ABSTRACT Symbolic music generation faces a fundamental trade-off between efficiency and quality. Fine-grained tokenizations achieve strong coherence but incur long sequences and high complexity, while compact tokenizations improve efficiency at the expense of intra-token dependencies. To address this, we adapt a delay-based scheduling mechanism (DP) that expands compound-like tokens across decoding steps, enabling autoregressive modeling of intra-token dependencies while preserving efficiency. Notably, DP is a lightweight strategy that introduces no additional parameters and can be seamlessly integrated into existing representations. Experiments on symbolic orchestral MIDI datasets show that our method improves all metrics over standard compound tok-enizations and narrows the gap to fine-grained tokenizations.

Guidance methods have demonstrated significant improvements in cross-modal audio generation, including text-to-audio (T2A) and video-to-audio (V2A) generation. The popularly adopted method, classifier-free guidance (CFG), steers generation by emphasizing condition alignment, enhancing fidelity but often at the cost of diversity. Recently, autoguidance (AG) has been explored for audio generation, encouraging the sampling to faithfully reconstruct the target distribution and showing increased diversity. Despite these advances, they usually rely on a single guiding principle, e.g., condition alignment in CFG or score accuracy in AG, leaving the full potential of guidance for audio generation untapped. In this work, we explore enriching the composition of the guidance method and present a mixture-of-guidance framework, AudioMoG. Within the design space, AudioMoG can exploit the complementary advantages of distinctive guiding principles by fulfilling their cumulative benefits. With a reduced form, AudioMoG can consider parallel complements or recover a single guiding principle, without sacrificing generality. We experimentally show that, given the same inference speed, AudioMoG approach consistently outperforms single guidance in T2A generation across sampling steps, concurrently showing advantages in V2A, text-to-music, and image generation. These results highlight a "free lunch" in current cross-modal audio generation systems: higher quality can be achieved through mixed guiding principles at the sampling stage without sacrificing inference efficiency. Demo samples are available at: https://audio-mog.github.io.

Recent works on object removal and insertion have enhanced their performance by handling object effects such as shadows and reflections, using diffusion models trained on counterfactual datasets. However, the performance impact of applying classifier-free guidance to handle object effects across removal and insertion tasks within a unified model remains largely unexplored. T o address this gap and improve efficiency in composite editing, we propose CrimEdit, which jointly trains the task embeddings for removal and insertion within a single model and leverages them in a classifier-free guidance scheme--enhancing the removal of both objects and their effects, and enabling controllable synthesis of object effects during insertion. CrimEdit also extends these two task prompts to be applied to spatially distinct regions, enabling object movement (repositioning) within a single denoising step. By employing both guidance techniques, extensive experiments show that CrimEdit achieves superior object removal, controllable effect insertion, and efficient object movement--without requiring additional training or separate removal and insertion stages.

This paper explores a novel lightweight approach LightFair to achieve fair text-to-image diffusion models (T2I DMs) by addressing the adverse effects of the text encoder. Most existing methods either couple different parts of the diffusion model for full-parameter training or rely on auxiliary networks for correction. They incur heavy training or sampling burden and unsatisfactory performance. Since T2I DMs consist of multiple components, with the text encoder being the most fine-tunable and front-end module, this paper focuses on mitigating bias by fine-tuning text embeddings. To validate feasibility, we observe that the text encoder's neutral embedding output shows substantial skewness across image embeddings of various attributes in the CLIP space. More importantly, the noise prediction network further amplifies this imbalance. To finetune the text embedding, we propose a collaborative distance-constrained debiasing strategy that balances embedding distances to improve fairness without auxiliary references. However, mitigating bias can compromise the original generation quality. To address this, we introduce a two-stage text-guided sampling strategy to limit when the debiased text encoder intervenes. Extensive experiments demonstrate that LightFair is effective and efficient. Notably, on Stable Diffusion v1.5, our method achieves SOTA debiasing at just $1/4$ of the training burden, with virtually no increase in sampling burden. The code is available at https://github.com/boyuh/LightFair.

In-context learning (ICL) enables Large Language Models (LLMs) to generate predictions based on prompts without additional fine-tuning. While prompt engineering has been widely studied, the impact of role design within prompts remains underexplored. This study examines the influence of role configurations in zero-shot and few-shot learning scenarios using GPT-3.5 and GPT-4o from OpenAI and Llama2-7b and Llama2-13b from Meta. We evaluate the models' performance across datasets, focusing on tasks like sentiment analysis, text classification, question answering, and math reasoning. Our findings suggest the potential of role-based prompt structuring to enhance LLM performance.

