Papers for
multimedia content creators
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
Reinforcement learning improves multi image editing consistency
DeCoRL: Decomposed Consistency Reinforcement Learning for Multi-Image Composition
Abstract: Multi-image composition (MICo) is a critical yet challenging capability for large-scale image editing models, where consistency degrades rapidly as the number of reference images increases. While supervised fine-tuning (SFT) can equip models with basic MICo ability, it provides only implicit supervision on the composed image and often fails to learn reliable object-level correspondence across multiple references. We propose DeCoRL, a reinforcement-learning (RL) framework that improves MICo consistency by decomposing holistic MICo consistency into object-level consistency. DeCoRL first establishes instruction-relevant object correspondences across reference images and localizes the corresponding regions in the generated image. It then uses a VLM-based reward model trained on single-object consistency annotations to produce fine-grained, object-wise feedback, which is aggregated into a multi-object consistency reward to guide RL post-training. We further introduce DeCo-Bench, a benchmark covering compositions from 2 to 6 reference images. Experiments on DeCo-Bench and the public MICo-Bench demonstrate consistent improvements in MICo consistency.
Graph method improves emotion recognition in artworks over vision language models
Beyond Vision-Language Models: Visual-Semantic Affect Graphs for Emotion Recognition in Artworks
Abstract: Emotion recognition in artworks remains challenging because affective meaning is often subjective, implicit, and shaped by relationships between visual form and semantic associations. While vision-language models provide strong multimodal representations, direct image-to-emotion prediction may overlook explicit relations among formal, semantic, and affective cues that support emotional interpretation. This paper proposes Visual-Semantic Affect Graphs (VSAG), a graph-based framework that represents artworks through interpretable feature nodes, including color, brushstroke, composition, light-shadow, line quality, and extracted visual entities. The graph learns feature-emotion associations for dominant emotion prediction, with valence-arousal structure modeled through an optional hierarchical auxiliary affect module. We evaluate VSAG and VSAG-TopK against direct VLM inference using Qwen3-VL and InternVL3.5 on EmoArt-5k. Results show that direct VLMs provide useful but modest baselines, while graph-based models improve dominant emotion prediction. In particular, the hierarchical valence-arousal variant of VSAG achieves the best overall performance, improving accuracy, macro-F1, and weighted-F1 over both direct VLM inference and the emotion-only VSAG baseline. Per-class results indicate that graph-based models and VLMs capture partly complementary signals, while valence-arousal structure is most useful when integrated as a coarse-to-fine affect hierarchy rather than as a simple post-hoc correction. These findings suggest that explicit visual-semantic structure can support both prediction and interpretation in artwork emotion recognition.
Direct forcing improves few-step streaming text to video generation
Direct Forcing: A Single-Stage Training Recipe for Few-Step Autoregressive Text-to-Video Generation
Abstract: Few-step autoregressive video generation enables low-latency streaming, yet prevailing training pipelines separate causal adaptation from distribution matching. Recent representation-based text-to-video methods retain this separation by starting from ODE-initialized checkpoints, even when the target distribution is defined directly by reference videos. We introduce Direct Forcing, a single-stage framework that uses reference videos both as paired examples of how a history continues and as a target population for self-generated rollouts. Causal Continuation Alignment learns paired spatiotemporal correspondence under reference histories, while Rollout Distribution Matching uses maximum mean discrepancy to align the representation distributions of self-generated and reference videos. Together, these complementary signals allow a pretrained video model to learn causal continuation and the target video distribution in a single training stage. Dynamic Gradient Balancing adapts the relative gradient contributions of continuation alignment and distribution matching during training. Starting from bidirectional Wan2.1-1.3B, Direct Forcing learns a four-step autoregressive generator in 250 updates without a separate initialization stage or online diffusion-score supervision, achieving 84.63 VBench Total and 86.01 Quality. Ablation studies further demonstrate the benefits of dynamic gradient balancing and reveal distinct representation preferences: deeper features benefit continuation alignment, whereas intermediate features yield better performance for rollout distribution matching.
Creative prompt expansion improves video and audio quality in generation
VideoGen Model 1: Prompt Expansion Versus Formatting in Audiovisual Generation
Abstract: Creative prompt expansion improved overall audiovisual quality scores over raw prompts in a controlled pilot of VideoGen Model 1, our post-trained version of MiniMax H3. We compared the raw, reformatted, and expanded versions of 30 prompts, holding the generator, seed, and sampling settings fixed. Each clip was rated separately on five dimensions. Their mean formed an overall score on a 1-10 scale. Compared with formatting alone, overall scores increased by 0.63 points (95% interval: [+0.11, +1.15]) under VA-Judger and 0.15 points (95% interval: [-0.03, +0.35]) under Gemini 3.8 Flash, whose overall change remained uncertain. Prompt-alignment scores increased by 0.97 points under VA-Judger and 0.50 under Gemini. VA-Judger also showed higher audio quality; visual-quality ratings changed little under either judge.
