable to those of the baselines. A user study further shows that participants prefer TCR on all five evaluated dimensions.

2026-09-02T09:39:50Z Yichen Liu Quanwei Zhang Haozhe Wang Donghao Zhou Xiaojie Li Yang Shi Jiaming Liu Ruihua Huang Yingtian Zou Daquan Zhou http://arxiv.org/abs/2609.02343v1 2026-09-02T09:18:27Z

Recent advances in audio-language modeling have been driven by large-scale audio captioning datasets. However, existing datasets remain limited by low semantic diversity, generic descriptions lacking acoustic details, and one-to-one audio-caption mappings that poorly reflect the inherent ambiguity of auditory perception. We introduce SonicCaps, a large-scale audio captioning dataset comprising ~15M captions paired with ~700k audio clips, generated using a multi-modal large language model (Qwen3-Omni) conditioned on both audio and text. To explicitly promote diversity, we generate around 24 captions per audio via structured prompt engineering and few- shot generation, spanning main descriptions, rephrased variants (verbosity, style) and semantic tags. Human evaluation shows that SonicCaps is rated significantly higher than existing captioning datasets, with fine-grained analyses indicating that our captions are perceived as more descriptive and precise, which strongly correlates with quality judgments. Finally, training CLAP models on SonicCaps with a multi-caption sampling strategy consistently improves audio retrieval and zero-shot classification, with stronger generalization across public and commercial benchmarks. We release both SonicCaps and two specialized CLAP models on hugging face: https://huggingface.co/datasets/Zineb/SonicCaps.

2026-09-02T09:18:27Z Zineb Lahrichi Marc Ferras Gaël Richard Geoffroy Peeters http://arxiv.org/abs/2609.02268v1 2026-09-02T08:15:52Z

With the rapid proliferation of generative models on Machine Learning as a Service (MLaaS) platforms, reliably tracing the provenance of synthetic media without modifying generator architectures or parameters remains a major challenge. In this work, we propose a self-referential retrosynthesis framework for explainable AI provenance forensics under a fixed-generator setting. The framework leverages a jointly optimized encoder-decoder pair to implement a self-embedding mechanism that enables round-trip consistency verification. During inference, client inputs are first encoded and then processed by the generator to produce outputs with high visual fidelity. For forensic verification, the consistency between the resynthesized image and the query image is analyzed to determine whether the image originates from the target generative model. Our approach eliminates the need for watermark embedding or modifications to the generation process. Experimental results show that images generated from encoded inputs maintain visual quality comparable to original generator outputs, while decoded images reliably trace back to their corresponding source inputs. Furthermore, the framework provides interpretable evidence for generative content provenance, establishing a practical tool for explainable generative AI forensics.

2026-09-02T08:15:52Z 12 pages, 10 figures. This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible Yijie Lin Ching-Chun Chang Isao Echizen Hui Li Chin-Chen Chang http://arxiv.org/abs/2609.02082v2 2026-09-03T07:46:30Z

Visual modality enhances the capabilities of multimodal large language models (MLLMs) but also introduces a safety concern: a benign textual query may convey harmful intent when grounded in a visual image. We term this cross-modal safety drift and our pilot studies show that the safety response rate for such requests is substantially lower than that for requests containing explicitly unsafe text. This paper aims to systematically study this issue. First, we conduct an empirical analysis to identify representative unsafe response patterns. Building on these, we interpret model representations and attentions, revealing that visually risky cues receive limited attention and weakly trigger refusal. Motivated by the observation that safety signals from unsafe text processing can be transferred, we propose safety-awareness representation transfer (SRT), a lightweight direction-refinement method that mitigates cross-modal safety drift with a frozen MLLM backbone. Experiments across multiple benchmarks and models show that SRT effectively improves safety in diverse cross-modal settings while preserving utility. Code is available at https://github.com/cucu220123/safety-awareness.

2026-09-02T04:12:35Z Accepted to Findings of EMNLP 2026 Tianqi Xiao Shiyao Cui Minghao Zhang Junxiao Yang Renmiao Chen http://arxiv.org/abs/2609.01535v1 2026-09-01T17:04:47Z

This paper explores whether large language models (LLMs) can design video coding tools, a highly challenging task due to the intricate algorithmic coupling of tool modifications. In particular, we present an empirical case study on the Planar mode, a long-standing intra prediction tool in video coding standards. Our experiments operate within a generation-and-evaluation loop, with the LLM generating new Planar predictors, encoder trials evaluating their coding performance, and the LLM re-generating refined implementations based on the evaluation feedback. We first examine directly replacing the default Planar mode in the Fraunhofer Versatile Video Encoder (VVenC) under its faster preset. Experimental results demonstrate that the LLM-generated mode can outperform the conventional Planar mode on this lightweight toolset, achieving 0.18% bitrate savings with 0.4% complexity overhead on the standard benchmark. We further extend our evaluation to the Enhanced Compression Model (ECM). Leveraging newly introduced directional Planar modes, we investigate two integration strategies: directly replacing them, and introducing the LLM-generated predictor as an additional prediction mode with new syntax elements. The empirical results suggest that both strategies can yield coding gains under a constrained low-resolution setting. Overall, this study offers preliminary evidence and practical insights, highlighting both the potential and open challenges of LLM-based coding tool design.