The creation of high-quality multimodal datasets remains fundamental for advancing role-playing capabilities in large language models (LLMs). While existing works predominantly focus on text-based persona simulation, Audio Role-Playing (ARP) presents unique challenges due to the need for synchronized alignment of semantic content and vocal characteristics. To address this gap, we propose AudioRole, a meticulously curated dataset from 13 TV series spanning 1K+ hours with 1M+ character-grounded dialogues, providing synchronized audio-text pairs annotated with speaker identities and contextual metadata. In addition, to demonstrate the effectiveness of the dataset, we introduced ARP-Eval, a dual-aspect evaluation framework that assesses both response quality and role fidelity. Empirical validation showing GLM-4-Voice trained on AudioRole (which we called ARP-Model) achieve an average Acoustic Personalization score of 0.31, significantly outperforming the original GLM-4-voice and the more powerful model MiniCPM-O-2.6, which specifically supports role-playing in one-shot scenarios. The ARP-Model also achieves a Content Personalization score of 0.36, surpassing the untrained original model by about 38% and maintaining the same level as MiniCPM-O-2.6. AudioRole features dialogues from over 115 main characters, 6 trained ARP-Models that role-play different characters, and evaluation protocols. Together, they provide an essential resource for advancing audio-grounded role-playing research.

Real-world videos often show routine activities punctuated by memorable, surprising events. However, most Video-LLMs process videos by sampling frames uniformly, likely missing critical moments that define a video's narrative. We introduce SPIKE, an inference-time framework that quantifies Bayesian Surprise as the belief update triggered by new visual evidence in the video stream, identifying moments where new visual evidence conflicts with prior beliefs. SPIKE effectively localizes surprise in videos, strongly correlated with humans on positive (FunQA) and negative (Oops!) surprise benchmarks. Since the beliefs of zero-shot Video-LLMs are often suboptimal, we develop SPIKE-RL, which leverages GRPO to optimize belief hypotheses based on a reward signal from the video caption. SPIKE and SPIKE-RL guide query-agnostic surprise-weighted frame sampling, which allocates more frames to interesting moments in the video. With this strategy, we achieve consistent performance gains on five downstream benchmarks over uniform sampling. By enabling Video-LLMs to track beliefs and register surprise, our work paves the way for more robust models that can revise their understanding in response to new information.

Language models (LMs) encode substantial factual knowledge, but often produce answers judged as incorrect. We hypothesize that many of these answers are actually correct, but are expressed in alternative surface forms that are dismissed due to an overly strict evaluation, leading to an underestimation of models' parametric knowledge. We propose Retrieval-Constrained Decoding (RCD), a decoding strategy that restricts model outputs to unique surface forms. We introduce YAGO-QA, a dataset of 19,137 general knowledge questions. Evaluating open-source LMs from 135M to 70B parameters, we show that standard decoding undervalues their knowledge. For instance, Llama-3.1-70B scores only 32.3% F1 with vanilla decoding but 46.0% with RCD. Similarly, Llama-3.1-8B reaches 33.0% with RCD, outperforming the larger model under vanilla decoding. We publicly share the code and dataset at https://github.com/Rajjaa/disambiguated-LLM.

Text-to-music models have revolutionized the creative landscape, offering new possibilities for music creation. Yet their integration into musicians workflows remains underexplored. This paper presents a case study on how TTM models impact music production, based on a user study of their effect on producers creative workflows. Participants produce tracks using a custom tool combining TTM and source separation models. Semi-structured interviews and thematic analysis reveal key challenges, opportunities, and ethical considerations. The findings offer insights into the transformative potential of TTMs in music production, as well as challenges in their real-world integration.

As large language models continue to develop, the feasibility and significance of text-based symbolic music tasks have become increasingly prominent. While symbolic music has been widely used in generation tasks, LLM capabilities in understanding and reasoning about symbolic music remain largely underexplored. To address this gap, we propose ABC-Eval, the first open-source benchmark dedicated to the understanding and instruction-following capabilities in text-based ABC notation scores. It comprises 1,086 test samples spanning 10 sub-tasks, covering scenarios from basic musical syntax comprehension to complex sequence-level reasoning. Such a diverse scope poses substantial challenges to models' ability to handle symbolic music tasks. We evaluated seven state-of-the-art LLMs on ABC-Eval, and the results reveal notable limitations in existing models' symbolic music processing capabilities. Furthermore, the consistent performance of individual baselines across different sub-tasks supports the reliability of our benchmark.