Code2Video Compass benchmark evaluates code based video creation
Code2Video-Compass: A Comprehensive Benchmark for General-Purpose Code-Based Video Generation across T2V, I2V, and V2V
Abstract: Code-based video generation offers fine-grained control, targeted editing, and component reuse compared with direct video generation. This approach requires multimodal large language models (MLLMs) to translate visual and temporal requirements into executable programs that produce complete videos. However, existing code-based benchmarks primarily target specialized tasks, leaving complete video creation across diverse creative categories and input modalities insufficiently evaluated. We introduce Code2Video-Compass, a general-purpose benchmark for code-based video generation across text-to-video (T2V), image-to-video (I2V), and video-to-video (V2V) settings. Code2Video-Compass comprises 2,104 tasks spanning 10 categories, 24 subtypes, and three complexity dimensions covering construction, dynamic animation, and state dependency. It combines tasks curated from textual instructions and visual references with corresponding T2V, I2V, and V2V tasks derived from complete professional and user-created videos. We propose a dual-level evaluation framework combining 12 objective metrics and five subjective metrics. Objective evaluation covers operational performance, video quality, text-video alignment, and reference fidelity, while subjective evaluation uses an MLLM judge to evaluate instruction following and holistic video quality. Extensive evaluation of 15 recent proprietary and open-weight MLLMs from nine families reveals that even leading models face challenges in reliably generating high-quality complete videos aligned with textual and visual requirements, with performance generally declining as construction, dynamic animation, and state-dependency complexity increases. We anticipate that Code2Video-Compass will advance reliable, coherent, and controllable code-based video generation and inspire broader progress in programmatic visual creation.
Multimodal model improves emotion recognition for oddly satisfying videos
PI-MLLM: Psychophysics-Inspired Multimodal Large Language Model for Oddly Satisfying Video Emotion Analysis
Abstract: Oddly Satisfying Videos (OSVs) can evoke strong viewer responses through repetitive motion, visual tactility, and ASMR-like sounds despite containing few explicit semantic emotion cues. This makes OSVs a challenging setting for Video Emotion Analysis (VEA), since most existing methods focus on semantic video content and under-model the physical sensory dynamics that shape viewer responses. To support Oddly Satisfying Video Emotion Analysis (OSVEA), we construct the Oddly Satisfying Video Dataset (OSVD), comprising 12,000 annotated 5-second clips. Each clip provides viewer-emotion labels under video-only, audio-only, and audio-visual conditions, together with explanations grounded in visual and acoustic sensory evidence. Building on OSVD, we propose PI-MLLM, a psychophysics-inspired multimodal large language model that integrates Modality-Specific Dynamics Alignment (MSDA) and Coalition-Based Audio-Visual Attribution (CAVA). MSDA uses auxiliary prediction objectives to align modality representations with visual motion and acoustic energy-onset dynamics. CAVA evaluates emotion scores across modality coalitions and applies Shapley attribution to identify the dominant modality and determine whether visual and auditory evidence is synergistic or conflicting. Experiments show that PI-MLLM improves both viewer-side emotion prediction and sensory-grounded explanation quality, suggesting the value of physical sensory cues to video emotion analysis.
Audio video consistency measured at event level improves generation quality
AVC-Reward: Event-Centric Audio-Video Consistency Evaluation
Abstract: As joint audio-video generation models become more capable, they can generate audio-visual scenes exhibiting three forms of complexity: multi-event composition, multi-layer audio mixture, and asymmetric cross-modal relations. However, current evaluations largely reuse metrics from earlier simple and unidirectional tasks. They typically evaluate an audio-video clip as a whole and rely on simple predefined relations, making them unsuitable for these increasingly complex scenes. We propose an event-centric evaluation framework that disentangles audio and visual events at the event level, analyzes each event independently, cross-validates relations across modalities, and aggregates the judgments into a coarse-to-fine, three-level consistency metric. The metric assesses global narrative harmony, event-level temporal synchronization, and fine-grained attribute alignment, providing a general formulation for audio-video consistency evaluation across diverse generation settings and cross-modal relations. We instantiate the metric as AVC-Reward, a trainable audio-video reward model. To evaluate the agreement between automatic judgments and human perception across generation paradigms and relation complexities, we construct AVC-Benchmark, which is primarily built from diverse positive and negative samples generated by various types of audio-video generation models. AVC-Reward achieves state-of-the-art alignment with human judgment among automatic methods, provides interpretable rationales that localize failures, and improves the audio-video consistency of a joint generation model when used as an online reinforcement-learning signal. Project page: https://tsuki0512.github.io/AVC-Reward.github.io/
Unified multimodal model generates consistent images and text over long sequences
Scaling Unified Multimodal Models to Long-Horizon Interleaved Generation
Abstract: Unified multimodal models (UMMs) provide an interface for interleaved text-image generation, where generated visual states can directly inform subsequent reasoning and synthesis. However, existing systems are still primarily demonstrated on short trajectories or simplified visual content. We develop a large UMM natively trained on interleaved multimodal data, with strong capabilities in image synthesis, text rendering, complex composition, and multimodal planning, and study how to extend these native capabilities to substantially longer autoregressive trajectories. The main challenge is that generated images continuously accumulate in context: irrelevant or imperfect visual states interfere with future synthesis, while previously established characters and scenes must still be recovered after long intervals. We introduce a visual-memory framework that maintains persistent visual references, selectively retrieves scene-relevant historical states while preserving the complete semantic history, and uses few-step self-rollout to adapt the model to imperfect generated histories. We further introduce LongMIG-Bench for evaluating long-horizon text-to-multi-image and image-to-multi-image generation. Our model achieves high-fidelity long-horizon interleaved generation with strong character and scene consistency, demonstrating how visual memory can scale native interleaved generation to substantially longer trajectories.