2026-09-01T17:04:47Z Yingwen Zhang Meng Wang Liqiang He Shiqi Wang http://arxiv.org/abs/2609.01287v1 2026-09-01T14:23:27Z

Multi-talker automatic speech recognition (MT-ASR) remains challenging under overlapping speech. Hard diarization-based segmentation introduces irreversible errors, whereas serialized output training (SOT) avoids explicit segmentation but does not condition a pretrained encoder on speaker activity. We propose Soft Posterior Speaker Injection (SPSI): a lightweight head predicts frame-level speaker posteriors $\hat{\mathbf{P}}$ and injects them into Whisper through multi-layer feature-wise linear modulation (FiLM) and decoder speaker-memory prompts. On controlled two-speaker LibriSpeech overlap, SPSI reduces utterance-mean constrained permutation word error rate (cpWER) from 50.7\% (SOT) to 49.6\% (one-sided paired bootstrap $p{\approx}0.006$), with a larger reduction in the high-overlap bin (60.4\%$\to$58.8\%). Same-backbone speaker-auxiliary objectives and voice activity detection (VAD) pipelines do not outperform SOT; zero-shot (ZS) LibriCSS is comparable. Freeze-posterior adaptation with overlap-heavy (OV-heavy) continuation reduces held-out LibriCSS cpWER (sessions 8--9) to 32.4\% (versus 37.5\% for SOT). Ablations indicate complementary encoder FiLM and decoder prompts, and that the effective signal is a \emph{soft} simplex-valued speaker share.

2026-09-01T14:23:27Z This paper is submitted to ICASSP2027 Jian Zhu Cheng Luo http://arxiv.org/abs/2609.01277v1 2026-09-01T14:12:57Z

Although pretrained joint audio-visual diffusion models offer rich control over \emph{what} to generate, they provide no explicit control over \emph{when} an utterance should occur. To address this, we study \emph{inference-time speech scheduling}, a novel task that places coupled speech and visual articulation within user-specified begin--end intervals without finetuning the backbone model. We uncover two intrinsic properties of the denoising process that enable this task. First, a timing-sensitive text-to-audio cross-attention head exposes each utterance's model-implied source span along the latent timeline. Second, the predicted clean latent already organizes coupled speech and visual articulation, allowing their temporal placement to be edited without regenerating the content. Building on these discoveries, we propose \textbf{TimeSteer}, a training-free framework that localizes each utterance's source span through \textbf{Source Span Localization} and transfers the associated audio-visual latent content from the source interval to the specified target interval through \textbf{Region-Aware Latent Remapping}. We further introduce \textbf{SpeechShift}, the first benchmark for interval-level speech scheduling in joint audio-visual generation. Experiments across two representative backbones show that TimeSteer substantially improves interval controllability over training-free baselines while maintaining competitive overall generation quality.

2026-09-01T14:12:57Z Chao Zhou Yiling Chen Qi Chu Tao Gong Nenghai Yu Tianyi We http://arxiv.org/abs/2609.01168v2 2026-09-02T12:27:54Z

Text-to-image (T2I) models remain vulnerable to jailbreak attacks that elicit Not-Safe-For-Work (NSFW) content, despite increasingly being guarded by heterogeneous, multi-layer safety stacks combining text filters, image classifiers, and cross-modal detectors. Existing jailbreak studies either optimize against individual filters or query the complete pipeline with aggregate feedback, making it difficult to identify the active constraint and adapt to conflicts across safety layers. In this paper, we introduce the Detection Surface, a unified geometric framework that characterizes the decision boundaries induced by heterogeneous T2I safety filters and their joint effect on the jailbreak search space. This formulation reveals that successful evasion is governed by a sparse and non-convex region shaped by cross-layer conflicts, where mutations that bypass one filter may increase exposure to another. Motivated by this analysis, we propose CRACK, a multi-agent debate framework for adaptive jailbreak search that decomposes jailbreak search into exploration, diagnosis, and arbitration. CRACK coordinates an Attack Agent, a Defense Agent, and a Judge Agent to iteratively generate prompt mutations, obtain layer-specific diagnostic feedback, and optimize mutation strategies through reward-guided refinement. Through repeated rounds of debate, CRACK adapts its search direction to the evolving cross-layer constraints while preserving the original harmful intent. Extensive experiments across multiple T2I models, datasets, and safety configurations show that CRACK achieves Attack Success Rates (ASR) of up to 99.63% under composite defenses, while requiring fewer queries than existing methods and maintaining semantic fidelity.

