Papers for

video content platform teams

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.

LinMorph improves video diffusion model speed and quality with learned weight conversion

LinMorph: Learning to Morph Weights for Efficient Video DiT Linearization

Abstract: Post-hoc linearization is an attractive way to accelerate pretrained Video Diffusion Transformers (Video DiTs), yet existing methods largely treat the quality loss induced by attention replacement as a recovery problem. We show that a more fundamental bottleneck lies upstream: \emph{initialization}. Random initialization discards pretrained knowledge, whereas manual weight reuse transfers it through rigid and incomplete source--target correspondences. We introduce \textbf{LinMorph}, which recasts Video DiT linearization as a learnable softmax-to-linear weight transformation problem. Instead of copying predefined counterparts, LinMorph learns a cross-parameter morphism that lets each target Gated DeltaNet (GDN) parameter draw jointly from pretrained query, key, value, and output projections, including for parameters with no direct softmax counterpart. A single morphism is shared across layers and learned through local teacher alignment, while producing layer-specific initializations from their respective pretrained weights. The resulting weights are then materialized as independent GDN modules for progressive distillation and joint recovery.Since the morphism operates in weight space rather than token space, it can be learned on short, low-resolution clips and directly reused at larger video scales. Across pretrained Video DiTs, LinMorph enables markedly faster recovery and better final generation quality than random initialization and heuristic weight reuse. Results show that efficient linearization depends not only on how a converted model is recovered, but critically on how pretrained knowledge is transferred before recovery. Project page: https://lonelyling.github.io/LinMorph.

Tue 13 OctComputer Vision and Pattern RecognitionArtificial Intelligence
The gist
Video Diffusion Transformers generate video clips but can be slow and hard to adapt efficiently. The authors identified that existing methods to speed them up do not transfer learned knowledge well when changing model parts. They created LinMorph, a new way to transform the internal weights of a video model so that it can be accelerated with less loss of quality. By learning how to best convert weights instead of copying them manually or using random guesses, LinMorph makes these video models faster and better at generating videos.
Open → 2610.17914v1

Small video language models learn better reasoning with new training trick

Efficient Reasoning Distillation: Small Video-Language Models via Synthetic CoT and Difficulty-Aware Fine-Tuning

Abstract: We present an efficient method to distill reasoning capabilities into compact video-language models (VLMs) for video question answering (VideoQA). Our approach fine-tunes a 2B-parameter model using only $\sim$900 uncertainty-selected examples, each augmented with synthetic chain-of-thought (CoT) rationales generated by a 4B teacher. Despite its minimal compute cost - under two hours on a single A100 GPU - our method enables the 2B model to outperform VLMs up to 4$\times$ larger, and generalize across CinePile, ActivityNet-QA, and MLVU, approaching the performance of its own 4B teacher. A key finding is that placing CoT rationales after the answer - contrary to standard prompting - substantially improves reasoning in compact models. This insight challenges prevailing CoT conventions and reveals new alignment strategies under limited model capacity. Our findings offer a practical blueprint for training deployable, reasoning-rich VLMs suited for mobile and edge applications.

Mon 14 SeptMachine LearningArtificial IntelligenceComputer Vision and Pattern Recognition
The gist
Video question answering models that understand videos and answer questions usually need to be very big to reason well. The authors show a way to teach a smaller model to reason almost as well by fine-tuning it with a few carefully chosen examples that include step-by-step explanations generated by a larger model. They found that putting these explanations after the answer instead of before works better for small models. This method is fast and uses less computing power, making smarter video language models easier to deploy on mobile and edge devices.
Open → 2609.16255v1