VoTSpeech separates voice design from speech generation in TTS systems
VoTSpeech: Decoupling Voice Design from Speech Generation in Instruction-Following TTS
Abstract: Instruction-following text-to-speech (TTS) typically leaves voice design implicit, as conventional speaker embeddings encode reference audio rather than text-driven targets. We propose VoTSpeech, which decouples voice design from speech generation via an explicit intermediate voice latent. Built on a shared language model backbone, VoTSpeech first predicts a continuous voice latent from text instructions using a flow-based module. This latent serves as a shared condition for both the language model and the Acoustic DiT, guiding autoregressive sequence modeling and acoustic latent generation, respectively, before an audio VAE decoder converts the acoustic latents into waveforms. To capture speaker identity and complementary acoustic nuances, the voice latent is jointly supervised by speaker embeddings and temporally pooled audio VAE features. With task-specific adaptation, VoTSpeech performs strongly on InstructTTSEval-ZH, demonstrating competitive capabilities against strong open-source models while maintaining high naturalness. These results confirm that explicit voice latent design and its conditioning interface effectively empower text-driven voice design for controllable speech synthesis. Code, model, and audio samples are available at https://github.com/ywbn/VoTSpeech .
Large vision language models stop seeing things that are not there
From Suppression to Repair: Mitigating Object Hallucination in Large Vision-Language Models via Localized Distribution Alignment
Abstract: Object hallucination remains a major obstacle for large vision-language models (LVLMs) to generate reliable content. An intuitive mitigation strategy is to suppress hallucination-related components in hidden representations. However, these components may also contain useful information, and suppressing them can weaken the model's multimodal capabilities. In this paper, we propose ResOT, a training-free method that repairs representations at inference time through localized distribution alignment. Specifically, ResOT projects dominant hallucinated directions away from the faithful subspace, forming a low-dimensional residual subspace for intervention. Within this subspace, ResOT uses Gaussian optimal transport (OT) to align the hallucinated distribution with the faithful one. The resulting map defines repair targets with minimal changes to the original representations. At inference, ResOT adaptively controls how far each token state moves toward its OT target. Experiments on three representative LVLMs show that ResOT substantially reduces object hallucination while improving image caption quality and multimodal performance across multiple benchmarks. Code will be released.
Objective metrics better predict human judgments for co speech gestures
Perceptually Grounded and Semantics-Aware Evaluation for Holistic Co-Speech Gesture Generation
Abstract: Holistic and semantics-aware co-speech gesture generation has advanced rapidly, yet evaluation remains behind: objective metrics do not consistently reflect human perception, and semantic appropriateness remains difficult to quantify. We present a perceptually grounded and semantics-aware benchmark that combines standardized model comparison, human-centered metric validation, and fine-grained semantic evaluation. We first curate a list of 13 objective metrics covering different aspects, including distributional similarity, geometric fidelity, kinematic quality, cross-modal synchrony, and semantic appropriateness. For the semantic-appropriateness category, we propose a new metric, Semantic Gesture Preservation (SGP), which measures how far semantic gestures in the ground truth are preserved in the generated gestures. For this, we augment the BEAT2 dataset's annotations using a multi-modal LLM. We then conduct a perceptual study where 101 participants score generated gestures among five dimensions, including human-likeness, motion diversity, absence of animation errors, speech timing and content match. We systematically analyze objective metric--subjective score correlations. Unlike Semantic Score (SC), which shows no significant association with the evaluated perceptual dimensions, SGP is selectively aligned with speech-aware human judgments. We construct five target-specific composite metrics aligned with the subjective dimensions. These composites improve perceptual alignment across all five dimensions, with the largest gains for absence of animation errors and content match, indicating that complementary objective signals can better approximate human judgments than individual metrics alone. Overall, our results show that objective metrics require validation against subjective evaluations.