2026-09-01T12:45:27Z 16 pages, 11 figures Kaiyan Wen Shijie Zhang Lu Yu Guangdong Bai http://arxiv.org/abs/2609.00913v1 2026-09-01T08:41:58Z

Interactions with cold items negatively impact real-time personalization of ID-based recommender systems. This is because the use of such interactions degrades user preference estimates, whereas excluding cold items from the user profile prevents real-time recommendation updates. In industrial scenarios, one heuristic often applied to address this shortcoming at inference time is to replace, i.e., "swap", cold-start items by their most similar "warm" neighbor, where similarity is inferred from the items' side information. In this paper, we demonstrate that sequential models, most often used for real-time personalization, are not robust to such swaps, and propose SwapRec, an approach to address this issue. SwapRec relies on using the same swap heuristics already at training time. We apply SwapRec to state-of-the-art models for sequential recommendation and analyze its impact by means of quantitative experiments in three recommendation domains (online shopping, movie, music). The experimental results show that, irrespective of the underlying sequential architecture, our easy-to-implement SwapRec approach allows for substantially more accurate recommendations when in presence of interactions with cold items, simultaneously leading to a larger percentage of cold items in the recommendation lists.

2026-09-01T08:41:58Z Accepted at DaQuaMRec @ RecSys 2026: Second International Workshop on Data Quality-Aware Multimodal Recommendation Marta Moscati Jan Malte Lichtenberg Davide Abbattista Antonio De Candia Laura Boggia Matteo Ruffini http://arxiv.org/abs/2609.00742v1 2026-09-01T05:23:48Z

As AI video generators achieve cinematic realism, reliable detection becomes essential for safeguarding digital trust. We identify cross-scale coupling mismatch as a new forensic signal, where scale refers to the level of abstraction (semantic dynamics vs. pixel-level residuals): in natural videos, macro-level temporal dynamics and micro-level residual patterns are intrinsically coupled by the unified imaging physics pipeline, whereas AI generators, whose training objectives do not explicitly preserve this joint distribution, systematically violate this coupling. Detecting such mismatch is challenging because it requires independently extracting information at both scales while simultaneously quantifying their cross-scale relationship. We propose RIFT (Representation Inconsistency Forensics on Trajectories), an orthogonal forensic framework that addresses this through three interlocking components: a macro stream that builds a dynamic baseline of expected temporal evolution via differential geometry and persistent homology on learned manifold trajectories, a micro stream that acts as a sensitive forensic probe via steganalytic filtering and temporal modeling, and a coupling divergence module that measures the conditional dependency between the two streams. Gram-Schmidt orthogonality guarantees the information-theoretic validity of this measurement. Experiments on two benchmarks (VidProM, 120K videos, 7 generators; GenVidBench, 68K videos, 4 generators) demonstrate that RIFT achieves 99.33% and 99.72% F1-score respectively, with 97.87% unseen-generator detection rate in leave-one-out evaluation, while exhibiting encoder agnosticism: scaling from ViT-S/14 (22M) to ViT-L/14 (300M) changes F1 by less than 0.1%, and switching to a different encoder family (DINOv1) reduces F1 by only 0.73 pp. Code is available at https://github.com/Litsay/RIFT

2026-09-01T05:23:48Z 12 pages, 4 figures, 11 tables. Accepted at ACM Multimedia 2026 (MM '26), Rio de Janeiro, Brazil. This version includes the supplementary material as Appendix A-E Siyu Li Jin Yang Weiheng Liang http://arxiv.org/abs/2609.00551v1 2026-09-01T01:38:41Z

Multimodal memory offers a scalable interface for long-video question answering, but existing methods often retrieve captions, frames, transcripts, summaries, or graph facts as isolated fragments. Although searchable, such fragments are not generation-ready: language models must reconstruct cross-modal and temporal alignments at inference time, when context is limited and attribution is difficult. We propose EM^2Mem, an event-centric multimodal memory framework that binds heterogeneous evidence to event anchors during memory construction. Each event-indexed memory cell aligns multimodal records, temporal context, graph-linked relations, semantic facts, and provenance, enabling compact evidence readout over grounded multimodal events rather than modality-specific fragments. Across three long-video QA benchmarks, EM^2Mem improves average accuracy over the strongest memory baseline by 2.0, 2.4, and 3.7 points, improves strict event-level Top-5 evidence recall by 7.0 points, and reduces per-query latency by 4.67 times and total inference tokens by 63.66% (The code will be integrated into https://github.com/zjunlp/LightMem).

2026-09-01T01:38:41Z Accepted by EMNLP 2026 findings Yijun Chen Yaqi Zheng Yanya Li Boyi Xiao Buqiang Xu Shuofei Qiao Jizhan Fang Xinle Deng Yunzhi Yao Xuehai Wang Liuxin Zhang Hui Li Huajun Chen Shumin Deng