Steerspeech lets text to speech voices express emotion with precision
Steerspeech: Activation Steering For Emotion Control In Generated Speech
Abstract: Pretrained text-to-speech (TTS) models can generate expressive speech, but reliable inference-time emotion control remains challenging: prompts and reference audio offer coarse, inconsistent control, whereas specialized conditioning and model adaptation require costly training. We present SteerSpeech, a lightweight activation-steering framework that controls emotion by injecting steering vectors into hidden activations. For each target emotion we train a lightweight low-rank transform, using a multi-expert objective that encourages monotonic emotion control while preserving speaker identity and linguistic content, constraining steering drift, and keeping the TTS backbone frozen. To optimize through discrete speech tokens, we introduce a two-pass generation-and-replay pipeline using a straight-through estimator to backpropagate expert supervision through sampled tokens. At inference, a target-emotion steering direction is optimized with its respective transform and injected into the base TTS model. Objective and subjective evaluations with Qwen3-TTS across seen, unseen, and accented speakers show stronger continuous emotion control with limited speaker and content degradation. SteerSpeech achieves 1.08x-7.12x baseline target-emotion scores and for a representative emotion subjectively, it receives 78.1%-96.8% intensity preference and 1.43x-1.46x speaker-identity preservation at high steering strengths.
Salt++ improves few-step streaming video and audio generation quality
Salt++: Context-Aligned Post-Training for Few-Step Streaming Multimodal Generation
Abstract: Few-step streaming audio--video generation requires both causal modeling and step distillation, yet standard training recipes face two context-related challenges. Teacher forcing pairs clean history with a noisy target, but supervises predictive contextual representations only indirectly through velocity prediction. Meanwhile, directly reusing bidirectional score models in causal Distribution Matching Distillation (DMD) creates a mismatch between generation and scoring contexts. We address these challenges with Salt++, a two-stage post-training framework comprising Causal Self-Flow (CSF) and context-aligned autoregressive DMD. CSF exploits contextual information asymmetry by varying the history while keeping the noisy target fixed: a noise-mixed-history student aligns its intermediate representations with those of a clean-history exponential-moving-average teacher. This self-supervised signal encourages the student to extract semantic information and improves cross-modal alignment. Context-aligned AR DMD shares the causal mask and prefix across generator sampling, fake-score training, and real-score evaluation to match generated and reference distributions under a block-conditional KL objective. With calibrated teacher guidance, it performs clean-prefix few-step distillation and then adapts to generated histories without switching objectives or requiring separate consistency distillation. At 480p, Salt++ improves visual and motion quality by 57% and 45% over OmniForcing on JavisBench under the same 4-step causal setting. A separate scale-wise post-training stage extends Salt++ to 4-step $1664\times960$ generation, outperforming bidirectional LTX-2 on six of seven reported metrics. Project page: https://xingtongge.github.io/Saltpp
Joint audio video generation improves fidelity and sync with new RL method
AV-GRPO: Modality-Anchored Decoupling Diffusion Reinforcement Learning for Joint Audio-Video Generation
Abstract: Recent years have witnessed major progress in joint audio-video generation. Existing models still suffer from limited per-modality fidelity, insufficient text-modality alignment and weak cross-modal synchronization. While reinforcement-learning post-training offers a promising remedy, directly adapting it to joint audio-video generation is challenging. Heterogeneous multimodal rewards entangle learning signals and complicate credit assignment. Joint optimization of two modality towers is computationally expensive given their divergent dynamics. Moreover, synchronization evaluation difficulty depends on paired samples, preventing fair reward comparisons. We propose AV-GRPO, a modality-anchored online diffusion RL framework, and 5DAV, a decoupled, difficulty-controllable training dataset. AV-GRPO includes three key modules: (1) modality-anchored rollouts to disentangle learning signals and stabilize difficulty; (2) trajectory-locked frozen-tower optimization to reduce cost and reassign credit; (3) adaptive objectives and perturbation strengths tailored to modality-specific dynamics. This converts coupled multimodal preference learning into unimodal subproblems for precise reward attribution and better synchronization. Our 5DAV dataset decouples samples across five dimensions for systematic training. Experiments on JavisBench and VABench demonstrate AV-GRPO outperforms LTX-2.3 in generation quality, semantic alignment and cross-modal synchronization under LoRA and full fine-tuning. Ablations confirm our designs. Code and data: https://github.com/zhiyuxu03/AV-GRPO